1、INTERNATIONAL TELECOMMUNICATION UNION CCITT TH E INTERNATIONAL TELEGRAPH AND TELEPHONE CONSULTATIVE COMMITTEE DATA COMMUNICATION NETWORKS NETWORK ASPECTS X.140 (09/92) GENERAL QUALITY OF SERVICE PARAMETERS FOR COMMUNICATION VIA PUBLIC DATA NETWORKS Recommendation X.140 CCITT RECMN*X.140 72 4862591 0
2、57b853 820 m FOREWORD The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to s
3、tandardizing telecommunications on a worldwide basis. The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCIT between Plenary Assemblies is covered b
4、y the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). Recommendation X.140 was revised by Study Group VI1 and was approved under the Resolution No. 2 procedure on the 10th of September 1992. CCITT NOTES 1) relecommunication adminisiration and a recognized private operating agency, 2
5、) In this Recommendation, the expression “Administration” is used for conciseness to indicate both a A list of abbreviations used in this Recommendation can be found in Annex D. O IT 1993 All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, elec
6、uoriic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. CCITT RECMN*X*L40 92 4Bb259L 057b854 7b7 Recommendation X.140 GENERAL QUALITY OF SERVICE PARAMETERS FOR COMMUNICATION VIA PUBLIC DATA NETWORKS (Malaga-Torremolinos, 1984; amended at Melbourne, 198
7、8; revised 1992) The CCIT, considering (a) that users of data transmission services need general parameters which express their quality of service requirements without reference to any particular service or the means of its provision; (b) that providers of data transmission services need similar gen
8、eral parameters for representing offered services, and for relating user quality of service requirements to network performance capabilities; (c) that Recommendations X.130 and X.131 define protocol-specific performance parameters and objec- tives for circuit-switched public data networks; (d) that
9、Recommendation X.134 specifies portion boundaries and packet-layer reference events for defining packet- sw itched performance parameters; (e) that Recommendations X.135, X.136 and X.137 define protocol-specific performance parameters and values for packet-switched public data networks; (f) that Rec
10、ommendation X.200 defines the Reference Model of Open Systems Interconnection (OSI) for CCIT applications; (8) that Recommendation X.213 defines the OS1 network service; (h) that Recommendation X.300 defines general principles and arrangements for interworking among public data networks, and between
11、 public data networks and other networks, ununimously recommends that the general parameters defined below be used in specifying the end-to-end quality of public data network services as seen from the users point of view. 1 Scope and application 1.1 (PDNs). The parameters have two essential characte
12、ristics: This Recommendation defines a set of general quality of service (QOS) parameters for public data networks 1) they focus on performance efsects which are observable at the network interfaces, rather than their causes within the network; and 2) their definitions are based on protocol-independ
13、ent events (e.g. access request) rather than protocol- specific interface events (e.g. issuance of an X.21 call request signal). These characteristics m or 2) by the fact that at least one bit of user information is input to the system before access timeout, in networks that do not provide a readyfo
14、r data or equivalent signal. In connection-oriented services, there is the additional requirement that the intended called user must have been contacted and committed to the data communication session during the access attempt. This requirement distinguishes successful access outcomes from incorrect
15、 access outcomes, as discussed in Q 2.1.2 below. Access delay is divided into user-dependent and network-dependent components. Values for the network- dependent components are specified in network-specific Recommendations (e.g. Recommendation X. 135). 2.1.2 Incorrect access probability Incorrect acc
16、ess probability is the ratio of total access attempts that result in incorrect access to total access attempts in a specified sample. Incorrect access is essentially the case of a “wrong number”. It occurs when the nehvork establishes a physical or virtual circuit connection to a user other than the
17、 one intended by the call originator, and then does not correct the error before the start of user infomation transfer. Incorrect access can only occur in connection-oriented services, since the network does not establish a connection between users in connectionless services. Incorrect access is dis
18、tinguished from successful access (in connection-oriented services) by the fact that the intended called user is not contacted and committed to the data communication session during the access attempt. Values for network-specific parameters corresponding to incorrect access probability are contained
19、 in nehvork-specific Recommendations (e.g. X.136). 2.1.3 Access denial probability attempts in a specified sample.1) communication session. access outcome is indicated in one of two ways: Access denial probability is the ratio of total access attempts that result in access denial to total access Acc
20、ess denial (also termed network blocking) can occur in two ways: 1) 2) the network issues a blocking signal to the originating user during the access period (preventing the start of user information transfer); or the network delays excessively in responding to user actions during the access period,
21、with the result that user information transfer is not initiated before access timeout. Access denial is distinguished from service outage by the fact that some active response (Le. interface signal) is issued by the network during the access attempt. This ratio and all other probability ratios efine
22、d in this Recommendation are actually estimates of the true probability values. G Recommendation X.140 (09/92) -_-_ -_ CCITT RECMN*X=340 92 4862591 0576860 TbO An access attempt can also fail as a result of user blocking. Such failures are excluded from network performance measurement. User blocking
23、 is defined as any case where an access attempt fails as a result of incorrect performance or non-performance on the part of a user. Examples of user blocking include the following: a) either the originating or the called user issues a termination (or blocking) signal to the network during the acces
24、s period (preventing the start of the user information transfer); or b) the originating or the called user delays excessively in responding to network actions during the access period, with the result that user information transfer is not initiated before access timeout. An example of the latter is
25、the case where the called user does not answer an incoming call. Access timeout occurs (Le. an access attempt is considered to have failed for performance assessment purposes) whenever the duration of an individual access attempt exceeds a specified value. A procedure for distinguishing access denia
26、l from user blocking is described in 9 3. Values for network-specific parameters corres- ponding to access denial probability are contained in network-specific Recommendations (e.g. X. 136). Note - Delay to access denial is not included as a parameter because its effect on the users is considered to
27、 be insignificant. 2.2 user information transfer parameters Performance of the user information transfer function is described by six parameters: user information transfer delay, user information transfer rate, user information error probability, extra user information delivery probability, user inf
28、ormation misdelivery probability, and user information loss probability. 2.2.1 ser information transfer deZay User information transfer delay is the value of elapsed time between the start of transfer and successful transfer of a specified user information unit (e.g. block). The start of user inform
29、ation unit transfer occurs, for any given user information unit, when two conditions have been met: 1) ail bits in the unit are physically present within the network facility; and 2) the network has been authorized to transmit them. Authorization may either be an explicit user action (e.g. typing ca
30、rriage return at a buffered CRT terminal) or an implicit part of inputting the user information itself (e.g. typing a single character at an unbuffered asynchronous terminal). The successful transfer outcome is declared (on end of transfer) when an information unit is transferred from the source use
31、r to the intended destination user within the specified transfer timeout period, and the delivered unit has exactly the form and content intended by the source. The form or content of an information unit successfully delivered to a destination user may differ from that input by the source if desired
32、 conversions are performed within the network. The end of user information unit transfer records the output of user information units to the destination user in essentially the same way as the start of transfer records their input at the source. It is defined to occur when: a) all bits in the unit a
33、re physically present within the destination user facility; and b) the destination user has been notified that the information is available for use. The notification may be explicit or implicit. Recommendation X.140 (09/92) 7 - CCITT RECMN*X-340 92 = 4862593 0576863 7T7 = The user information unit u
34、sed in defining user information transfer delay is a contiguous group of user information bits delimited at the source user-network interface for transfer to a destination user as a unit. The specific number of bits in such a unit may be defined by the provider in specifying an offered service, or b
35、y the user in specifying a service requirement. User information transfer delay is divided into user-dependent and network- dependent components. Values for the network-dependent components are specified in network-specific Recommenda- tions (e.g. X.135). 2.2.2 User information transfer rate User in
36、formation transfer rate is the total number of successfully-transferred user information units in an individual transfer sample divided by the input/output time for that sample. The input/output time for a transfer sample is the larger of the input time or the output time for that sample (see Figure
37、 5LX.140). The sample input time begins when the transfer sample (defined above) begins, and ends when either: I) all digits in the sample have been input to the network, and the network has been authorized to transmit them; or 2) sample inpudoutput timeout occurs. The sample output time begins when
38、 the first user information digit in the sample is delivered by the network to the destination user. It ends when either: - the last digit of user information in the sample is delivered to the destination user; or - sample inpudoutput timeout occurs. As noted earlier, either the input or the output
39、of a transfer sample may be delayed excessively by a user (rejected sample). Such failures are excluded from network performance measurement. As in the case of user information transfer denial probability, rejected samples are distinguished from valid transfer samples using the procedure described i
40、n 0 3. Note - A “maximum user information transfer rate” which excludes the effect of user inputloutput delays can be calculated using the procedure described in 0 3. Values for network-specific parameters corresponding to user information transfer rate are contained in network-specific Recommendati
41、ons (e.g. X. 135). 2.2.3 User information error probability User information error probability is the ratio of total incorrect user information units to total successfully transferred user information units plus incorrect user information units in a specified sample. A transferred user information u
42、nit is defined to be an incorrect user information unit when the value of one or more digits in the unit is in error, or when some, but not all, digits in the unit are lost digits or extra digits (Le. digits that were not present in the original signal). Bit error ratio is a limiting case of user in
43、formation error probability in which the user information unit length, on which the error performance is based, is a single binary digit. The proportion of errored seconds is a particular case of user information error probability in which the user information unit length is defined as one second. T
44、he number of digits contained in each user information unit in this case is numerically equal to the digit rate per second. This parameter is usually expressed in the form of the percentage of its complement, i.e. as a percentage of error-free seconds (% EFS). A similar parameter, the percentage of
45、error-free deciseconds (% EFdS), can be defined based on a user information unit length of 100 ms. Values for network-specific parameters corresponding to user information error probability are contained in network-specific Recommendations (e.g. X. 136). 8 Recommendation X.140 (09J92) Case 1 -No rat
46、e conversion: Wi = Wo T07157693 Case 2 - Rare increase: Wi Wo Case 3 - Rafe reduction: Wi c Wo User information transfer rate = Bs Max Wi or Wo B1 = Total successful transfer outcomes in the transfer sample FIGURE 5B.140 User information transfer rate 2.2.4 Extru user infomation delivery probability
47、 Extra user information delivery probability is the ratio of total (unrequested) extra information units to total information units received by a destination user in a specified sample. An information unit received by a particular destination user is declared to be an extra information unit when non
48、e of the bits in the unit were input to the system by the source user for delivery to that destination. Unless misdelivered user information units are explicitiy identified in a measurement process, they will be counted as extra information units. Values for network-specific parmeters corresponding
49、to extra user information delivery probability are contained in network-specific Recommendations (e.g. X.136). 2.2.5 User information misdelivery probability User information misdelivery probability is the ratio of total misdelivered user information units to total user information units transferred between a specified source and destination user in a specified sample. A misdelivered user information unit is a user information unit transferred from a source user to a particular destination user that was actually intended for delivery to a different destination user. It is cons
