1、CCITT RECMN*E.434 92 m 4862571 0573594 727 m INTERNATIONAL TELECOMMUNICATION UNION CCITT E.434 TH E INTERNATIONAL TELEGRAPH AND TELEPHONE CONSULTATIVE COMMITTEE TELEPHONE NETWORK AND ISDN QUALITY OF SERVICE, NETWORK MANAGEMENT AND TRAFFIC ENGINEERING SUBSCRIBER-TO-SUBSCRIBER MEASUREMENT OF PUBLIC SW
2、ITCHED TELEPHONE NETWORK Recommendation E.434 Geneva, 1992 INTERNATIONAL TELECOMMUNICATION UNION CCITT THE INTERNATIONAL TELEGRAPH AND TELEPHONE CONSULTATIVE COMMITTEE TELEPHONE NETWORK AND ISDN QUALITY OF SERVICE, NETWORK MANAGEMENT AND TRAFFIC ENGINEERING E.434 CCITT RECMN*E*434 72 m 4862573 05735
3、75 Bb5 m SU BSCR I B ER-TO-S U BSC R I B ER MEASUREMENT OF PUBLIC SWITCHED TELEPHONE NETWORK Recommendation E.434 Geneva, 1992 FOREWORD The CCIIT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union 0. CCI” is responsibl
4、e for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. The Plenary Assembly of CC which meets every four years, establishes the topics for study and approves Recommendations prepared by its St
5、udy Groups. The approval of Recommendations by the members of CCIT between Plenary Assemblies is covered by the procedure lai8 down in CC Resolution No. 2 (Melbourne, 1988). procedure on the 16th of June 1992. Recommendation.E.434 was prepared by Study Group II and was approved under the Resolution
6、No. 2 1) telecommunication Administration and a recognized private operating agency. 2) 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 A. O ITU 1992 All rights reserved. No p
7、art of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. CCITT RECMN*E.434 92 E 48b259L 0573597 638 E CTE LE TE ISC IN I sc Recommendation E.434 LE CTE SUBSCRIBER-TOS
8、UBSCRER MEASUREMENT OF THE PUBLIC SWITCHED TELEPHONE NETWORK 4 b 1 Introduction These measurements are taken: Subscriber-to-subscriber service quality measurements are taken end-to-end as defined in Figure 1jE.434. to relate network quality to customers expectations which are based upon previous exp
9、erience with the network; to ensure that objective measures are made of the network; to ensure that acceptable thresholds of quality are Wing met; to ensure that regulatory requirements are met; to ensure that the distant correspondents level of quality meets other network providers needs. - - - - -
10、 To assess the network service quality requires the collection of the following: - end-to-end network measurements results; - Related publications in this area are Recommendations E.424, E.431, E.433, E.435 and M.1235 and the Hand Book on Quality of Service and network peormance: Annex E, E-series R
11、eommendations which include numerical objectives to be published in 1992). customer perception of network quality. LE Localexchange TE Transit exchange CTE Customers terminating equipment ISC International switching centre IN International network FIGURE 1m.4 I Reference diagram for end-to-end measu
12、rements 2 Scope network (PSTN), detailing: The aim of this Recommendation is to define end-to-end measurements of the public switched telephone - - the measurements required; the methods and processes for testing; Subscriber-to-subscriber type test calls are defined in Recommendation E.424. Recommen
13、dation E.434 1 -_ L_ - _I CCITT RECNN*E*434 72 = 4862571 8573598 574 - - elements of an end-to-end system; the operation of an end-to-end system, with recommendations to define interfaces and protocols. 3 Methodology 3.1 End-to-end network measurements/results To obtain a duplication of the subscrib
14、er-to-subscriber connection, the measurement must replicate the total end-to-end connection, Test calls are required to be generated for the assessment of the network between subscriber terminals. The results wiii be used to assess four areas of performance monitoring: - connection setuphelease; - c
15、onnection retainability; - transmission quality after connection is established; - billing integrity. 3.2 Customer perception of network quality There are various methods to evaluate the customers perception of the network quality; market surveys (interviews), operator keyed trouble reports (OKTR),
16、service observations and customer complaints. This subjective data can be correlated with the subscriber-to-subscriber measurements. In this manner the objective measures can either support the customers perception or require further investigation by the network provider when the customers perceptio
17、n does not agree with the objective data. 4 A typical end-to-end test call system An end-to-end test call system would typically consist of three elements: - test call controler, - test cai1 transponder, - test cai1 responder. This is shown in diagrammatic form in Figure 2B.434. Network II II Transp
18、onder Responder 12 I I FIGURE m.434 Typical test call system configuration TO203070-92 2 Recommendation E.434 CCITT RECMNaE-434 92 m 48b2591 0573599 400 m 4.1 Interfaces The sytem is composed of two interfaces, I1 and 12, see Figure 2h3.434. The I1 interface can be broken down into 3 parts: - 2-wire
19、 connection to the network; - end-to-end protocol for communication between transponder/responder; - transmission measurement technique. It is envisaged that the connection point for Il can be at the test access point at the 2-wire main distribution kame (MDF) in an exchange/remote concentrator unit
20、 building or via a portable respondedtransponder to a network terminahg equipment (“E) socket. The 12 interface is the communication between the controller and transponder/responder. It is a 2-wire interface or a private wire link to the network. 4.2 Test call controller A controller is assigned to
21、a number of transponders/responders. The controller schedules the test calls, controls the transponderdresponders, receives and stores fest results. Analysis of test results can be done at the controller or at a high order processor via a standard interface. The controller-to-responder ratio is depe
22、ndent on the capacity of the controller. 4.3 Test call transponder and responder The test call transponder initiates test calls to responders operating under instructions received from the system controller. Both transponders/responders make measurements and record the results of the test calls. The
23、se results are transferred to the transponder to be uploaded to the controller via the I2 interface. An option could exist to upload test call results from the responder directly to the controller. Transponder/responder functionality could be combined in a single type of equipment. 5 Planning and sc
24、heduling 5.1 Goals When planning subscriber-to-subscriber test, it is necessary to decide the goals of the testing. These goals help determine the content and frequency of the test. i 5.2 Testplan A test plan is required to ensure that the test results provide valid information consistent with the g
25、oals of the testing. Statistical methods (estimations of distributions, sample size, confidence intervals, etc.) should be used to develop this test plan and to make inferences from the test results. Other consideration for the test plan include traffic patterns which reflect customer usage, time of
26、 day and : day of week calling. I 5.3 Duration of test period Test should be carried out within a period of time which is consistent with the goals of testing. In general, this will be to ensure network conditions are static during the period that the tests are being conducted. Recommendation E.434
27、3 - _ ._ CCITT RECMN*E-434 92 m 4862573 0573b00 T52 m 5.4 Number of test sites When determining the specific locations and numbers of test sites to be used in a given country, factor such as geographic diversity of subscriber and location of high usage centres should be considered, It would be ideal
28、 to have the transponderdresponders at each switching node. Test facilities are essential at the local switching unit to enable to test close to the subscriber. A measurement unit at each switching node would benefit any maintenance requirement and would provide the ability to test to any part of th
29、e network. However, this is dependent on unit costs and Administrations requirements. Portable responders could be used to provide flexibility to measure all parts of the network. For international testing, the number and location of tests sites should ensure that test calls WU be routed via gateway
30、s to provide measurements of service quality on the international routes typically encountered by subscriben in the originating country. 5.5 Scheduling of test programme This section addresses the test call programming and various aspects of scheduling tests. Two typ of testing should be considered:
31、 routine testing and on-demand testing described below. The recommendations in this section are primarily applicable to routine testing which has easily definable requirements because of its periodic nature. While many of the considerations described below apply to both routine and on-demand testing
32、, it is difficult to provide guidelines for on-demand testing because the requirements vary depending on the circumstances motivating the tests. 5.5.1 Routine testing Routine testing is conducted on a national and international basis at regularly defined intervals for the purpose of evaluating curre
33、nt service quality levels and identifying the trends in service quality as they evolve over time. Network providers need to know the quantity of test traffic being generated, and coordination and agreement is required on quantity, accounting and stopping test traffic for special reasons, i.e. high d
34、emand. For international testing, the quantity of test traffic being generated needs to be controlled and coordinated bilaterally. It is noted that reporting the quantity of failures, even if reoccuring, is essential for continuing the measurement in order to statistically represent the Quantity of
35、Service of the network under test. 5.5.2 On-demand testing On-demand testing may be conducted in response to special circumstances, changes to network conditions, or as a follow-up to previously completed routine testing. On-demand, testing may include testing to isolate or sectorize service quality
36、 problems, evaluation of the service quality levels on specific routes (for example, restoration facilities or a new international route), and testing from locations other than those that are usually used in routine testing. 5.5.3 Number of test calls The number of test calls depend on the parameter
37、s to be measured and the desired accuracy for those parameters. The number of test calls should be large enough to provide statistically valid results within the desired accuracy for each parameter measured, and small enough to allow the data to be collected over a fairly short period of time. Consi
38、deration should be given to trunk size and traffk volumes on the route when test call quantities are being determined. 4 Recommendation E.434 CCITT RECMN*E-Ll3Y 92 W 4862593 0573603 999 N 5.6 Availability of responders Another consideration in keeping the test period relatively short is to ensure th
39、e availability of the test devices. These responders must be dedicated for use during a particular test period to avoid scheduling conflicts for the use of the responders. Such conflicts may result in inaccurate and misleading test results: for example, test calls may have failed to complete because
40、 the terminating responder was in use for a different study rather than a genuine network failure. The availability of responders wili be achieved through careful coordination and scheduling. 6 Control centre It is recommended that each Administration have a nominated control centre responsible for
41、managing the - planning and scheduling tests; - coordinating tests between Administrations; - maintaining test equipment; - - reporting of test results; - subscriber-to-subscriber test program. The responsibilities of this centre should include: collecting and analysing test data: referring follow-u
42、p actions as required by test results. 6.1 Central timing of testing system National networks administrating the system should impose a centralised timing. This will ensure there is no clash between testing sub-systems. 6.2 Coordination of all test calls All routine and on-demand testing involving a
43、nother Administration needs to be coordinated. A single point The control of testing will be tiered. One point of contact will have overall responsibility centrally and wili of contact is required in each Administration. This contact will coordinate and agree testing programmes. assign responsibilit
44、y to other individuals within pre defined catchment areas. 7 Results Results should report all the key performance indicators and associated measurement methods. This includes the thresholds set, statistical spread and other factors affecting the results: for example, sample size, time of day and te
45、sting locations. All reports have to be completed on timely basis. Test resultdanalysed data should be made available to managers network investigation centre and maintenance. Previous resultdanalysed data may be stored on an off-line mass storage medium for later retrieval if required. Results shou
46、ld be available in a standard ASCIi format. 8 Service quality agreement It is recommended that there be a formal service quality agreement (SQA), agreed and supported at senior management levels, to ensure that a commitment is made at the working level. Such agreements are produced and agreed betwee
47、n Administrations. This wili ensure that: a) results are exchanged regularly, including: - - summary of ali results (key performance indicators) for the duration of the testing period, a detaied breakdown of the used for the summary; Recommendation E.434 5 _ -_-I - CCITT RECMN*E.434 72 m 4862573 057
48、3602 825 m b) measurement methods are compatible; c) targets and thresholds are set for transmission performance; d) commitment is made by the Administration to target service improvements: - recommendation of the opportunities for improvement, and - location(s) of faults The placement of units with
49、in the domain of other Administrations is dependent on the cooperation of the offering Administration. 1f.equipment is exchanged, an agreement would address installation, maintenance and the portability of units within the network. Another agreement may be required for the accounting of test calls. In lieu of the benefits of the quality improvements, the cost could be shared by the Administrations. It is recommended that a quarterly review on performance between Administrations be a suitable period to support the SQA. It is recommended that similar service a
