1、ITU-T RECMNIE.846 93 4862593 0591485 T57 H INTERNATIONAL TELECOMMUNICATION UNION ITU-T TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU (03/93) TELEPHONE NETWORK AND ISDN QUALITY OF SERVICE, NETWORK MANAGEMENT AND TRAFFIC ENGINEERING ACCESSIBILITY FOR 64 kbis CIRCUIT SWITCHED INTERNATIONAL END-TO-END
2、 ISDN CONNECTION TYPES ITU-T Recommendation E.846 (Previously “CCITT Recommendation”) FOREWORD The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the International Telecom- munication Union. The ITU-T is responsible for studying technical, operating and tariff questions
3、 and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. The World Telecommunication Standardization Conference (WTSC), which meets every four years, established the topics for study by the ITLJ-T Study Groups which, in their turn, produce Recommenda
4、tions on these topics. ITU-T Recommendation E.846 was revised by the ITU-T Study Group II (1988-1993) and was approved by the WTSC (Helsinki, March 1-12, 1993). NOTES 1 As a consequence of a reform process within the International Telecommunication Union (ITU), the CCIT ceased to exist as of 28 Febr
5、uary 1993. In its place, the ITU Telecommunication Standardization Sector (ITU-T) was created as of 1 March 1993. Similarly, in this reform process, the CCIR and the IFRF3 have been replaced by the Radiocommunication Sector. In order not to delay publication of this Recommendation, no change has bee
6、n made in the text to references containing the acronyms “CCITT, CCIR or IFRB” or their associated entities such as Plenary Assembly, Secretariat, etc. Future editions of this Recommendation will contain the proper terminology related to the new ITLJ structure. 2 telecommunication administration and
7、 a recognized operating agency. In this Recommendation, the expression “Administration” is used for conciseness to indicate both a O ITU 1994 All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying a
8、nd microfilm, without permission in writing from the ITU. ITU-T RECMN*E-646 93 4662591 059LY67 82T CONTENTS Introduction Scope Purpose . Definitions Short-Term-Accessibility . Objective for accessibility performance . Annex A . Success criterium for connection attempts Appendix I . Examples of STA d
9、istribution functions Appendix II . Evaluation of the STA distribution function Recommendation E.846 (03/93) Page 1 1 2 2 2 2 4 5 6 i ACCESSIBILITY FOR 64 kbih CIRCUIT SWITCHED INTERNATIONAL END-TO-END ISDN CONNECTION TYPES (Helsinki, 1993) 1 Introduction Accessibility performance of networks varies
10、 very much in time. One reason is changing traffic load. A second reason is changing traffic handling capacity of networks, e.g. because of equipment faults, preventive maintenance, etc. The importance of the second factor is increasing as networks are becoming more complex. As the quality of ISDN s
11、ervices depends on the end-to-end accessibility performance, an objective for accessibility performance should reflect the aspects which are important for users and should not depend on assumptions on network internal conditions. This objective has to cover fully operational as well as faulty networ
12、k states. Recommendation E.721 contains an objective for end-to-end blocking for circuit switched services in the evolving ISDN. This Recommendation is based on the assumption that the network and its components are fully operational and makes assumptions on the traffic load. It is therefore not suf
13、ficient for reliability engineering of networks nor is it sufficient to derive quality of service values. Recommendation E.845 contains an end-to-end objective for accessibility performance of the international telephone “Connection Accessibility” as a long-term average access probability. This is n
14、ot satisfactory, even when concentrating on busy hours only. For example, no access during 5% and full access during 95% of the busy hours will be perceived differently by users than a constant access probability of 95% during the same time period. But in both cases the long- term average access pro
15、bability is the same. When a connection attempt is blocked a user will retry if the connection is important for him. So, for users, both the probability to get through with the first attempt, and if this one is unsuccessful, the probability to get through with repeated attempts are important. For th
16、e first attempt the long term average success probability for connection attempts is an adequate measure. However, this quantity gives no indication about the probability to get through with repeated attempts in case the first one is unsuccessful. Therefore, the proportion of a short time interval w
17、hich allows a successful connection is used in this Recommendation for specifying an acceptable network performance for repeated attempts. I service including effects of equipment faults and traffic congestion. However, this Recommendation defines 2 Scope This is a Recommendation on network performa
18、nce. It covers the connection access phase for all network conditions from normal operation over degraded operation to unavailable states). It provides measures and objectives for accessibility performance of international circuit switched 64 kbit/s ISDN connection types (see Table 2A.340: AI, A4, A
19、7) between T-reference points. It covers effects of traffic load, failures, operational actions (planned and unplanned), etc. It does not cover blocking because of lack of channels at T-reference points nor catastrophic situations, e.g. earth quakes, floods, etc. It only covers connections which are
20、 all the way within ISDNs. Poor transmission quality of established connections is not considered, even at the beginning of the information transfer phase. Recommeadation E.846 (03/93) 1 1u-T RECMNxE.84b 93 4862533 0593489 GT2 I 3 Purpose Nd possble 4 b4 b Time T1 T2 This Recommendation sets end-to-
21、end accessibility performance objectives for combined traffic and reliability design of ISDN networks and for defining operations strategies for such networks. Examples of application of this Recom- mendation are comparison of different network concepts including reliability aspects, dimensioning of
22、 networks for fault-free and faulty states, calculation of required availability of network components, evaluation of quality of service. 4 Definitions In this Recommendation “accessibility performance” is defined as the ability of a network to provide a connection tothe correct destination within s
23、pecified delays when a correct request is received from the user side of the T-reference-poin t. The term “accessibility” stands here for measures for accessibility performance (for “measure” see 3011 in Supple- ment No. 6 to Series-E Recommendations). The explicit definition of the measure is given
24、 in clause 5. 5 Short-Term-Accessibility From a location in a network to a specific destination sometimes connection Set-up is possible and sometimes not because somewhere in the network no resource is free (e.g. because of high traffic load, faults, etc.). If an interval of length Ta is taken, the
25、proportion of the interval in which an additional connection is possible (within the specified delays) is a random variable (see Figure i). This random variable will be called “Short-Term-Accessibility” (STA). The distribution function of this random variable is the measure for accessibility perform
26、ance defined here for repeated attempts. This measure reflects the proportion of time during which the STA is equal to or less than a given value. For illustration, Appendix 1 shows examples of simulated STA distribution functions. Connection Set-up k Ta 4 i i STA= (Ti +T2 + T3Va FIGURE 1E.846 Examp
27、le for iilustrating the definition of the dom variable STA 6 Objective for accessibility performance To reflect successive trials of users when encountering blocked connection attempts, the following value for Ta is specified: Ta = 5 minutes The criteria for defining successful connection attempts a
28、re given in Annex A. 2 Recommendation E.W (03/93) ITU-T RECM“CE.846 93 M 4662571 0593490 314 The objective for accessibility performance is specified by means of a Long term average probability of successful connection attempts and a mask for the distribution function of the Short-Tem-Accessibility
29、as shown in Figure 2. In order to comply with this Recommendation, two conditions have to be met for ail time consistent one-hour intervals (one-hour period starting at the same time each day): 1) The average access probability has to be greater than or equal to Pa = 0.94 NOTE - To allow, in additio
30、n to blocking in normal busy hours, some contribution from exceptionally high load and degraded network states, this value is 1% lower than the corresponding value in Recommendation E.721. 2) Any actual distribution function of the Short Term Accessibility has to be below the mask given in Figure 2.
31、 Probability 1 .o 0.5 0.2 0.1 0.05 o .o2 (STA 5 x) 0.01 0.005 0.002 P“ = 0.001 0.0 02 0.4 x,= 0.6 0.6 X“ 1.0 X TOp402GWd02 FIGURE 2E.846 Mask for the distribution fundiai of the Short-Term-Accessibility (STA) Blocking at the T-reference point because of lack of B-channels is not included in the obje
32、ctive. Recommendation E.846 (0-3) 3 ITU-T RECMN*E-B4b 73 D 4862571 0591471 2.50 M Call phase Parameter Reference Events Codes a) Start stop Before dialling Activation time not yet defined of layer 1 (for basic access) Establishment delay Q5 a After dialling Connection setup P1, (en-bloc) P& delay (a
33、ccording Rec. 1.352) to the definition of 9 (overlap) P& The practical meaning of the mask is as follows. Assume that a large sample of non-overlapping intervals of duration Ta is taken. The proportion of those intervals having an average probability for successful connection attempts smaller than t
34、he value given on the x-axis, has to be less than the corresponding value on the mask. The mask in Figure 2 is defined by the following parameter values: Success citerium Activation time of layer 1 x b, Delay between SABME andUAyb) Connection setup delay a) Without Intelligent Network data base inqu
35、iry: 10 seconds b) With Intelligent Network data base inquiry: 12 seconds Pu = 0.001 Pt = 0.03 X, = 0.95 Xt = 0.6 The mask reflects the idea that network states with lower STA have to be less probable than states with higher STA. Pu can be interpreted as the maximum acceptable unavailability of the
36、traffic relation. The objective in this Recommendation is a minimum requirement. It is up to the concerned network operators to provide a better network performance. Annex A Success criterium for connection attempts (This annex forms an integral part of this Recommendation) The following table sets
37、the maximum delays between reference events for the characterization of a successful connection attempt. TABLE A.lE.846 Characterization of successful connection attempts 4 Recommendation E.846 (03193) ITU-T RECMN*E.Yb 93 = 4862.593 0593492 297 Appendix I Examples of STA distribution functions (This
38、 appendix does not form an integral part of this Recornmendation) To illustrate the Short-Term-Accessibility (STA) concept, this appendix contains simulated STA distribution functions for a fully operational international traffic relation. Distribution functions of the STA of three trunk groups in s
39、eries have been generated by simulation for Ta = 5 minutes for different trunk group size. These trunk groups represent an intemational and two national trunk groups of an international relation. The traffic offered to the trunk groups was poissonian with negative exponential holding time with a mea
40、n value of 3 minutes. The traffic streams on the three trunk groups were independent of each other. All blocked connection attempts were repeated with a probability of 0.75 and a mean delay of 66 seconds (a constant delay of 13 seconds plus a negative exponentially distributed interval). The offered
41、 traffic load was chosen for a blocking probability of 1.66 % on each trunk group without repeated attempts. The STA was evaluated in consecutive 5 minute intervals. The simulated time was 360 hours in each case (see Figure I. 1). 1 o. 1 h VI a 6 a 0.01 .- 8 L 0.001 0.0001 i! I : O 0.1 0.2 0.3 0.4 0
42、.5 0.6 0.7 0.8 0.9 1 X T0204030-94 FIGURE I.lE.846 Examples at Smilated Short-TermAccessibity (STA) distribution functims Recommendation E.846 (03/93) 5 ITU-T RECMM*E.846 73 4862573 0593473 023 Appendix II Evaluation of the STA distribution function (This appendixdoes not form an integral part of th
43、is Recommendation) NOTE - This appendix will be replaced later on by a Recommendation for compliance tests. The subject is for further study. By counting successful and total connection attempts over successive intervals Ta and dividing the first by the second, estimates of samples of the random var
44、iable STA are obtained*). They can be used to evaluate an empirical distribution function. This can be done by observing real traffic or additional test traffic. A sufficient number of connection sufficient attempts within the intervals Ta are necessary to achieve precision of the results. The probl
45、em with this direct kind of measurement is the long observation time necessary to get enough data for covering rare situations. It is the same problem as with measuring availability. To reduce the observation time, many traffic relations in the network which use the same configuration and type of eq
46、uipment can be measured simultaneously. An alternative is to calculate the distribution function of the STA from operational data such as availability of equipment and traffic measurement results. *) Unsuccessful attempts due to lack of channels at T-reference points have to be counted as successful. 6 Recommendation E.846 (03/93)
