1、INTERNATIONAL TELECOMMUNICATION UNION ITU-T TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU GENERAL ASPECTS OF DIGITAL TRANSMISSION SYSTEMS G.720 (07/95) CHARACTERIZATION OF LOW-RATE WITH NON-VOICE SIGNALS DIGITAL VOICE CODER PERFORMANCE ITU-T Recommendation G.720 (Previously “CCITT Recommendation”)
2、 FOREWORD The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (IT). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommen- dations on them with a view to standardizing telecommunications
3、 on a worldwide basis. The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. The approval of Recommendations by the Members of the IT-T is
4、covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). ITU-T Recommendation G.720 was prepared by ITU-T Study Group 15 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 10th of July 1995. NOTE In this Recommendation, the expression “Admi
5、nistration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. O ITU 1995 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 and micro
6、film, without permission in writing from the ITU. CONTENTS Introduction Scope References Abbreviations . Voiceband data performance 5.1 Test Set-up Architecture . 5.2 Impairments 5.3 Measurement Procedures 5.4 Transmit and echo levels 5.5 Performance measurements 6.6 Calibration Narrow-band signal p
7、erformance . 6.1 System No . 5 inter-register signalling 6.3 DTMF Signals 6.4 6.5 Circuit Continuity Tones 6.6 Echo canceller disabling tone . 6.7 Distortion, frequency and level . Coding delay 7.1 Total delay measurement method . Convergence time . 8.1 Encodeddecoder decorrelation . 8.2 Encoder/dec
8、oder state mismatch . 8.3 Attack time 8.4 Measurement method 6.2 R-2 Signalling . System No . 5 line signalling . Facsimile performance . Bibliography Recommendation 6.720 (07/95) Page 1 1 2 3 3 3 4 6 7 7 8 8 8 9 10 11 11 12 12 13 14 14 15 15 15 15 15 15 1 SUMMARY In this Recommendation, an evaluati
9、on methodology is presented for the characterization of low-rate digital speech (or more generally source) coder performance with non-voice signals. The types of non-voice signals considered include voiceband data, network signalling, circuit continuity tones, and dual-tone multi-frequency signallin
10、g. Also addressed is the measurement of input/output delay and convergence time. Finally, appropriate test configurations, performance metrics, and test procedures are defined. 11 Recommendation G.720 (07/95) - ITU-T RECflN*G.720 95 m 48b259L Ob023Y3 2bT m Recommendation 6.720 CHARACTERIZATION OF LO
11、W-RATE DIGITAL VOICE CODER PERFORMANCE WITH NON-VOICE SIGNALS (Geneva, 1995) 1 Introduction In selecting speech coding algorithms for network applications, characterization of their performance with a diversity of telephone network signals is normally required. Typical non-voice signais whose in-ban
12、d transmission over the telephone network is desirable include the following. Voiceband Data - Some transparency to voiceband data may be desirable particularly for applications such as thin-route satellite systems, digital mobile services, and circuit multiplication systems, such as DCME and PCME.
13、It is considered that such transparency should be afforded to Public Switched Telephone Network (PSTN) voiceband data modems operating at bit rates up to and including 2400 bit/s. These modems include those conforming to Recommendations V.22, V.22 bis, V.21, V.23, V.26 bis and V.26 ter. It is recogn
14、ized that other leased line and proprietary modems have found extensive use in the telephone network. Therefore, it is also desirable to use voiceband data signals from other modems which are not standardized by the KU-T. Supervisory and inter-register signals that are incorporated within various si
15、gnalling systems in the PSTN- These signals include two frequencies (2400 Hz and 2600 Hz) present within Signalling System No. 5 for line signalling, 15 multi-frequencies used by Signalling System No. 5 for register signalling, and all valid combinations (pairs) of the six high-frequency and six low
16、-frequency tones employed by Signalling System R-2. Dual-Tone Multi-Frequency (DTMF) signals - These signals may be used independently of a signalling system to provide end-user access to specialized services such as answering machine message retrieval, airline reservations and electronic banking. 0
17、 Tones employed by Signalling Systems No. 6 and No. 7 for the verification of end-to-end circuit continuity and the four high-group and four low-group frequencies, conforming to Recom- mendation 4.23. 0 The 2100 Hz tone (including any associated phase reversais) specified in Recommendation V.25, whi
18、ch is employed by the V-Series modems for disabling network echo control devices. Finally, the tones utilized in national signalling systems to provide ringing, busy or congestion information (see Recommendation 4.35) could require support depending upon the location of the coder within the network
19、(such as in land-mobile environments) and should therefore be included in these tests. This methodology has evolved over several years from performance evaluations conducted during the development of Recommendations G.726, G.727 and G.728. 2 Scope This Recommendation encompasses methods that can be
20、used for quantifying and characterizing the performance of low-rate digital speech codecs with non-voice signals. The signals used are primarily those employed within the telephone network and, consequently, these methods are most appropriate for speech codecs intended for network (as opposed to end
21、-user terminal) applications. The methods presented here can be used to quantify the performance of voice codecs that operate in the telephone band and that employ transmission rates that are less than or equal to 64 kbitfs. Recommendation G.720 (07/95) 1 3 References The following ITU-T Recommendat
22、ions and other references contain provisions which, through reference in this text, constitute provisions to this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; ail users of this Recommendation are ther
23、efore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of ITU-T Recommendations is regularly published. - - - ITU-T V-Series of Recommendations, Data communication over the telephone network. ITU-T Recommen
24、dation Clause 2 4.141 (1993), Line signalling - 2.1 signal code for line signalling. CCITT Recommendation Q.1.51 (1988), Signal code fur register signalling. - CCITT Recommendation Q 441 (1988), Signalling code. - CCITT Recommendation 4.271 (1988), General. - - - CCITT Recommendation Q.724 (1988), S
25、ignalling procedure Specifications of Signalling System No. 7 CCITT Recommendation Q.23 (1988), Technical features of push-button telephone sets. CCITT Recommendation V.25 (1988), Automatic answering equipment and/or parallel automatic calling equipment on the general switched telephone network incl
26、uding procedures for disabling of echo control devices for both manually and automatically established calls. CCI“ Recommendation Q.35 (1988), Technical characteristics of tones for the telephone service. CCITT Recommendation G.711 (1988), Pulse Code Modulation (PCM) of voice frequencies. CCITT Reco
27、mmendation G.726 (1990), 40, 32, 24,16 kbit/s Adaptive differential pulse code modulation (ADPCM). - - - - CCITT Recommendation 0.42 (1988), Equipment to measure non-linear distortion using the 4-tone intermodulation method. - CCITT Supplement No. 3.1, Measuring instrument requirements. Sinusoidal s
28、ignal generators and level measuring instruments, Green Book, Vol. IV.2, Supplement to Series-M, -N and -0 Recommendations, pp. 530-533, Geneva, 1972. CCITT Supplement No. 3.2, Noise measuring instruments for telecommunication circuits, Green Book, Vol. IV.2, Supplement to Series-M, -N and -0 Recomm
29、endations, pp. 534-548, Geneva, 1972. - - ITU-T O-Series of Recommendations, Specifications for measuring equipment. - CCITT Recommendation Q.112 (1988), Signal levels und signal receiver sensitivity. - CCITT Recommendation Q. 1 14 (1988), Typical transmission requirements for signal senders and rec
30、eivers. - CCITT Recommendation 4.451 (1988), Definitions. - - CCITT Recommendation Q.454 (1988), The sending part of the multifrequency signalling equipment. CCITT Recommendation 4.455 (1988), The receiving part of the multifrequency equipment. Range, speed, and reliability of interregister signalli
31、ng. CCITT Recommendation Q.143 (1988), Line signal sender. CCITT Recommendation G.144 (1988), Line signal receiver. ITU-T Recommendation G.165 (1993), Echo cancellers. ITU-T Recommendation O. 133 (1993), Equipment for measuring the performance of PCM encoders and decoders. - - - - - CCITT Recommenda
32、tion 0.91 (1988), Phase jitter measuring equipment for telephone-type circuits. 2 Recommendation 6.720 (07/95) ITU-T RECMN*G-720 95 4862573 Ob02345 032 - - CCITT Recommendation 0.8 i (1988), Group-delay measuring equipment for telephone-type circuits. CCIT Recommendation 0.131 (1988), Quantizing dis
33、tortion measuring equipment using a pseudo- random noise test signal. CCITT Recommendation O. 132 (1988), Quantizing distortion measuring equipment using a sinusoidal test signal. ITU-T Recommendation E.450 (1993), Facsimile qualis, of service on PSTN - General aspects. -T Recommendation E.45 1 (199
34、3), Facsimile call cut-offperformance. ITU-T Recommendation E.452 (1993), Facsimile modem speed reductions and transaction time. - - - - 4 Ab brevia tions For the purposes of this Recommendation, the following abbreviations are used: ASCII BER Bit Error Ratio CEFT CHER Character Error Ratio DCME Dig
35、ital Circuit Multiplication Equipment DTMF Dual-Tone Multi-Frequency FR Finite Impulse Response MBER Modem Bit Error Ratio MBLER Modem Block Error Ratio MSUR Modem Missed Start-up Ratio PCM Pulse Code Modulation PCME Packet Circuit Multiplication Equipment PSTN Public Switched Telephone Network S/N
36、Signal-to-Noise Ratio SNR Signal-to-Noise Ratio TLP Transmission Level Point American Standard Code for Information Interchange European Committee on Post and Telecommunications 5 Voiceband data performance In this clause, the procedure to be used for evaluating a speech codecs performance with 2- a
37、nd 4-wire modems is described. 5.1 Test Set-up Architecture An architecture to be employed for voiceband data modem measurements should be centred around a Bit Error Ratio (BER) test system which can provide the source data for the transmit direction of the modems (Figure 1). This system should perm
38、it the analysis of the receive modem data in order to compute single errors, blocks, block lengths, block errors, test duration, and error-free seconds, all of which are parameters that can be used in the interpretation of the received signal quality. In addition to the modems and BER test set, a re
39、ference 64 kbih Pulse Code Modulation (PCM) codec, the speech codec under evaluation, and an analogue impairment channel should be employed. All measurements shall then be performed with the analogue channel in tandem with the digitally (speech) encoded channel in the forward direction of transmissi
40、on, with the input to the digitally encoded channel defined as the O Transmission Level Point (O Transmission Level Point - TLP). In this Figure 1, the signal levels are defined with the analogue channel impairments disabled, and O = O dBmO, which makes the data level equal to -15 dBm or -15 dBmO. R
41、ecommendation 6.720 (07/95) 3 Direction under test b -1 5 dBm (O TLP) / - , I - - Analogue - encoded b chmnei Rec8ibe modem channel I (ZdBrn receive -9 dBm transmit = 60f 12 dB 1 25ms hhY FIGURE 1/G. 720 Modem test configuratiai architecture 5.2 Impairments The analogue channel shall permit the intr
42、oduction of two types of network impairments: variable and fixed. Variable impairments consist of the linear addition of controlled amounts of band-limited analogue white noise prior to the digitally encoded channel. Fixed impairments consist of group and attenuation distortion, phase jitter, and se
43、cond and third order non-linear distortion. The impairments to be injected in the analogue channel are in accordance with the configuration shown in Figure 2. The values of these impairments are given in Tables 1 and 2. These impairments correspond to stringent network requirements. However, they pr
44、ovide a frame of reference against which the speech coding algorithms can be compared with other ITU-T speech coding Recommendations. It is noted that in measuring second and third order non-linear distortion (Table i), a four-tone method shall be used which employs two pairs of equi-level tones wit
45、h a composite power equal to that of the voiceband data modem signal. In the first tone pair (denoted by A), the tones shall be 6 Hz apart and centred at 860 Hz. In the second pair (denoted by B), the tones shall be 16 Hz apart and centred at 1380 Hz. The second order non-linear distortion shall be
46、determined by measuring the power of the four B + A and four B - A intermodulation products, and the third order shall be determined by measuring the six 2B - A products. NOTE - The non-linear distortion unit provides, for every 1 dB increase in the input signal, an increase in the output second-ord
47、er component of 2 dB, and an increase in the output third-order component of 3 dB. 4 Recommendation G.720 (07/95) ITU-T RECMN*G.720 95 4862593 Ob02347 905 bandpass filter Analque impairnient channel and total distortion. It is noted that the requirement regarding frequency offsets impose a tolerance
48、 mask of f 6 Hz for signal senders (Recommendation Q.143) and t 15 Hz for signal receivers (Recommendation 4.144). Recommendation 4.143 also imposes a signal to total distortion ratio limit of 40 dB for signal senders. (Recommendation 0.132 deals with relevant methods for performing these measuremen
49、ts.) Compound tone This is defined by the simultaneous presence of the 2400 and 2600 Hz tones and is used to signify a clear forward and a release guard line condition. Performance with this compound tone shall be assessed without bit errors injected into the coder link under evaluation. The analogue impairment channel shall be adjusted so that the level of each tone is the same and equal to -9 dBmO per frequency at the input to the encoder (this calibration is made without analogue impairments). The start and termination of the two tones shall be t
