1、INTERNATIONAL TELECOMMUNICATION UNION ITU-T TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU G.726 - Annex A GENERAL ASPECTS OF DIGITAL TRANSMISSION SYSTEMS (1 1/94) 40, 32, 24, 16 kbis ADAPTIVE DIFFERENTIAL PULSE CODE MODULATION (ADPCM) ANNEX A: EXTENSIONS OF RECOMMENDATION G.726 FOR INPUT AND OUTPU
2、T USE WITH UNIFORM-QUANTIZED ITU-T Recommendation G.726 - ANNEX A (Previously “CCITT Recommendation”) ITU-T RECMN*G.726 ANNEX*A 94 = 4862591 0602332 484 FOREWORD The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (IT). The IT-T is r
3、esponsible for studying technical, operating and tariff questions and issuing Recommen- dations on them with a view to standardizing telecommunications on a worldwide basis. The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study
4、by the IT-T Study Groups which, in their turn, produce Recommendations on these topics. The approval of Recommendations by the Members of the IT-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). ITU-T Recommendation G.726 - Annex A was prepared by ITU-T S
5、tudy Group 15 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 1 st of november 1994. NOTE In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. O ITU 1995 Al
6、l 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 microfilm, without permission in writing from the IT. CONTENTS Page 1 A.l General 1 A.3 4 Annex A - Extensions of recommendation G.726 on 40,
7、 32, 24, 16 kbit/s adaptive differential pulse code modulation for use with uniform-quantized input and output A.2 Changes to the ADPCM algorithm . 1 Computational details of changes to Recommendation G.726 for uniform-quantized PCM . Recommendation 6.726 - Annex A (1l94) 1 ITU-T RECMN*G-Zb ANNEXUA
8、94 m 4862591 Ob02334 257 m Recommendation G.726 40, 32, 24, 16 kbits ADAPTIVE DIFFERENTIAL PULSE CODE MODULATION (ADPCM)l) (Geneva, 1990) Annex A Extensions of Recommendation 6.726 on 40,32,24,16 kbis adaptive differential pulse code modulation for use with uniform-quantized input and output (Geneva
9、, 1994) (This annex forms an integral part of this Recommendation) A.l General This annex extends the Adaptive Differential Pulse Code Modulation (ADPCM) algorithm specified in Recommendation G.726 to allow use of a uniform PCM interface at input and output. The principal application for this extens
10、ion is at the analogue end points of digital transmission links, especially telephone terminal equipment. This annex defines a 14-bit urtiform-quantized interface for Recommendation G.726. Use with other data widths requires scaling the data such that the most significant bit of the 16bit uniform PC
11、M corresponds to the most significant bit of the target data width. For example, a 13-bit input value must be left shifted by one bit to provide the input, SL. Likewise SO, the output value, must be right shifted by one bit to provide a 13-bit output. Subclause A.2 describes the changes in principle
12、s to use uniform-quantized PCM. Subclause A.3 provides the changed computational details from Recommendation G.726. Appendix UG.726 describes the digital test sequences. NOTE - Uniform-quantized PCM is defined by reference to Recommendation G.701 and to 3.aG.711. A.2 The core ADPCM algorithm is unch
13、anged, but the PCM conversion and formatting blocks EXPAND and COMPRESS as well as the Synchronous Coding Adjustment shall not be used. Because there was an implicit limiting function in the COMPRESS block, a new block LIMO has been introduced to explicitly limit the output to the range -8192 to +81
14、91, inclusive. Changes to the ADPCM algorithm I) This Recommendation completely replaces the text of Recommendations (3.721 and G.723 published in Volume 111.4 of the Blue Book. It should be noted that systems designed in accordance with the present Recommendation will be compatible with systems des
15、igned in accordance with Blue Book version, Recommendation G.726 - Annex A (11/94) 1 The input to the encoder is no longer s(k). but sl(k), the uniform PCM signal. The output of the decoder is no longer s&), but s,(k), the 14-bit limited version of sik), the reconstructed signal. The Synchronous Cod
16、ing Adjustment was introduced into Recommendation G.726 to compensate for cumulative distortion caused by multiple transcodings as a signal propagates through successive PCM and ADPCM links. However, it is not possible to use this technique with uniform-quantized PCM. Analysis has shown that cumulat
17、ive distortion is not possible at 16, 24 or 32 kbit/s. However, because cumulative distortion is possible at 40 kbit/s, a 40 kbit/s decoder as described in this annex, shall be used only at the analogue endpoint of a digital transmission chain. The 14-bit uniform-PCM output value shall not be re-com
18、pressed to A-law or plaw PCM as this bypasses the Synchronous Coding Adjustment of Recommendation G.726. When a uniform PCM interface is implemented, references to Figure UG.726 shall be replaced with references to Figure A.l. Difference signai AdaptiWqUantkW ADPCM - & I I ADPCM inpui Inverse adapti
19、w quantizer difference signal signal - a) Encoder Quantized Adaptive predictor 7 14-bit limiter i b) Decoder FIGURE A.llG.726 Simplified block diagrams When a uniform PCM interface is implemented, Figure A.2 shall be used instead of Figure 2/G.726. 2 Recommendation G.726 - Annex A (11/94) ITU-T RECM
20、N*G-726 ANNEX*A 94 = 4862593 Ob02336 02T Reconstructed signal r ADPCMoutput ri calculator - Adaptation Tone and transition td(k) detector - !Ml , Quantuer scale factor adaptation speed control a, (k) 1 FIGURE A.UG.726 Encoder block schematic When a uniform PCM interface is implemented, Figure A.3 sh
21、all be used instead of Figure 31G.726. ADPCM i(k) input - sr (k) ReNXEtrUCted - Imrseadaptiw d(k) signal c Output limiting calculator quantizer t transition detector contrd Quantizer scale 1 TI 51 4730-WdO3 FIGURE A.3lG.726 Decoder block schematic Recommendation 6.726 - Annex A (11/94) 3 1u-T RECMNx
22、G.72b ANNEX*A 94 m 48b259L Ob02337 Tbb m A.3 Computational details of changes to Recommendation G.726 for uniform-quantized PCM A.3.1 Input and output signais When a uniform PCM interface is implemented. Table A.l shall be used instead of Table UG.726. TABLE A. 11G.726 Input and output signais I Enc
23、oder Input Input output output Output Output Decoder Input Input Input Input Input Output Name SL R (optional) I I I I I I I I R (optional) so Number of bits 14 1 5 4 3 2 5 4 3 2 1 14 Description Uniform PCM input word Reset 40 kbitls ADPCM word 32 kbitls ADPCM word 24 kbitls ADPCM word 16 kbitls AD
24、PCM word 40 kbitls ADPCM word 32 kbiVs ADPCM word 24 kbitls ADPCM word 16 kbitls ADPCM word Reset Uniform PCM output word A.3.2 Description of internal processing variables For uniform PCM implementations, the internai processing variables shall be the same as in Table 61G.726, with the exception of
25、 the variables DLNX, DLX, DSX, DX, SL, SLX and SP, which shall not be used. A.3.3 When a uniform PCM interface is implemented, Figure A.4 shall be used instead of Figure 4/G.726. When a uniform PCM interface is implemented, the EXPAND block shall not be used. Changes to input PCM conversion and diff
26、erence signal computation T1514740-93Id4 FIGURE A.41G.726 Difference signal conputation 4 Recommendation G.726 - Annex A (11/94) . ITU-T RECMN*G*72b ANNEX*A 94 = 4862591 Ob02338 9T2 A.3.4 When a uniform PCM interface is implemented the blocks COMPRESS, EXPAND and SYNC (see 4.2.8lG.726) shall not be
27、used. Figure 1 VG.726 shall also not be used. In addition, 4.2.8/G.726 shall be replaced by the following subclause: Output PCM format conversion and synchronous coding adjustment deleted A.3.5 Output limiting (decoder only) See Figure AS. T1514750-83M05 FIGURE A.WG.726 output limiting LIMO (decoder only) Input: SR Function: output: so Limit output to 16bit twos compliment value I 8191, SR 8191 and SR 32767 and SR 57 344 Recommendation 6.726 - Annex A (1U94) 5
