ITU-T G 727 ANNEX A-1994 5- 4- 3- and 2-Bits Sample Embedded Adaptive Differential Pulse Code Modulation (ADPCM) - Annex A Extensions of Recommendation G 727 for Use with Uniform- o.pdf

1、 ITU-T RECMN*G-727 ANNEX*A 94 4862573 Ob02323 471 W INTERNATIONAL TELECOMMUNICATION UNION ITU-T TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU G.727 - Annex A (1 1 /94) GENERAL ASPECTS OF DIGITAL TRANSMISSION SYSTEMS 5, 41, 3- AND 2-bits SAMBPLE EMBEDDED ADAPTIVE DIFFERENTIAL PULSE CODE MODULATION

2、(ADPCM) ANNEX A: EXTENSIONS OF RECOMMENDATION G.727 INPUT AND OUTPUT FOR USE WITH UNIFORM-QUANTIZED ITU-T Recommendation G.727 - Annex A (Previously “CCITT Recommendation”) ITU-T RECMN*G-727 ANNEXaA 74 m 48b259L Ob02324 308 m FOREWORD The IT-T (Telecommunication Standardization Sector) is a permanen

3、t organ of the International Telecommunication Union (U). The IT-T is responsible 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

4、(WTSC), which meets every four years, establishes the topics for study by the IW-T Study Groups which, in their turn, produce Recommendations on these topics. The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1

5、-12, 1993). IT-T Recommendation G.727 - Annex A was prepared by IT-T Study Group 15 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 1st of November 1994. NOTE In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunic

6、ation administration and a recognized operating agency. O IT 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 microfilm, without permission in writing from the IT. CONTENTS Page 1 A.l

7、 General 1 A.3 3 Annex A - Extensions of Recommendation G.727 on 5-,4-, 3- and 2-bit sample embedded 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.727 for uniform

8、-quantized PCM . Recommendation G.727 - Annex A (11/94) i - ITU-T RECMN*G-727 ANNEXJA 94 m 4862593 O602326 180 m Recommendation 6.727 5, 4-, 3- AND 2 bits SAMPLE EMBEDDED ADAPTIVE DIFFERENTIAL PULSE CODE MODULATION (ADPCM) (Geneva, 1990) Annex A Extensions of Recommendation 6.727 on 5-,4-, 3- and 2-

9、bit sample embedded adaptive differential pulse code modulation for use with uniform-quantized input and output (Geneva, 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 Reco

10、mmendation G.727 to allow use of a uniform PCM interface at input and output. The principal application for this extension is at the analogue end points of digital transmission links, especially telephone terminal equipment. This annex defines a 1Cbit unifonn-quantized interface for Recommendation G

11、.727. Use with other data widths requires scaling the data such that the most significant bit of the 14-bit uniform PCM 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 inputs, SL. Likewise SO, the outp

12、ut value, must be right shifted by one bit to provide a 13-bit output. Clause A.2 describes the changes in principles to use uniform-quantized PCM. Clause A.3 provides the changed computational details from Recommendation G.727. NOTE - Uniform-quantized PCM is defined by reference to Recommendation

13、G.701 and to 3.6/G.711. A.2 Changes to the ADPCM algorithm The core ADPCM algorithm is unchanged, 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

14、new block LIMO has been introduced to explicitly limit the output to the range -8192 to +8191, inclusive. The input to the encoder is no longer s(k), but SI), the uniform PCM signal. The output of the decoder is no longer s&), but s,(k), the 14-bit limited version of s,(k), the feed-forward reconstr

15、ucted signal. The Synchronous Coding Adjustment was introduced into Recommendation G.727 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

16、. Analysis has shown that cumulative 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 end point of a digital transmission chain. The 14-bit uniform-PCM

17、 output value should not be re-compressed to A-law or p-law PCM as this bypasses the Synchronous Coding Adjustment. Recommendation 6.727 - Annex A (iU94) 1 ITU-T RECMN*G.727 ANNEX*A 94 m 48b2591 O602327 017 m signal * Adapw quantizer When a uniform PCM interface is implemented, Figure A.l shall be u

18、sed instead of Figure VG.727. . ADPCM i wtput 112 kbis unifm-PCM input Signal 1) estimate Bit masking COnvert to PCM Feed-forward inverseadaptive quantizer + Signal Quantizer I II I I Adaptive predictor Feed-back -b inverseadaptive Bit masking quantizer I I t T1515880-94/dOl FIGURE A.liG.727 Simplif

19、ied block diagrams 2 Recommendation G.727 - Annex A (1Y94) ITU-T RECMN*G.727 ANNEX*A 9Y Y862591 Ob02328 T53 9 Difference - d(k) signai computcdion When a uniform PCM interface is implemented, Figure A.2 shall be used instead of Figure 2/G.727. Adaptive Fdback -t* Inverse A&ive e bit quantizer adapti

20、ve predictor masking quantizer t Reconstructed signal ri calculator t A I 1- A i t a,(W 1 Quantizer 4 scale factor a,(k) adaptation T1514770-93602 Adaptation Tone and 4 detector fd(k) transt speed control FIGURE A.uG.727 Encoder block schematic When a uniform PCM interface is implemented, Figure A.3

21、 shall be used instead of Figure 3/G.727. A.3 Computational details of changes to Recommendation G.727 for uniform-quantized PCM A.3.1 Input and output signais When a uniform PCM interface is implemented, Table A. 1 shall be used instead of Table UG.727. A.3.2 Description of internai processing vari

22、ables For uniform PCM implementation, the internai processing variables shall be the same as in Table 6/G.727, with the exception of the variable: 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 Figur

23、e 4lG.727. When a uniform PCM interface is implemented, the EXPAND block shall not be used. Changes to input PCM conversion and difference signal computation Recommendation G.727 - Annex A (11/94) 3 ITU-T RECMN*G=727 ANNEX*A 94 E 48625911 Ob02329 99T I I II Encoder Input Input output Decodificador I

24、nput Input output Y) - lt,(k: + Y,(k) t t - + - Quantizer Adaptation Tone and transition adaptation 4 conird detector tdk) scale factor a, (k) T FIGURE A.3/G.727 Decoder block schematic TABLE A.UG.727 Input and output signais Name SL R (optional) I I“ R (optional) so Number of bits 14 1 C+E C+E 1 14

25、 4 Recommendation G.727 - Annex A (11/94) Description Uniform PCM input word Reset C = core bits, E = enhancement bits C = core bits, E = enhancement bits Reset Uniform PCM output word SL a-. SUBTA T1514790-WdO4 FIGURE A.4lG. 727 Difference signai computation A.3.4 When a uniform PCM interface is im

26、plemented the blocks COMPRESS, EXPAND and SYNC (see 6.2.10/G.727) shall not be used. Figure 13/G.727 shall also not be used. In addition, 6.2.10/G.727 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. T15 14800-93JdO5 FIGURE A.YG.721 output limiting LIMO (decoder only) Input: SRw output: so Function: Limit output to 16bit twos compliment value I 8191, SRw 8191 and SRw 57 344 I 57344, SRw 32 767 and SRw 57 344 Recommendation 6.727 - Annex A (11/94) 5