ETSI TS 126 445-2016 Universal Mobile Telecommunications System (UMTS) LTE Codec for Enhanced Voice Services (EVS) Detailed algorithmic description (V13 3 0 3GPP TS 26 445 version .pdf

1、 ETSI TS 126 445 V13.1.0 (2016-05) Universal Mobile Telecommunications System (UMTS); LTE; Codec for Enhanced Voice Services (EVS); Detailed algorithmic description (3GPP TS 26.445 version 13.1.0 Release 13) floppy3TECHNICAL SPECIFICATION ETSI ETSI TS 126 445 V13.1.0 (2016-05)13GPP TS 26.445 version

2、 13.1.0 Release 13Reference RTS/TSGS-0426445vd10 Keywords LTE,UMTS ETSI 650 Route des Lucioles F-06921 Sophia Antipolis Cedex - FRANCE Tel.: +33 4 92 94 42 00 Fax: +33 4 93 65 47 16 Siret N 348 623 562 00017 - NAF 742 C Association but non lucratif enregistre la Sous-Prfecture de Grasse (06) N 7803/

3、88 Important notice The present document can be downloaded from: http:/www.etsi.org/standards-search The present document may be made available in electronic versions and/or in print. The content of any electronic and/or print versions of the present document shall not be modified without the prior

4、written authorization of ETSI. In case of any existing or perceived difference in contents between such versions and/or in print, the only prevailing document is the print of the Portable Document Format (PDF) version kept on a specific network drive within ETSI Secretariat. Users of the present doc

5、ument should be aware that the document may be subject to revision or change of status. Information on the current status of this and other ETSI documents is available at https:/portal.etsi.org/TB/ETSIDeliverableStatus.aspx If you find errors in the present document, please send your comment to one

6、of the following services: https:/portal.etsi.org/People/CommiteeSupportStaff.aspx Copyright Notification No part may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm except as authorized by written permission of ETSI. The content

7、of the PDF version shall not be modified without the written authorization of ETSI. The copyright and the foregoing restriction extend to reproduction in all media. European Telecommunications Standards Institute 2016. All rights reserved. DECTTM, PLUGTESTSTM, UMTSTMand the ETSI logo are Trade Marks

8、 of ETSI registered for the benefit of its Members. 3GPPTM and LTE are Trade Marks of ETSI registered for the benefit of its Members and of the 3GPP Organizational Partners. GSM and the GSM logo are Trade Marks registered and owned by the GSM Association. ETSI ETSI TS 126 445 V13.1.0 (2016-05)23GPP

9、TS 26.445 version 13.1.0 Release 13Intellectual Property Rights IPRs essential or potentially essential to the present document may have been declared to ETSI. The information pertaining to these essential IPRs, if any, is publicly available for ETSI members and non-members, and can be found in ETSI

10、 SR 000 314: “Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notified to ETSI in respect of ETSI standards“, which is available from the ETSI Secretariat. Latest updates are available on the ETSI Web server (https:/ipr.etsi.org/). Pursuant to the ETSI IPR Policy, no i

11、nvestigation, including IPR searches, has been carried out by ETSI. No guarantee can be given as to the existence of other IPRs not referenced in ETSI SR 000 314 (or the updates on the ETSI Web server) which are, or may be, or may become, essential to the present document. Foreword This Technical Sp

12、ecification (TS) has been produced by ETSI 3rd Generation Partnership Project (3GPP). The present document may refer to technical specifications or reports using their 3GPP identities, UMTS identities or GSM identities. These should be interpreted as being references to the corresponding ETSI delive

13、rables. The cross reference between GSM, UMTS, 3GPP and ETSI identities can be found under http:/webapp.etsi.org/key/queryform.asp. Modal verbs terminology In the present document “shall“, “shall not“, “should“, “should not“, “may“, “need not“, “will“, “will not“, “can“ and “cannot“ are to be interp

14、reted as described in clause 3.2 of the ETSI Drafting Rules (Verbal forms for the expression of provisions). “must“ and “must not“ are NOT allowed in ETSI deliverables except when used in direct citation. ETSI ETSI TS 126 445 V13.1.0 (2016-05)33GPP TS 26.445 version 13.1.0 Release 13Contents Intelle

15、ctual Property Rights 2g3Foreword . 2g3Modal verbs terminology 2g3Foreword . 14g31 Scope 14g32 References 14g33 Definitions, abbreviations and mathematical expressions 16g33.1 Definitions 16g33.2 Abbreviations . 17g33.3 Mathematical Expressions 18g34 General description of the coder 19g34.1 Introduc

16、tion 19g34.2 Input/output sampling rate 19g34.3 Codec delay 19g34.4 Coder overview 20g34.4.1 Encoder overview . 20g34.4.1.1 Linear Prediction Based Operation . 21g34.4.1.2 Frequency Domain Operation . 21g34.4.1.3 Inactive Signal coding . 22g34.4.1.4 Source Controlled VBR Coding 22g34.4.2 Decoder ove

17、rview . 22g34.4.2.1 Parametric Signal Representation Decoding (Bandwidth Extension) . 22g34.4.2.2 Frame loss concealment 22g34.4.3 DTX/CNG operation. 23g34.4.3.1 Inactive Signal coding . 23g34.4.4 AMR-WB-interoperable option 23g34.4.5 Channel-Aware Mode . 23g34.5 Organization of the rest of the Tech

18、nical Standard 23g35 Functional description of the encoder 24g35.1 Common processing . 24g35.1.1 High-pass Filtering . 24g35.1.2 Complex low-delay filter bank analysis 24g35.1.2.1 Sub-band analysis . 24g35.1.2.2 Sub-band energy estimation 25g35.1.3 Sample rate conversion to 12.8 kHz . 26g35.1.3.1 Co

19、nversion of 16, 32 and 48 kHz signals to 12.8 kHz 26g35.1.3.2 Conversion of 8 kHz signals to 12.8 kHz 26g35.1.3.3 Conversion of input signals to 16, 25.6 and 32 kHz . 28g35.1.4 Pre-emphasis . 28g35.1.5 Spectral analysis . 29g35.1.5.1 Windowing and DFT. 29g35.1.5.2 Energy calculations . 30g35.1.6 Ban

20、dwidth detection . 31g35.1.6.1 Mean and maximum energy values per band 31g35.1.7 Bandwidth decision. 33g35.1.8 Time-domain transient detection 36g35.1.9 Linear prediction analysis . 37g35.1.9.1 LP analysis window 37g35.1.9.2 Autocorrelation computation. 37g35.1.9.3 Adaptive lag windowing . 38g35.1.9

21、.4 Levinson-Durbin algorithm . 38g35.1.9.5 Conversion of LP coefficients to LSP parameters 39g35.1.9.6 LSP interpolation 40g35.1.9.7 Conversion of LSP parameters to LP coefficients 40g3ETSI ETSI TS 126 445 V13.1.0 (2016-05)43GPP TS 26.445 version 13.1.0 Release 135.1.9.8 LP analysis at 16kHz . 41g35

22、.1.10 Open-loop pitch analysis 42g35.1.10.1 Perceptual weighting . 42g35.1.10.2 Correlation function computation . 43g35.1.10.3 Correlation reinforcement with past pitch values 44g35.1.10.4 Normalized correlation computation . 45g35.1.10.5 Correlation reinforcement with pitch lag multiples . 45g35.1

23、.10.6 Initial pitch lag determination and reinforcement based on pitch coherence with other half-frames 46g35.1.10.7 Pitch lag determination and parameter update 47g35.1.10.8 Correction of very short and stable open-loop pitch estimates . 48g35.1.10.9 Fractional open-loop pitch estimate for each sub

24、frame. 50g35.1.11 Background noise energy estimation 51g35.1.11.1 First stage of noise energy update . 51g35.1.11.2 Second stage of noise energy update . 53g35.1.11.2.1 Basic parameters for noise energy update . 53g35.1.11.2.2 Spectral diversity . 54g35.1.11.2.3 Complementary non-stationarity . 54g3

25、5.1.11.2.4 HF energy content . 55g35.1.11.2.5 Tonal stability 55g35.1.11.2.6 High frequency dynamic range 59g35.1.11.2.7 Combined decision for background noise energy update 59g35.1.11.3 Energy-based parameters for noise energy update 61g35.1.11.3.1 Closeness to current background estimate . 61g35.1

26、.11.3.2 Features related to last correlation or harmonic event . 61g35.1.11.3.3 Energy-based pause detection . 62g35.1.11.3.4 Long-term linear prediction efficiency 62g35.1.11.3.5 Additional long-term parameters used for noise estimation 63g35.1.11.4 Decision logic for noise energy update . 64g35.1.

27、12 Signal activity detection 67g35.1.12.1 SAD1 module 68g35.1.12.1.1 SNR outlier filtering 69g35.1.12.2 SAD2 module 71g35.1.12.3 Combined decision of SAD1 and SAD2 modules for WB and SWB signals . 74g35.1.12.4 Final decision of the SAD1 module for NB signals 74g35.1.12.5 Post-decision parameter upda

28、te . 75g35.1.12.6 SAD3 module 76g35.1.12.6.1 Sub-band FFT 76g35.1.12.6.2 Computation of signal features 77g35.1.12.6.3 Computation of SNR parameters . 80g35.1.12.6.4 Decision of background music 81g35.1.12.6.5 Decision of background update flag 82g35.1.12.6.6 SAD3 Pre-decision 83g35.1.12.6.7 SAD3 Ha

29、ngover 85g35.1.12.7 Final SAD decision . 85g35.1.12.8 DTX hangover addition . 87g35.1.13 Coding mode determination 89g35.1.13.1 Unvoiced signal classification . 90g35.1.13.1.1 Voicing measure 91g35.1.13.1.2 Spectral tilt 91g35.1.13.1.3 Sudden energy increase from a low energy level 92g35.1.13.1.4 To

30、tal frame energy difference . 93g35.1.13.1.5 Energy decrease after spike . 93g35.1.13.1.6 Decision about UC mode . 93g35.1.13.2 Stable voiced signal classification . 95g35.1.13.3 Signal classification for FEC. 95g35.1.13.3.1 Signal classes for FEC . 96g35.1.13.3.2 Signal classification parameters 96

31、g35.1.13.3.3 Classification procedure 97g35.1.13.4 Transient signal classification . 98g35.1.13.5 Modification of coding mode in special cases 99g35.1.13.6 Speech/music classification. 100g3ETSI ETSI TS 126 445 V13.1.0 (2016-05)53GPP TS 26.445 version 13.1.0 Release 135.1.13.6.1 First stage of the s

32、peech/music classifier . 100g35.1.13.6.2 Scaling of features in the first stage of the speech/music classifier . 102g35.1.13.6.3 Log-probability and decision smoothing . 103g35.1.13.6.4 State machine and final speech/music decision . 104g35.1.13.6.5 Improvement of the classification for mixed and mu

33、sic content . 107g35.1.13.6.6 Second stage of the speech/music classifier 111g35.1.13.6.7 Context-based improvement of the classification for stable tonal signals . 113g35.1.13.6.8 Detection of sparse spectral content 117g35.1.13.6.9 Decision about AC mode . 119g35.1.13.6.10 Decision about IC mode 1

34、19g35.1.14 Coder technology selection . 119g35.1.14.1 ACELP/MDCT-based technology selection at 9.6kbps, 16.4 and 24.4 kbps 120g35.1.14.1.1 Segmental SNR estimation of the MDCT-based technology 120g35.1.14.1.2 Segmental SNR estimation of the ACELP technology 126g35.1.14.1.3 Hysteresis and final decis

35、ion . 127g35.1.14.2 TCX/HQ MDCT technology selection at 13.2 and 16.4 kbps . 128g35.1.14.3 TCX/HQ MDCT technology selection at 24.4 and 32 kbps 130g35.1.14.4 TD/Multi-mode FD BWE technology selection at 13.2 kbps and 32 kbps . 133g35.2 LP-based Coding 134g35.2.1 Perceptual weighting. 134g35.2.2 LP f

36、ilter coding and interpolation . 135g35.2.2.1 LSF quantization . 135g35.2.2.1.1 LSF weighting function . 135g35.2.2.1.2 Bit allocation . 138g35.2.2.1.3 Predictor allocation 139g35.2.2.1.4 LSF quantizer structure . 139g35.2.2.1.5 LSFQ for voiced coding mode at 16 kHz internal sampling frequency : BC-

37、TCVQ 144g35.2.2.1.6 Mid-frame LSF quantizer 151g35.2.3 Excitation coding 152g35.2.3.1 Excitation coding in the GC, VC and high rate IC/UC modes 152g35.2.3.1.1 Computation of the LP residual signal 153g35.2.3.1.2 Target signal computation . 153g35.2.3.1.3 Impulse response computation 154g35.2.3.1.4 A

38、daptive codebook 154g35.2.3.1.5 Algebraic codebook. 156g35.2.3.1.6 Combined algebraic codebook 166g35.2.3.1.7 Gain quantization. 180g35.2.3.2 Excitation coding in TC mode 185g35.2.3.2.1 Glottal pulse codebook search . 185g35.2.3.2.2 TC frame configurations 189g35.2.3.2.4 Pitch period and gain coding

39、 in the TC mode . 191g35.2.3.2.5 Update of filter memories 194g35.2.3.3 Excitation coding in UC mode at low rates . 194g35.2.3.3.1 Structure of the Gaussian codebook 194g35.2.3.3.2 Correction of the Gaussian codebook spectral tilt . 195g35.2.3.3.3 Search of the Gaussian codebook 196g35.2.3.3.4 Quant

40、ization of the Gaussian codevector gain 197g35.2.3.3.5 Other parameters in UC mode . 198g35.2.3.4 Excitation coding in IC and UC modes at 9.6 kbps 198g35.2.3.4.2 Gaussian noise generation . 200g35.2.3.4.3 Gain coding . 200g35.2.3.4.4 Memory update 202g35.2.3.5 Excitation coding in GSC mode 202g35.2.

41、3.5.1 Determining the subframe length 203g35.2.3.5.2 Computing time-domain excitation contribution . 203g35.2.3.5.3 Frequency transform of residual and time-domain excitation contribution . 204g35.2.3.5.4 Computing energy dynamics of transformed residual and quantization of noise level . 205g35.2.3.

42、5.6 Find and encode the cut-off frequency 205g35.2.3.5.7 Band energy computation and quantization. 207g35.2.3.5.8 PVQ Bit allocation 207g35.2.3.5.9 Quantization of difference signal. 208g35.2.3.5.10 Spectral dynamic and noise filling 208g3ETSI ETSI TS 126 445 V13.1.0 (2016-05)63GPP TS 26.445 version

43、 13.1.0 Release 135.2.3.5.11 Quantized gain addition, temporal and frequency contributions combination 208g35.2.3.5.12 Specifics for wideband 8kbps 209g35.2.3.5.13 Inverse DCT 210g35.2.3.5.14 Remove pre-echo in case of onset detection 210g35.2.4 Bass post-filter gain quantization 211g35.2.5 Source C

44、ontrolled VBR Coding . 211g35.2.5.1 Principles of VBR Coding 211g35.2.5.2 EVS VBR Encoder Coding Modes and Bit-Rates 212g35.2.5.3 Prototype-Pitch-Period (PPP) Encoding . 212g35.2.5.3.1 PPP Algorithm . 212g35.2.5.3.2 Amplitude Quantization 213g35.2.5.3.3 Phase Quantization 214g35.2.5.4 Noise-Excited-

45、Linear-Prediction (NELP) Encoding . 214g35.2.5.5 Average Data Rate (ADR) Control for the EVS VBR mode 214g35.2.6 Coding of upper band for LP-based Coding Modes . 218g35.2.6.1 Bandwidth extension in time domain 218g35.2.6.1.1 High band target signal generation 219g35.2.6.1.2 TBE LP analysis 220g35.2.

46、6.1.3 Quantization of linear prediction parameters. 222g35.2.6.1.4 Interpolation of LSF coefficients . 225g35.2.6.1.5 Target and residual energy calculation and quantization . 227g35.2.6.1.6 Generation of the upsampled version of the lowband excitation . 227g35.2.6.1.7 Non-Linear Excitation Generati

47、on 228g35.2.6.1.8 Spectral flip of non-linear excitation in time domain 229g35.2.6.1.9 Down-sample using all-pass filters 229g35.2.6.1.10 Adaptive spectral whitening 230g35.2.6.1.11 Envelope modulated noise mixing. 230g35.2.6.1.12 Spectral shaping of the noise added excitation 232g35.2.6.1.13 Post p

48、rocessing of the shaped excitation . 232g35.2.6.1.14 Estimation of temporal gain shape parameters 234g35.2.6.1.15 Estimation of frame gain parameters . 237g35.2.6.1.16 Estimation of TEC/TFA envelope parameters. 239g35.2.6.1.17 Estimation of full-band frame energy parameters . 241g35.2.6.2 Multi-mode

49、 FD Bandwidth Extension Coding . 243g35.2.6.2.1 SWB/FB Multi-mode FD Bandwidth Extension . 244g35.2.6.2.2 WB Multi-mode FD Bandwidth Extension . 255g35.2.6.3 Coding of upper band at 64 kb/s . 259g35.2.6.3.1 Coding in normal mode . 260g35.2.6.3.2 Coding in transient mode . 264g35.3 MDCT Coding Mode . 267g35.3.1 General description . 267g35.3.2 Time-to-frequency transformations 267g35.3.2.1 Transform sizes and MDCT configurations 267g35.3.2.2 Long block transformation (ALDO window) 267g35.3.2.2.1 Folding and on-the-fly window decimation. 269g35.3.2.2.2 eDCT . 271g35.3.2.3 Trans