SIST ETS 300 580-2 E3:2003

SLOVENSKI STANDARD
01-december-2003
'LJLWDOQLFHOLþQLWHOHNRPXQLNDFLMVNLVLVWHPID]D±*RYRUVSROQRKLWURVWMR±
GHO3UHNRGLUDQMH*60UD]OLþLFD
Digital cellular telecommunications system (Phase 2) (GSM); Full rate speech; Part 2:
Transcoding (GSM 06.10 version 4.2.1)
Ta slovenski standard je istoveten z: ETS 300 580-2 Edition 3
ICS:
33.070.50 Globalni sistem za mobilno Global System for Mobile
telekomunikacijo (GSM) Communication (GSM)
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

EUROPEAN ETS 300 580-2
TELECOMMUNICATION December 2000
STANDARD Third Edition
Source: SMG Reference: RE/SMG-110610PR2
ICS: 33.020
Key words: Digital cellular telecommunications system, Global System for Mobile communications (GSM)
R
GLOBAL SYSTEM FOR
MOBILE COMMUNICATIONS
Digital cellular telecommunications system (Phase 2);
Full rate speech;
Part 2: Transcoding
(GSM 06.10 version 4.2.1)
ETSI
European Telecommunications Standards Institute
ETSI Secretariat
Postal address: F-06921 Sophia Antipolis CEDEX - FRANCE
Office address: 650 Route des Lucioles - Sophia Antipolis - Valbonne - FRANCE
Internet: secretariat@etsi.fr - http://www.etsi.org
Tel.: +33 4 92 94 42 00 - Fax: +33 4 93 65 47 16
Copyright Notification: No part may be reproduced except as authorized by written permission. The copyright and the
foregoing restriction extend to reproduction in all media.

© European Telecommunications Standards Institute 2000. All rights reserved.

Page 2
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
Whilst every care has been taken in the preparation and publication of the present document, errors in

content, typographical or otherwise, may occur. If you have comments concerning its accuracy, please
write to "ETSI Standards Making Support Dept." at the address shown on the title page.

Page 3
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
Contents
Intellectual Property Rights.7
Foreword.7
1 General.9
1.1 Scope .9
1.1.2 Normative references .9
1.1.3 Abbreviations.10
1.2 Outline description .10
1.3 Functional description of audio parts .10
1.4 PCM format conversion .11
1.5 Principles of the RPE-LTP Encoder.11
1.6 Principles of the RPE-LTP Decoder .11
1.7 Sequence and subjective importance of encoded parameters.12
2 Transmission characteristics.14
2.1 Performance characteristics of the analogue/digital interfaces .14
2.2 Transcoder delay .14
3 Functional description of the RPE-LTP Codec.15
3.1 Functional description of the RPE-LTP Encoder .15
3.1.1 Offset compensation.16
3.1.2 Preemphasis.16
3.1.3 Segmentation .16
3.1.4 Autocorrelation .16
3.1.5 Schur Recursion.16
3.1.6 Transformation of reflection coefficients to Log.-Area Ratios .16
3.1.7 Quantization and coding of Log.-Area Ratios .17
3.1.8 Decoding of the quantized Log.-Area Ratios .17
3.1.9 Interpolation of Log.-Area Ratios.18
3.1.10 Transformation of Log.-Area Ratios into reflection coefficients .18
3.1.11 Short Term Analysis Filtering .18
3.1.12 Sub-segmentation .18
3.1.13 Calculation of the LTP parameters.18
3.1.14 Coding/Decoding of the LTP lags.19
3.1.15 Coding/Decoding of the LTP gains.19
3.1.16 Long term analysis filtering.20
3.1.17 Long term synthesis filtering.20
3.1.18 Weighting Filter.20
3.1.19 Adaptive sample rate decimation by RPE grid selection.21
3.1.20 APCM quantization of the selected RPE sequence.21
3.1.21 APCM inverse quantization .23
3.1.22 RPE grid positioning.23
3.2 Decoder.23
3.2.1 RPE decoding section .23
3.2.2 Long Term Prediction section.23
3.2.3 Short term synthesis filtering section.23
3.2.4 Postprocessing.23
4 Computational details of the RPE-LTP Codec .27
4.1 Data representation and arithmetic operations.27
4.2 Fixed point implementation of the RPE-LTP Coder.30
4.2.0 Scaling of the input variable .30
4.2.1 Downscaling of the input signal.31
4.2.2 Offset compensation.31
4.2.3 Preemphasis.31
4.2.4 Autocorrelation .31
4.2.5 Computation of the reflection coefficients .32
4.2.6 Transformation of reflection coefficients to Log.-Area Ratios .33

Page 4
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
4.2.7 Quantization and coding of the Log.-Area Ratios. 33
4.2.8 Decoding of the coded Log.-Area Ratios. 33
4.2.9 Computation of the quantized reflection coefficients. 34
4.2.9.1 Interpolation of the LARpp[1.8] to get the LARp[1.8] . 34
4.2.9.2 Computation of the rp[1.8] from the interpolated
LARp[1.8] . 34
4.2.10 Short term analysis filtering . 34
4.2.11 Calculation of the LTP parameters . 35
4.2.12 Long term analysis filtering . 36
4.2.13 Weighting filter . 36
4.2.14 RPE grid selection . 37
4.2.15 APCM quantization of the selected RPE sequence . 37
4.2.16 APCM inverse quantization . 38
4.2.17 RPE grid positioning . 38
4.2.18 Update of the reconstructed short term residual signal dp[-120.-1] . 38
4.3 Fixed point implementation of the RPE-LTP Decoder . 39
4.3.1 RPE decoding section . 39
4.3.2 Long term synthesis filtering . 39
4.3.3 Computation of the decoded reflection coefficients. 40
4.3.4 Short term synthesis filtering section. 40
4.3.5 Deemphasis filtering . 40
4.3.6 Upscaling of the output signal. 41
4.3.7 Truncation of the output variable . 41
4.4 Tables used in the fixed point implementation of the rpe-ltp coder and decoder . 41
5 Digital test sequences. 43
5.1 Input and output signals . 43
5.2 Configuration for the application of the test sequences . 43
5.2.1 Configuration 1 (encoder only) . 43
5.2.2 Configuration 2 (Decoder only). 43
5.3 Test sequences . 44
5.3.1 Test sequences for configuration 1. 44
5.3.2 Test sequences for configuration 2. 44
Annex 1 (informative): Codec Performance. 48
A1.1 Introduction. 48
A1.2 Speech performance. 48
A1.2.1 Single encoding . 48
A1.2.2 Speech performance when interconnected with coding systems on an
analogue basis. 49
A1.2.2.1 Performance with 32 kbit/s ADPCM (G.721, superseded
by G.726). 49
A1.2.2.2 Performance with another RPE-LTP codec . 49
A1.2.2.3 Performance with encoding other than RPE-LTP and 32
kbit/s ADPCM (G.721, superseded by G.726). 49
A1.3 Non-speech performance. 49
A1.3.1 Performance with single sine waves. 49
A1.3.2 Performance with DTMF tones . 50
A1.3.3 Performance with information tones . 50
A1.3.4 Performance with voice-band data . 50
A1.4 Delay . 50
A1.5 References . 52
Annex 2 (informative): Subjective relevance of the speech coder output bits . 53
Annex 3 (informative): Format for test sequence distribution . 55
A3.1 Type of files provided . 55
A3.2 File format description . 55
Annex 4 (informative): Test sequence diskette. 56
Annex 5 (informative): Change Request History. 57

Page 5
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
History.58

Page 6
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
Blank page
Page 7
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
Intellectual 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 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 (http://www.etsi.org/ipr).
Pursuant to the ETSI IPR Policy, no investigation, 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 European Telecommunication Standard (ETS) has been produced by the Special Mobile Group
(SMG) Technical Committee of the European Telecommunications Standards Institute (ETSI).
The present document specifies the full rate speech transcoding within the digital cellular
telecommunications system.
NOTE: The present document is a reproduction of recommendation T/L/03/11 "13 kbit/s
Regular Pulse Excitation - Long Term Prediction - Linear Predictive Coder for use in
the digital cellular telecommunications system".
A 3,5 inch diskette (Annex 4 informative) is supplied with the present document, the diskette contains test
sequences, as described in clause 5.
The diskette contain files labelled as follows:
Diskette 1 ETS 300 580-2, annex 4: Test sequences for the GSM Full Rate speech codec;
Test sequences SEQ01.xxx to SEQ05.xxx.
The specification from which the present document has been derived was originally based on CEPT
documentation, hence the presentation of the present document may not be entirely in accordance with
the ETSI/PNE Rules.
Transposition dates
Date of adoption of this ETS: 1 December 2000

Date of latest announcement of this ETS (doa):
31 March 2001
Date of latest publication of new National Standard
or endorsement of this ETS (dop/e): 30 September 2001

Date of withdrawal of any conflicting National Standard (dow): 30 September 2001

Page 8
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
Blank page
Page 9
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
1 General
1.1 Scope
The transcoding procedure specified in the present document is applicable for the full-rate traffic channel
(TCH) in digital cellular telecommunications system. The use of this transcoding scheme for other
applications has not been considered.
In recommendation GSM 06.01, a reference configuration for the speech transmission chain of the digital
cellular telecommunications system is shown. According to this reference configuration, the speech
encoder takes its input as a 13 bit uniform PCM signal either from the audio part of the mobile station or
on the network side, from the PSTN via an 8 bit/A-law to 13 bit uniform PCM conversion. The encoded
speech at the output of the speech encoder is delivered to a channel encoder unit which is specified in
Rec.GSM 05.03. In the receive direction, the inverse operations take place.
The present document describes the detailed mapping between input blocks of 160 speech samples in
13 bit uniform PCM format to encoded blocks of 260 bits and from encoded blocks of 260 bits to output
blocks of 160 reconstructed speech samples. The sampling rate is 8000 sample/s leading to an average
bit rate for the encoded bit stream of 13 kbit/s. The coding scheme is the so-called Regular Pulse
Excitation - Long Term prediction - Linear Predictive Coder, here-after referred to as RPE-LTP.
The recommendation also specifies the conversion between A-law PCM and 13 bit uniform PCM.
Performance requirements for the audio input and output parts are included only to the extent that they
affect the transcoder performance. The recommendation also describes the codec down to the bit level,
thus enabling the verification of compliance to the recommendation to a high degree of confidence by use
of a set of digital test sequences. These test sequences are also described and are available in the
diskette (informative annex 4) which is attached to the back cover of the present document.
1.1.2 Normative references
The present document incorporates by dated and undated reference, provisions from other publications.
These normative references are cited at the appropriate places in the text and the publications are listed
hereafter. For dated references, subsequent amendments to or revisions of any of these publications
apply to the present document only when incorporated in it by amendment or revision. For undated
references, the latest edition of the publication referred to applies.
[1] GSM 01.04 (ETR 100): " Digital cellular telecommunications system (Phase 2);
Definitions, abbreviationsand acronyms".
[2] GSM 05.03 (ETS 300 575): " Digital cellular telecommunications system
(Phase 2); Channel coding".
[3] GSM 06.01 (ETS 300 580-1): " Digital cellular telecommunications system
(Phase 2); Full rate speech processingfunctions".
[4] GSM 11.10 (ETS 300 607): " Digital cellular telecommunications system
(Phase 2); Mobile Station (MS) conformity specification".
[5] ETS 300 085: "Integrated Services Digital Network (ISDN);3,1kHz telephony
teleservice Attachment requirements for handset terminals (Candidate
NET 33)".
[6] ITU-T Recommendation G.711: "Pulse code modulation (PCM) of voice
frequencies".
[7] ITU-T Recommendation G.712: "Transmission performance characteristics of
pulse code modulation".
[8] ITU-T Recommendation G.726: "40, 32, 24, 16 kbit/s adaptive differential pulse
code modulation (ADPCM)".
Page 10
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
[9] ITU-T Recommendation Q.35: "Technical characteristics of tones for the
telephone service".
[10] ITU-T Recommendation V.21: "300 bits per second duplex modem
standardised for use in the general switched telephone network".
[11] ITU-T Recommendation V.23: "600/1 200-band modem standardised for use in
the general switched telephone network".
1.1.3 Abbreviations
Abbreviations used in the present document are listed in GSM 01.04.
1.2 Outline description
The specification is structured as follows:
Section 1.3 contains a functional description of the audio parts including the A/D and D/A functions.
Section 1.4 describes the conversion between 13 bit uniform and 8 bit A-law samples. Sections 1.5 and
1.6 present a simplified description of the principles of the RPE-LTP encoding and decoding process
respectively. In section 1.7, the sequence and subjective importance of encoded parameters are given.
Section 2 deals with the transmission characteristics of the audio parts that are relevant for the
performance of the RPE-LTP codec.
Some transmission characteristics of the RPE-LTP codec are also specified in section 2. Section 3
presents the functional description of the RPE-LTP coding and decoding procedures, whereas section 4
describes the computational details of the algorithm. Procedures for the verification of the correct
functioning of the RPE-LTP are described in section 5.
Performance and network aspects of the RPE-LTP codec are contained in annex 1.
1.3 Functional description of audio parts
The analogue-to-digital and digital-to-analogue conversion will in principle comprise the following
elements:
1) Analogue to uniform digital
- microphone,
- input level adjustment device,
- input anti-aliasing filter,
- sample-hold device sampling at 8 kHz,
- analogue-to-uniform digital conversion to 13 bits representation.

The uniform format shall be represented in two’s complement.

2) Uniform digital to analogue
- conversion from 13 bit /8kHz uniform PCM to analogue,
- a hold device,
- reconstruction filter including x/sin x correction,
- output level adjustment device,
- earphone or loudspeaker.
In the terminal equipment, the A/D function may be achieved either

- by direct conversion to 13 bit uniform PCM format.
- or by conversion to 8 bit/A-law companded format, based on a standard A-law codec/filter
according to CCITT rec. G.711/714, followed by the 8-bit to 13-bit conversion according to
the procedure specified in section 1.4.

For the D/A operation, the inverse operations take place.

Page 11
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
In the latter case it should be noted that the specifications in CCITT recommendation G.714 (superseded
by G.712) are concerned with PCM equipment located in the central parts of the network. When used in
the terminal equipment, the present document does not on its own ensure sufficient out-of-band
attenuation.
The specification of out-of-band signals is defined in section 2 between the acoustic signal and the digital
interface to take into account that the filtering in the terminal can be achieved both by electronic and
acoustical design.
1.4 PCM format conversion
The conversion between 8 bit A-law companded format and the 13-bituniform format shall be as defined
in CCITT Recommendation G.721 (superseded by G.726), section 4.2.1, sub-block EXPAND and section
4.2.7, sub-block COMPRESS. The parameter LAW = 1 should be used.
1.5 Principles of the RPE-LTP Encoder
A simplified block diagram of the RPE-LTP encoder is shown in Fig1.1. In this diagram the coding and
quantization functions are not shown explicitly.
The input speech frame, consisting of 160 signal samples (uniform 13 bit PCM samples), is first pre-
processed to produce an offset-free signal, which is then subjected to a first order pre-emphasis filter.
The 160 samples obtained are then analyzed to determine the coefficients for the short term analysis filter
(LPC analysis). These parameters are then used for the filtering of the same 160 samples. The result is
160 samples of the short term residual signal. The filter parameters, termed reflection coefficients, are
transformed to log.area ratios, LARs, before transmission.
For the following operations, the speech frame is divided into 4 sub-frames with 40 samples of the short
term residual signal in each. Each sub-frame is processed blockwise by the subsequent functional
elements.
Before the processing of each sub-block of 40 short term residual samples, the parameters of the long
term analysis filter, the LTP lag and the LTP gain, are estimated and updated in the LTP analysis block,
on the basis of the current sub-block of the present and a stored sequence of the 120 previous
reconstructed short term residual samples.
A block of 40 long term residual signal samples is obtained by subtracting 40 estimates of the short term
residual signal from the short term residual signal itself. The resulting block of 40 long term residual
samples is fed to the Regular Pulse Excitation analysis which performs the basic compression function of
the algorithm.
As a result of the RPE-analysis, the block of 40 input long term residual samples are represented by one
of 4 candidate sub-sequences of 13 pulses each. The subsequence selected is identified by the RPE grid
position (M). The 13 RPE pulses are encoded using Adaptive Pulse Code Modulation (APCM) with
estimation of the sub-block amplitude which is transmitted to the decoder as side information.
The RPE parameters are also fed to a local RPE decoding and reconstruction module which produces a
block of 40 samples of the quantized version of the long term residual signal.
By adding these 40 quantized samples of the long term residual to the previous block of short term
residual signal estimates, a reconstructed version of the current short term residual signal is obtained.
The block of reconstructed short term residual signal samples is then fed to the long term analysis filter
which produces the new block of 40 short term residual signal estimates to be used for the next sub-block
thereby completing the feedback loop.
1.6 Principles of the RPE-LTP Decoder
The simplified block diagram of the RPE-LTP decoder is shown in fig 1.2. The decoder includes the same
structure as the feed-back loop of the encoder. In error-free transmission, the output of this stage will be
the reconstructed short term residual samples. These samples are then applied to the short term
synthesis filter followed by the de-emphasis filter resulting in the reconstructed speech signal samples.

Page 12
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
1.7 Sequence and subjective importance of encoded parameters
As indicated in fig 1.1 the three different groups of data are produced by the encoder are:
- the short term filter parameters,
- the Long Term Prediction (LTP) parameters
- the RPE parameters.
The encoder will produce this information in a unique sequence and format, and the decoder must
receive the same information in the same way. In table 1.1, the sequence of output bits b1 to b260 and
the bit allocation for each parameter is shown.
The different parameters of the encoded speech and their individual bits have unequal importance with
respect to subjective quality. Before being submitted to the channel encoding function the bits have to be
rearranged in the sequence of importance as given in GSM 05.03. The ranking has been determined by
subjective testing and the procedure used is described in annex 2.
==================================================================
Parameter Parameter  Parameter     Var. Number  Bit no.
number   name        name of bits (LSB-MSB)
==================================================================

==================================================================
1            LAR 1   6   b1  - b6
2            LAR 2   6   b7 - b12
FILTER     3   Log. Area     LAR 3   5   b13 - b17
4   ratios      LAR 4   5   b18 - b22
PARAMETERS   5   1 - 8       LAR 5   4   b23 - b26
6            LAR 6   4   b27 - b30
7            LAR 7   3   b31 - b33
8            LAR 8   3   b34 - b36
==================================================================
Sub-frame no.1
==================================================================
LTP      9   LTP lag       N1   7   b37 - b43
PARAMETERS  10   LTP gain      b1   2   b44 - b45
------------------------------------------------------------------
11   RPE grid position  M1   2   b46 - b47
RPE      12   Block amplitude  Xmax1   6   b48 - b53
PARAMETERS  13   RPE-pulse no.1  x1(0)   3   b54 - b56
14   RPE-pulse no.2  x1(1)   3   b57 - b59
..    ...                  ...
25   RPE-pulse no.13  x1(12)  3   b90 - b92
==================================================================
Sub-frame no.2
==================================================================
LTP      26   LTP lag       N2   7   b93 - b99
PARAMETERS  27   LTP gain      b2   2   b100- b101
------------------------------------------------------------------
28   RPE grid position  M2   2   b102- b103
RPE      29   Block amplitude  Xmax2   6   b104- b109
PARAMETERS  30   RPE-pulse no.1  x2(0)   3   b110- b112
31   RPE-pulse no.2  x2(1)   3   b113- b115
..    ...                  ...
42   RPE-pulse no.13  x2(12)  3   b146- b148
==================================================================

Table 1.1a: Encoder output parameters in order of occurrence and bit allocation within the speech
frame of 260 bits/20 ms
Page 13
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
Sub-frame no.3
==================================================================
LTP      43   LTP lag       N3   7   b149- b155
PARAMETERS  44   LTP gain       b3   2   b156- b157
------------------------------------------------------------------
45   RPE grid position  M3   2   b158- b159
RPE      46   Block amplitude  Xmax3  6   b160- b165
PARAMETERS  47   RPE-pulse no.1   x3(0)  3   b166- b168
48   RPE-pulse no.2   x3(1)  3   b169- b171
..    ...                   ...
59   RPE-pulse no.13  x3(12)  3   b202- b204
==================================================================
Sub-frame no.4
==================================================================
LTP      60   LTP lag       N4   7   b205- b211
PARAMETERS  61   LTP gain       b4   2   b212- b213
------------------------------------------------------------------
62   RPE grid position  M4   2   b214- b215
RPE      63   Block amplitude  Xmax4  6   b216- b221
PARAMETERS  64   RPE-pulse no.1   x4(0)  3   b222- b224
65   RPE-pulse no.2   x4(1)  3   b225- b227
..    ...                   ...
76   RPE-pulse no.13  x4(12)  3   b258- b260
==================================================================

Table: 1.1b: Encoder output parameters in order of occurrence and bit allocation within the
speech frame of 260 bits/20 ms

Reflection
coefficients coded as
Log. - Area Ratios
Short term
(36 bits/20 ms)
LPC
analysis
RPE parameters
(1) (2)
RPE grid
Input (47 bits/5 ms)
Short term
Pre-
selection
+
analysis
processing
signal
and coding
filter
-
(3)
(4) (5)
RPE grid
Long term
+ decoding and
analysis
positioning
filter
LTP parameters
LTP
(9 bits/5 ms)
analysis
(1) Short term residual
To
(2) Long term residual (40 samples)
(3) Short term residual estimate (40 samples)
radio
(4) Reconstructed short term residual (40 samples)
(5) Quantized long term residual (40 samples)
subsystem
Figure 1.1: Simplified block diagram of the RPE - LTP encoder

Page 14
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
Reflection coefficients coded
as Log. - Area Ratios
(36 bits/20 ms)
RPE grid
Short term
Output
Post-
decoding and
+ synthesis
processing
signal
positioning filter
RPE
parameters Long term
(47 bits/5 ms) synthesis
filter
LTP
parameters
(9 bits/5 ms)
From
radio
subsystem
Figure 1.2: Simplified block diagram of the RPE - LTP decoder
2 Transmission characteristics
This section specifies the necessary performance characteristics of the audio parts for proper functioning
of the speech trancoder. Some transmission performance characteristics of the RPE-LTP transcoder are
also given to assist the designer of the speech transcoder function. The information given here is
redundant and the detailed specifications are contained in recommendation GSM 11.10.
The performance characteristics are referred to the 13 bit uniform PCM interface.
NOTE: To simplify the verification of the specifications, the performance limits may be
referred to an A-law measurement interface according to CCITT Recommendation
G.711. In this way, standard measuring equipment for PCM systems can be utilized
for measurements. The relationship between the 13 bit format and the A-law
companded shall follow the procedures defined in section 1.4.
2.1 Performance characteristics of the analogue/digital interfaces
Concerning 1) discrimination against out-of-band signals (sending) and 2) spurious out-of-band signals
(receiving), the same requirements as defined in ETSI standard TE 04-15 (digital telephone, candidate
NET33) apply.
2.2 Transcoder delay
Consider a back to back configuration where the parameters generated by the encoder are delivered to
the speech decoder as soon as they are available.
The transcoder delay is defined as the time interval between the instant a speech frame of 160 samples
has been received at the encoder input and the instant the corresponding 160 reconstructed speech
samples have been out-put by the speech decoder at an 8 kHz sample rate.
The theoretical minimum delay which can be achieved is 20 ms. The requirement is that the transcoder
delay should be less than 30 ms.

Page 15
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
3 Functional description of the RPE-LTP Codec
The block diagram of the RPE-LTP-coder is shown in fig 3.1. The individual blocks are described in the
following sections.
3.1 Functional description of the RPE-LTP Encoder
The Preprocessing section of the RPE-LTP encoder comprises the following two sub-blocks:
* Offset compensation (3.1.1)
* Preemphasis (3.1.2)
The LPC analysis section of the RPE-LTP encoder comprises the following five sub-blocks:
* Segmentation (3.1.3)
* Auto-Correlation (3.1.4)
* Schur Recursion (3.1.5)
* Transformation of reflection coefficients to Log.-Area Ratios (3.1.6)

* Quantization and coding of Log.-Area Ratios (3.1.7)

The Short term analysis filtering section of the RPE-LTP comprises the following four sub-blocks:
* Decoding of the quantized Log.-Area Ratios (LARs) (3.1.8)

* Interpolation of Log.-Area Ratios (3.1.9)

* Transformation of Log.-Area Ratios into reflection coefficients (3.1.10)

* Short term analysis filtering (3.1.11)

The Long Term Predictor (LTP) section comprises 4 sub-blocks working on subsegments (3.1.12) of the
short term residual samples.
* Calculation of LTP parameters (3.1.13)

* Coding of the LTP lags (3.1.14) and the LTP gains (3.1.15)

* Decoding of the LTP lags (3.1.14) and the LTP gains (3.1.15)

* Long term analysis filtering (3.1.16), and Long term synthesis filtering (3.1.17)

The RPE encoding section comprises five different sub-blocks:
* Weighting filter (3.1.18)
* Adaptive sample rate decimation by RPE grid selection (3.1.19)

* APCM quantization of the selected RPE sequence (3.1.20)

* APCM inverse quantization (3.1.21)

* RPE grid positioning (3.1.22)

PREPROCESSING SECTION
Page 16
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
3.1.1 Offset compensation
Prior to the speech encoder an offset compensation,by a notch filter is applied in order to remove the
offset of the input signal s to produce the offset-free signal s .
o of
s (k) = s (k) - s (k-1) + alpha*s (k-1)      (3.1.1)
of o o of
-15
alpha = 32735*2
3.1.2 Preemphasis
The signal s is applied to a first order FIR preemphasis filter leading to the input signal s of the analysis
of
section.
s(k) = s (k) - beta*s (k-1)           (3.1.2)
of of
-15
beta= 28180*2
LPC ANALYSIS SECTION
3.1.3 Segmentation
The speech signal s(k) is divided into non-overlapping frames having a length of T = 20 ms (160
samples). A new LPC-analysis of order p=8 is performed for each frame.
3.1.4 Autocorrelation
The first p+1 = 9 values of the Auto-Correlation function are calculated by
ACF(k)=  ∑ s(i)s(i-k)   ,k = 0,1.,8     (3.2)
i=k
3.1.5 Schur Recursion
The reflection coefficients are calculated as shown in Fig 3.2 using the Schur Recursion algorithm. The
term "reflection coefficient" comes from the theory of linear prediction of speech (LPC), where a vocal
tract representation consisting of series of uniform cylindrical sections is assumed. Such a representation
can be described by the reflection coefficents or the area ratios of connected sections.
3.1.6 Transformation of reflection coefficients to Log.-Area Ratios
The reflection coefficients r(i), (i=1.8), calculated by the Schur algorithm, are in the range
-1 <= r(i) <= + 1
Due to the favourable quantization characteristics, the reflection coefficients are converted into Log.-Area
Ratios which are strictly defined as follows:
1 + r(i)
Logarea(i) = log (----------)          (3.3)
1 - r(i)
Since it is the companding characteristic of this transformation that is of importance, the following
segmented approximation is used.
r(i)            ;     |r(i)| < 0.675
LAR(i) = sign[r(i)]*[2|r(i)|-0.675] ; 0.675 <= |r(i)| < 0.950
sign[r(i)]*[8|r(i)|-6.375] ; 0.950 <= |r(i)| <= 1.000

Page 17
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
(3.4)
with the result that instead of having to divide and obtain the logarithm of particular values, it is merely
necessary to multiply, add and compare these values.
The following equation (3.5) gives the inverse transformation.
LAR’(i)              ; |LAR’(i)|<0.675
r’(i)=sign[LAR’(i)]*[0.500*|LAR’(i)|
+0.337500]            ; 0.675<=|LAR’(i)|<1.225
sign[LAR’(i)]*[0.125*|LAR’(i)|
+0.796875]            ; 1.225<=|LAR’(i)|<=1.625
(3.5)
3.1.7 Quantization and coding of Log.-Area Ratios
The Log.-Area Ratios LAR(i) have different dynamic ranges and different asymmetric distribution
densities. For this reason, the transformed coefficients LAR(i) are limited and quantized differently
according to the following equation (3.6), with LAR (i) denoting the quantized and integer coded version
c
of LAR(i).
LAR (i) = Nint{A(i)*LAR(i) + B(i)}          (3.6)
c
with
Nint{z} = int{z+sign{z}*0.5}            (3.6a)
Function Nint defines the rounding to the nearest integer value, with the coefficients A(i), B(i), and
different extreme values of LAR (i) for each coefficient LAR(i) given in table 3.1.
c
Table 3.1: Quantization of the Log.-Area Ratios LAR(i)
LAR No i  A(i)   B(i)  Minimum  Maximum
LAR (i)  LAR (i)
c c
1   20.000  0.000   -32    +31
2   20.000  0.000   -32    +31
3   20.000  4.000   -16    +15
4   20.000  -5.000   -16    +15
5   13.637  0.184   - 8    + 7
6   15.000  -3.500   - 8    + 7
7    8.334  -0.666   - 4    + 3
8    8.824  -2.235   - 4    + 3

SHORT-TERM ANALYSIS FILTERING SECTION
The current frame of the speech signal s is retained in memory until calculation of the LPC parameters
LAR(i) is completed. The frame is then read out and fed to the short term analysis filter of order p=8.
However, prior to the analysis filtering operation, the filter coefficients are decoded and preprocessed by
interpolation.
3.1.8 Decoding of the quantized Log.-Area Ratios
In this block the quantized and coded Log.-Area Ratios (LAR (i)) are decoded according to equation
c
(3.7).
Page 18
ETS 300 580-2: December 2000 (GSM 06.10 version 4.2.1)
LAR’’(i) = ( LAR (i) - B(i) )/ A(i)        (3.7)
c
3.1.9 Interpolation of Log.-Area Ratios
To avoid spurious transients which may occur if the filter coefficients are changed abruptly, two
subsequent sets of Log.-Area Ratios are interpolated linearly. Within each frame of 160 analysed speech
samples the short term analysis filter and the short term synthesis filter operate with four different sets of
coefficients derived according to table 3.2.
Table 3.2: Interpolation of LAR parameters (J=actual segment)
k         LAR’ (i) =
J
0.12  0.75*LAR’’J-1(i) + 0.25*LAR’’J(i)
13.26  0.50*LAR’’J-1(i) + 0.50*LAR’’J(i)
27.39  0.25*LAR’’J-1(i) + 0.75*LAR’’J(i)
40.159                     LAR’’J(i)

3.1.10 Transformation of Log.-Area Ratios into reflection coefficients
The reflection coefficients are finally determined using the inverse transformation according to equation
(3.5).
3.1.11 Short Term Analysis Filtering
The Short term analysis filte
...