Enhanced Variable Rate Codec, Speech Service Options 3, 68, 70, and 73 for Wideband Spread Spectrum Digital Systems

Transcription

1 GPP C.S00-D Version.0 October 00 Enhanced Variable Rate Codec, Speech Service Options,, 0, and for Wideband Spread Spectrum Digital Systems 00 GPP GPP and its Organizational Partners claim copyright in this document and individual Organizational Partners may copyright and issue documents or standards publications in individual Organizational Partner's name based on this document. Requests for reproduction of this document should be directed to the GPP Secretariat at Requests to reproduce individual Organizational Partner's documents should be directed to that Organizational Partner. See for more information.

2 GPP C.S00-D v.0 REVISION HISTORY Revision C.S00-0 v.0 Enhanced Variable Rate Codec (EVRC) December C.S00-0- Enhanced Variable Rate Codec, Speech Service Option for Wideband Spread Spectrum Systems Addendum December C.S00-0- Enhanced Variable Rate Codec, Speech Service Option for Wideband Spread Spectrum Digital Systems Addendum C.S00-0- v.0 EVRC TTY/TDD Extension Release 0 Addendum C.S00-A v.0 C.S00-B v.0 C.S00-C v.0 C.S00-D v.0 C.S00-D v.0 C.S00-D v.0 Enhanced Variable Rate Codec, Speech Service Option for Wideband Spread Spectrum Digital Systems Enhanced Variable Rate Codec, Speech Service Options and for Wideband Spread Spectrum Digital Systems Enhanced Variable Rate Codec, Speech Service Options,, and 0 for Wideband Spread Spectrum Digital Systems Enhanced Variable Rate Codec, Speech Services Options,, 0, and for Wideband Spread Spectrum Digital Systems Enhanced Variable Rate Codec, Speech Services Options,, 0, and for Wideband Spread Spectrum Digital Systems Enhanced Variable Rate Codec, Speech Services Options,, 0, and for Wideband Spread Spectrum Digital Systems Date December May 00 May 00 May 00 February 00 May 00 January 00 October 00

3 GPP C.S00-D v.0 0 FOREWORD (This foreword is not part of this Standard.) The scope of this document specifies the technical requirements that form a standard for Service Options,, 0, and, enhanced variable rate, two-way speech service options known as EVRC-A, EVRC-B, EVRC-WB, and EVRC-NW respectively. These service options communicate at one of four channel rates corresponding to the 00 bps, 00 bps, 00 bps, and 00 bps frame rates. However, Service Options and 0 do not use 00 bps channel frame rate. The speech coder source bit-rate corresponding to the above mentioned channel rates are 0 bps, 000 bps, 000 bps, and 00 bps. Service Options, 0, and have the capability to operate in multiple capacity operating points at different average rates of operation that can be used for obtaining different system capacity while trading off speech quality gracefully. This standard does not address the quality or reliability of Service Options,, 0, and nor does it cover equipment performance or measurement procedures. i

4 GPP C.S00-D v NOTES. The appropriate version of the Minimum Performance Standard for the enhanced variable rate codecs, EVRC- A, EVRC-B, EVRC-WB and EVRC-NW provides specifications and measurement methods.. Base station refers to the functions performed on the landline side, which are typically distributed among a cell, a sector of a cell, a mobile switching center, and a personal communications switching center..this document uses the following verbal forms: Shall and shall not identify requirements to be followed strictly to conform to the standard and from which no deviation is permitted. Should and should not indicate that one of several possibilities is recommended as particularly suitable, without mentioning or excluding others; that a certain course of action is preferred but not necessarily required; or that (in the negative form) a certain possibility or course of action is discouraged but not prohibited. May and need not indicate a course of action permissible within the limits of the standard. Can and cannot are used for statements of possibility and capability, whether material, physical, or causal... Footnotes appear at various points in this specification to elaborate and further clarify items discussed in the body of the specification... Unless indicated otherwise, this document presents numbers in decimal form. Binary numbers are distinguished in the text by the use of single quotation marks. In some tables, binary values may appear without single quotation marks if table notation clearly specifies that values are binary. The character x is used to represent a binary bit of unspecified value. For example xxx0000 represents any - bit binary value such that the least significant five bits equal Hexadecimal numbers (base ) are distinguished in the text by use of the form 0xhh, where hh represents a string of hexadecimal digits. For example, 0xFA represents a number whose binary value is and whose decimal value is... Full-Rate and Rate are used interchangeably in this document. The same rule applies to Half-Rate and Rate /, Quarter-Rate and Rate /, and Eighth-Rate and Rate /... The following conventions apply to mathematical expressions in this standard: o x indicates the largest integer less than or equal to x:. =,.0 =, and -. o x indicates the smallest integer greater than or equal to x:. =,.0 =, and. =. o x indicates the absolute value of x: =, =. o indicates exclusive OR. o min(x, y) indicates the minimum of x and y. o max(x, y) indicates the maximum of x and y. o In figures, indicates multiplication. In formulas within the text, multiplication is implicit. For example, if h(n) and p L (n) are functions, then h(n) p L (n) = h(n) p L (n). o x mod y indicates the remainder after dividing x by y: x mod y = x - (y x/y). o round(x) is traditional rounding: round(x) = sign(x) x 0., where sign x 0 x. x 0 ii

5 GPP C.S00-D v.0 o indicates summation. If the summation symbol specifies initial and terminal values, and the initial value is greater than the terminal value, then the value of the summation is 0. For example, if N=0, and if f(n) represents an arbitrary function, then N n f(n) 0. o The bracket operator, [ ], isolates individual bits of a binary value. VAR[n] refers to bit n of the binary representation of the value of the variable VAR, such that VAR[0] is the least significant bit of VAR. The value of VAR[n] is either 0 or. o Unless otherwise specified log(x) denotes logarithm at base 0 throughout this document. iii

6 GPP C.S00-D v.0 REFERENCES The following standards contain provisions which, through reference in this text, constitute provisions of this Standard. At the time of publication, the editions indicated were valid. All standards are subject to revision, and parties to agreements based on this Standard are encouraged to investigate the possibility of applying the most recent editions of the standards indicated below. 0 0 NORMATIVE REFERENCES N. C.S000-E, Upper Layer (Layer ) Signalling Standard for cdma000 Spread Spectrum Systems, September 00. N. C.S00-0, Discontinuous Transmission (DTX) of Speech in cdma000 Systems, December 00. N. TIA--B (R00), Mobile Station-Base Station Compatibility Standard for Wideband Spread Spectrum Cellular Systems, October 00. N. C.S00-E, Medium Access Control (MAC) Standard for cdma000 Spread Spectrum Systems, September 00. N. ANSI/EIA/TIA--A, Telecommunications Telephone Terminal Equipment Transmission Requirements for Digital Wireline Telephones, October. N. ITU-T Recommendation G., Pulse Code Modulation (PCM) of Voice Frequencies, November. N. ITU-T Recommendation G., Transmission Performance Characteristics of Pulse Code Modulation, November 00. N. IEEE Standard -00, IEEE Standard Methods for Measuring Transmission Performance of Analog and Digital Telephone Sets, Handsets, and Headsets, April 00. N. IEEE Standard - (R00), Method for Determining Objective Loudness Ratings of Telephone Connections, September 00. N0. C.S000-E, Physical Layer Standard for cdma000 Spread Spectrum Systems, September 00. N. ITU-T Recommendation V., Operational and interworking requirements for DCEs operating in the text telephone mode, November 000. N. ITU-T Recommendation P., Transmission Characteristics for Wideband (0-000 Hz) Digital Handset Telephones, June 00. N. ITU-T Recommendation P., Transmission Characteristics for Wideband (0-000 Hz) Digital Hands-free Telephony Terminals, June INFORMATIVE REFERENCES S. C.S00-0 v.0, Enhanced Variable Rate Codec (EVRC), December. S. C.R00-G, Administration of ParameterValue Assignments for cdma000 Spread Spectrum Standards, June 00. S. ITU-T Recommendation G., Digital network echo cancellers, March 00. R. Proakis, J. G. and Manolakis, D. G., Introduction to Digital Signal Processing, Macmillan, New York,. R. Rabiner, L. R. and Schafer, R. W., Digital Processing of Speech Signals, Prentice-Hall Inc., New Jersey. R. Crochiere, R. E. and Rabiner L. R., Multirate Digital Signal Processing, Prentice-Hall Inc., New Jersey. R. Oppenheim, A. V. and Schafer, R. W., Digital Signal Processing, Prentice-Hall Inc., New Jersey. R. Dellar, J. R., Proakis, J. G., Hansen, J. H. L., Discrete-Time Processing of Speech Signals, Macmillan, New York,.Reserve. R. Alejandro, A., Acoustical and Environmental Robustness in Automatic Speech Recognition, Kluwer Academic Publishers, Boston,. iv

7 GPP C.S00-D v.0 0 R. Atal, B. S. and Schroeder, M. R., Stochastic coding of speech at very low bit rates, Proc Int. Conf. Comm., Amsterdam,, pp 0-. R. Kleijn, W. B., Kroon, P., and Nahumi, D., The RCELP Speech-Coding Algorithm, European Transactions on Telecommunications, Vol, No. Sept/Oct, pp -. R. Nahumi, D and Kleijn, W. B., An Improved kb/s RCELP Coder, IEEE Workshop on Speech Coding,. R0. U. Mittal, J. P. Ashley, and E. M. Cruz-zeno, Low Complexity Joint Optimization of Excitation Parameters in Analysis-by-Synthesis Speech Coding, Proceedings Eurospeech 00, pp. -. R. Laflamme, C., Adoul, J-P., Salami, R., Morissette, S., and Mabilleau, P., kbps wideband speech coding technique based on algebraic CELP, Proc. ICASSP, pp. -. R. Salami, R., Laflamme, C., Adoul, J-P., et.al, A Toll Quality kbits Speech coder for Personal Communication Systems (PCS), IEEE Trans. on Vehicle Technology,. R. Ashley, J.P., Cruz-Zeno, E.M., Mittal, U., Peng,W., Wideband coding of speech using a scalable pulse codebook, 000 IEEE Workshop on Speech Coding, September 000, pp -0. R. Gerson, I.A.; Jasiuk, M.A., Techniques for improving the performance of CELP-type speech coders, Selected Areas in Communications, IEEE Journal, Volume 0, Issue, June Page(s):. v

8 GPP C.S00-D v TABLE OF CONTENTS GENERAL...-. Scope Requirements Language References Normative References Informative References Service Option Number Allowable Delays Allowable Transmitting Speech Codec Encoding Delay Allowable Receiving Speech Codec Decoding Delay Special Cases Blanked Packets Null Traffic Channel Data All Zeros Packet Terms and Numeric Information... - REQUIRED MULTIPLEX OPTION SUPPORT...-. Interface to Multiplex Option Transmitted Packets Received Packets Negotiation for Service Option Procedures Using Service Option Negotiation Initialization and Connection Initialization and Connection in the Mobile Station Initialization and Connection in the Base Station Service Option Control Orders Procedures Using Service Negotiation Initialization and Connection Service Option Control Messages Service Option Group Assignments Negotiation for Service Option Procedures Using Service Negotiation Initialization and Connection Service Option Control Messages Negotiation for Service Option Procedures Using Service Negotiation... - vi

9 GPP C.S00-D v Initialization and Connection Service Option Control Messages Negotiation for Service Option Procedures Using Service Negotiation Initialization and Connection Service Option Control Messages... - AUDIO INTERFACES...-. Input Audio Interface Input Audio Interface in the Mobile Station Conversion and Scaling Digital Audio Input Analog Audio Input Transmit Level Adjustment Band Pass Filtering Echo Return Loss Input Audio Interface in the Base Station Sampling and Format Conversion Transmit Level Adjust Line Echo Canceling Output Audio Interface Output Audio Interface in the Mobile Station Band Pass Filtering Receive Level Adjustment Output Audio Interface in the Base Station Receive Level Adjustment... - OVERVIEW AND SPEECH ENCODER...-. Service Option Service Option Pre-Processing Rate/Coding Scheme Decision Open-Loop Decision Closed-Loop Decision Capacity Operating Point-Switching and Dimming Arbitrary Capacity Operating Point Operation Full-Rate CELP and Half-Rate CELP Full-Rate PPP and Quarter-Rate PPP and Special Half-Rate PPP... - vii

10 GPP C.S00-D v Background PPP Model Further Processing of PPP Frames Amplitude Quantization Phase Quantization Special Half-Rate PPP Frame Detection of Outliers Frame Erasure Handling Bad Rate/Packet Handling Service Options 0 and Leverage from EVRC-B to Obtain EVRC-WB-LB Bit Allocation Coding the High-Frequency Band of Wideband Speech Click Remover High-Band Analysis High-Band Excitation Generation Silence Encoder Decoder Details Input Signal Preprocessing Fast Automatic Gain Control for Wideband Inputs High-Pass Filter Noise Suppression Frequency Domain Conversion Channel Energy Estimator Channel SNR Estimator Voice Metric Calculation Band SNRs for Rate Determination Algorithm Spectral Deviation Estimator Peak-to-Average Ratio Background Noise Update Decision SNR Estimate Modification SNR Variability Estimator Channel Gain Computation Frequency Domain Filtering Background Noise Estimate Update Time Domain Signal Reconstruction Zero Delay Overlap/Add... - viii

11 GPP C.S00-D v Analysis Filterbank Generation of the Lowband Signal, s LB Generation of the Highband Signal, s HB Interpolation to khz Resampling to khz Decimation to khz Spectral Reversal Decimation to khz Spectral Shaping Model Parameter Estimation Formant Filter Parameter Calculation Direct Form LPC Parameter Calculation Generation of Spectral Transition Indicator (LPCFLAG) Direct Form LPC to LSP Conversion Generation of the Short-Term Prediction Residual LSP Interpolation LSP to Direct Form LPC Conversion Generation of Residual Samples Calculation of the Delay Estimate and Long-Term Prediction Gain Non-Exhaustive Open Loop Delay Search Refinement of Open Loop Delay Estimate for SO Refinement of Open Loop Delay Estimate for SO, SO 0, and SO Long-Term Prediction Gain Calculation Smoothed Delay Estimate and LTP Gain for SO Smoothed Delay Estimate and LTP Gain for SO, SO 0, and SO Composite Delay and Gain Calculations for SO Composite Delay and Gain Calculations for SO, SO 0, and SO Determining the Data Rate for SO and Voice Activity Detection for SO, SO 0, and SO Estimating the Data Rate Based on Current Signal Parameters Computing Band Energy Calculating Rate Determination Thresholds SO SO, SO 0, and SO Comparing Thresholds Performing Hangover Hangover Table for SO Hangover Table for SO, SO 0, and SO... - ix

12 GPP C.S00-D v Constraining Rate Selection Updating RDA Parameters Updating the Smoothed Band Energy Updating the Background Noise Estimate Updating the Signal Energy Estimate Mode Decision and Rate Decision for SO, SO 0, and SO Pitch-based NACF Pitch Based NACF Computation Decimation of Formant Residual (Current Frames and Look-Ahead Frames) NACF Computation for the Two Subframes of Current Frame NACF Computation for the Look-Ahead Frame Speech Coding Mode Decision Algorithm Energy and Zero Crossing Rate Rate-Decision for SO, SO 0, and SO Constraining Rate Selection Quantization of LSP Parameters for SO Computation of Weights Error Matrix Computation Adjustment of Quantization Error Quantization Search Generation of Quantized LSP Parameters Quantization of LSP Parameters for SO, SO 0, and SO Full-Rate, Half-Rate, and Quarter-Rate Unvoiced: Delta LSP VQ Full-Rate LSP and Unvoiced Half-Rate LSP VQ for SO 0 and SO WB Mode Half-Rate LSPVQ for SO, SO 0, and SO NB Mode Quarter-Rate Unvoiced LSPVQ for SO and SO NB Mode Quarter-Rate Voiced LSPVQ for SO and SO NB Mode QPPP LSP Quantization Weight Computation Constructing the MAVQ Memory During Frames That Are Not QPPP Eighth-Rate LSPVQ for SO, SO 0, and SO NB Mode Eighth-Rate LSPVQ SO 0 and SO WB Mode Encoding at Rates / and for SO, SO, SO 0, and SO LSP Quantization RCELP Shift State Update Delay Encoding Rates / and Subframe Processing Interpolation of LSP Parameters... - x

13 GPP C.S00-D v LSP to LPC Conversion Interpolated Delay Estimate Calculation Modification of the Original Residual Generation of the Weighted Modified Original Speech Vector Zero Input Response Calculation Impulse Response Calculation Calculation of the Adaptive Codebook Contribution Closed-Loop Gain Calculation Fixed Codebook Search Target Vector Generation Perceptual Domain Target Vector Conversion of the Target Vector to the Residual Domain Delay Calculation for Current Subframe Fixed Codebook Search Fixed Codebook Gain Quantization Combined Excitation Vector Computation Excitation Input to Highband Analysis (SO 0 and SO Wideband Mode) Encoder State Variable Update Computation of the Adaptive Codebook Contribution Delay Contour Computation Mapping of the Adaptive Codebook to the Delay Contour Closed Loop Delay Contour Adjustment Modification of the Residual Mapping of the Past Modified Residual to the Delay Contour Calculation of the Residual Shift Frame Parameters Search for Pulses in the Subframe Residual Location of the First Pulse in the Residual Location of a Pulse Inside of the Lag Window Shift Frame Boundary Calculation Shift Decision Peak to Average Ratio Calculation Matching the Residual to the Delay Contour Computation of the Shift Range Generation of a Temporary Modified Residual Signal for Matching Matching the Temporary Modified Residual to the Target Residual Adjustment of the Accumulated Shift Modification of the Residual Modified Target Residual Update... - xi

14 GPP C.S00-D v Computation of the ACELP Fixed Codebook Contribution for SO Algebraic Codebook Structure, Rate Algebraic Codebook Search Pre-Setting of Pulse Signs Non-Exhaustive Pulse Position Search Codeword Computation of the Algebraic Codebook Algebraic Codebook Structure, Rate / Computation of the Factorial Fixed Codebook Contribution for SO, SO 0, and SO Fixed Codebook (FCB) Structure, Rate Codebook Search Pre-Setting of Pulse Signs Non-Exhaustive Pulse Position Search Codeword Computation of the Factorial Codebook Converting a Code-Word to a Code-Vector Codebook Structure, Rate / Converting a Rate / Code-Word to a Code-Vector Fixed Codebook Gain Calculation Voiced Encoding (Full-Rate PPP and Quarter-Rate PPP) Modification of the Original Residual Prototype Pitch Period Processing Alignment Extraction between Two DFS, X and X Pole-Filtering a DFS Zero-Filtering a DFS Applying a Phase Shift to a DFS Energy of a DFS, X Converting a DFS from Cartesian to Polar Representation Converting a DFS from Polar to Cartesian Representation Normalizing the Band Energy of a DFS in Polar Representation Converting the DFS Amplitudes to Equivalent Rectangular Bandwidth (ERB) Amplitudes LPC Power-Spectrum Current to Previous PPP Energy Ratio Full Search Alignment between Two DFS Fine Search Alignment between Two DFS Weighted Alignment Search between Two DFS, X and X Zero Padding a DFS to Have the Same Lag as another DFS Band-Limited Correlation between Two DFS Speech Domain Energy of a DFS... - xii

15 GPP C.S00-D v Grouping DFS Harmonics into ERB Slots Zero Inserting into a DFS Peak to Average Ratio of PPP from its DFS Converting ERB Amplitudes to DFS Amplitudes Prototype Pitch Period Extraction Restoring PPP Memories if Last Frame Is Non-PPP or Full-Rate PPP Correlation between Current and Previous PPP Open-Loop Coding Scheme Change Quarter-Rate PPP Quantization Delta Delay Quantization ERB Differential Amplitude Quantization ERB Differential Amplitude Dequantization Evaluating the Quality of Quantization Obtaining the Dequantized Current PPP DFS Copying the Phase Spectrum of Previous PPP DFS Closed Loop Bump-Up from QPPP to Full-Rate CELP Coding Scheme FPPP DFS Amplitude Vector Quantization ERB Amplitude Dequantization Phase Quantization Global Alignment Band Alignment PPP Synthesis Calculating the Expected Alignment Shift Computing the Actual Alignment Shift and Shifting the Current PPP Computation of the Cubic Phase Track for the Synthesis of One Frame of Residual Generating the Whole Frame of Residual Perceptual Weighting Filter Update Encoding at Rate Generic Audio for SO 0 and SO LPC Analysis/Quantization Short-Term Residual Generation Formant Emphasis Modified Discrete Cosine Transform (MDCT) MDCT Residual Line Spectrum Quantization Global Scale Factor Quantization Noise Injection Inverse Modified Discrete Cosine Transform (IMDCT) Formant Deemphasis... - xiii

16 GPP C.S00-D v Encoding at Rate / Unvoiced for SO and SO NB and Rate / Unvoiced for SO 0 and SO WB Computation and Quantization of Gains Random Number Generation Creation of Sparse Nonzero Excitation Shaping the Excitation Perceptual Weighting Filter Update Encoding at Rate / LSP Quantization Interpolation of LSP Parameters LSP to LPC Conversion Impulse Response Computation Calculation of the Frame Energy Gain for SO Gain Quantization for SO Calculation of the frame energy gain Gain Quantization for SO, SO 0, and SO NB Mode Gain Quantization for SO 0 and SO WB Mode Generation of Rate / Excitation Perceptual Weighting Filter Update Random Number Generation Uniform Pseudo-Random Number Generation Algorithm Gaussian Pseudo-Random Number Generator Discontinuous Transmission (DTX) Coding the High-Frequency Band of Wideband Speech Declick Algorithm High-Band LPC Analysis and Quantization High-Band Excitation Generation Computation and Quantization of Gains for High-Band Excitation Packet Formatting for SO Packet Formatting for SO Special Half NELP Frame Packet Level Signaling Interworking Function Packet Formatting for SO Packet Level Signaling Interworking Function Packet Formatting for SO... - SPEECH DECODER...-. Frame Error Detection... - xiv

17 GPP C.S00-D v Received Packet Type Processing for SO Received Packet Type Processing for SO Received Packet Type Processing for SO Received Packet Type Processing for SO Packet Sanity Check for Bad Rate Detection for SO Delay Parameter Checking Delta Delay Parameter Checking Packet Sanitycheck for Badrate Detection for SO Delay Parameter Checking Delta Delay Parameter Checking LSP Checking Unused Bit Checking Invalid Filter ID Invalid Transition Rate / and CELP Decoding for SO, SO, SO 0, and SO Decoding of the LSP Parameters Delay Decoding and Frame Erasure Delay Contour Reconstruction Delay Decoding Frame Erasure Delay Contour Reconstruction for Rate for SO Frame Erasure Delay Contour Reconstruction for Rate Good Frame Following an Erased Frame for SO Delay Reconstruction Reconstruction of the Delay Contour Warping of the Adaptive Codebook Memory Frame Erasure Delay Contour Reconstruction for Rate Good Frame Following a QPPP frame Which Follows an Erased Frame in SO Delay Reconstruction Reconstruction of the Delay Contour Warping of the Adaptive Codebook Memory Smoothing of the Decoded Delay Rates / and Subframe Decoding Interpolation of LSP Parameters LSP to LPC Conversion Bandwidth Expansion Interpolated Delay Estimate Calculation Calculation of the Adaptive Codebook Contribution SO SO... - xv

18 GPP C.S00-D v Calculation of the Fixed Codebook Gain Computing of the Reconstructed ACELP Fixed Codebook Excitation for SO Computing of the Reconstructed Fixed Codebook Excitation in SO Decoder Total Excitation Generation Adaptive Codebook Memory Update Additional Excitation Frame Processing Frame Expansion or Compression in CELP Frames in SO PPP Smoothing of CELP Frame After Erasure in SO Pitch Prefilter (SO 0 and SO WB Mode) Synthesis of the Decoder Output Signal Rate and Rate / PPP Decoding (FPPP and QPPP) for SO Decoding of the LSP Parameters in FPPP Decoding of the LSP Parameters in QPPP Delay Decoding in FPPP Delay Decoding in QPPP Previous Frame Delay PPP (QPPP and FPPP) Interpolation of LSP Parameters and LSP to LPC Conversion Restoring PPP Memories If Last Frame Is Non-PPP or Full-Rate PPP ERB Amplitude Dequantization of Full-Rate PPP Phase Dequantization of Full-Rate PPP or Special Half-Rate PPP (Packet Level Signaling) Frame -..0 Obtaining the Dequantized Current DFS in QPPP Copying the Phase Spectrum of Previous PPP DFS PPP Synthesis Frame Expansion or Compression in PPP Synthesis Synthesis of the Decoder Output Signal in PPP Decoding at Rate Generic Audio for SO 0 and SO MDCT Residual Line Spectrum Decoding Global Scale Factor Decoding Noise Injection Decoding Inverse Modified Discrete Cosine Transform (IMDCT) Formant Deemphasis Decoding of Rate / Unvoiced Decoding or Special Rate / NELP Decoding SO and Rate / Unvoiced Decoding SO 0 and SO Frame Expansion or Compression in Rate / NELP Decoding Rate / Decoding for SO Decoding of LSP Parameters Decoding of the Frame Energy Vector... - xvi

19 GPP C.S00-D v Rate / Subframe Decoding Rate / Excitation Generation Interpolation of LSP Parameters LSP to LPC Conversion Synthesis of Decoder Output Signal Rate / Decoding for SO, SO 0, and SO DTX Off Mode: SO and SO Decoding of the LSP Parameters Decoding of the Frame Energy Gain for SO and SO 0 NB Mode Decoding of the Frame Energy Gain for SO 0 WB Mode Rate / Subframe Decoding Rate / Excitation Generation Interpolation of LSP LSP to LPC Conversion Estimation of the LPC Prediction Gain (for SO 0 WB Mode) Synthesis of Decoder Output Signal DTX On Mode: SO, SO 0, and SO Mechanism I Synthesizing Comfort Noise Generating Encoded Packet from Synthesized Noise Mechanism II Narrowband Mode Wideband Mode Adaptive Postfilter for SO Tilt Compensation Filter The Short Term Residual Filter The Long-Term Postfilter Gain Normalization and Short-Term Postfilter Adaptive Postfilter for SO, SO 0, and SO The Short-Term Residual Filter The Long-Term Postfilter Gain Normalization and Short-Term Postfilter Background Noise Estimation (for SO ) Frequency Domain Conversion Channel Energy Estimator Background Noise Estimate Background Noise Estimation (SO 0 and SO in WB Mode)... - xvii

20 GPP C.S00-D v Frequency Domain Conversion Channel Energy Estimator Background Noise Estimate Suppressed Eighth-Rate Likelihood Predictor Generation of High Band Signal for khz Decoding for SO 0 and SO Decoding of Synthesis Filterbank for khz Decoding for SO 0 and SO Processing the Lowband Signal, s LB Upsample-by- and Lowpass Filter Shaping Filter Processing the High Band Signal, s HB Interpolation to khz Spectral Reversal Interpolation to khz Resampling to khz Combining the Lowband and Highband... - TTY/TDD EXTENSION FOR SO...-. Introduction Overview TTY/TDD Extension TTY Onset Procedure Encoder TTY Onset Procedure Decoder TTY Onset Procedure TTY_MODE Processing TTY_SILENCE Processing TTY Header, Baud Rate, and Character Format Transporting the TTY Information in the Speech Packet Half-Rate TTY Mode Interoperability with. Baud-Only TTY Extensions Reflected Baudot Tones TTY/TDD Processing Recommendation TTY Encoder Processing TTY Encoder Inputs Dit Classification Dits to Bits TTY Character Classification TTY Baud Rate Determination TTY State Machine... - xviii

21 GPP C.S00-D v TTY/TDD Decoder Processing TTY Decoder Inputs Decoding the TTY/TDD Information Baudot Generator Tone Generator... - EVRC-B, EVRC-WB, AND EVRC-NW DATA CAPABILITIES...-. EVRC-B, EVRC-WB, and EVRC-NW Data Capabilities TTY/TDD Payload Format Baudot Code Payload DTMF Payload Format TTY/TDD Operation TTY Header and Character Format TTY Encoder Processing TTY Encoder Inputs Dit Classification Dits to Bits TTY Character Classification TTY Baud Rate Determination TTY State Machine TTY/TDD Decoder Processing TTY Decoder Inputs Decoding the TTY/TDD Information Baudot Generator Tone Generator DTMF DTMF Detector... - CODEBOOK MEMORIES AND CONSTANTS...- xix

22 GPP C.S00-D v LIST OF FIGURES Figure.- Speech Encoder Top-Level Diagram... - Figure.- Speech Encoder Top-Level Diagram... - Figure.- Speech Decoder Top-Level Diagram... - Figure...- Continuous Range of COPs... - Figure...- Illustration of the Principles of PPP Coding Figure...- Block Diagram of PPP Coding Scheme... - Figure...- A Block Diagram of Amplitude Quantization and Dequantization Modules... - Figure...- An Illustrative Non-Uniform Band Partition Used PPP Quantization... - Figure...- Illustration of the Effect of Spectral Downsampling and Upsampling... - Figure...- Demonstration of the Effect of a Single-Band Alignment on a Single PPP... - Figure...- Demonstration of the Effect of Multi-Band Alignments on a Single PPP... - Figure.- EVRC-WB Conceptual Overview... - Figure.- Wideband Encoder... - Figure.- Wideband Decoder... - Figure...- High-Band Analysis Figure...- Generation of the High-Band Excitation Signal... - Figure..- High-Band Synthesis... - Figure..- Noise Suppression Block Diagram... - Figure.- Analysis Filterbank... - Figure..- Pitch-Based NACF Computation... - Figure...- High Level Diagram of Mode Decision... - Figure...- nacf_at_pitch[]>voicedth Second Subframe NACF of the Current Frame Is High... - Figure...- nacf_at_pitch[]<unvoicedth Second Subframe NACF of the Current Frame Is Low... - Figure...- nacf_at_pitch[]>=unvoicedth && nacf_at_pitch[]<=voicedth... - Figure.- RCELP Encoding Block Diagram... - Figure.- Rate Generic Audio Encoding Block Diagram... - Figure.- Excitation Coding for Rate / Unvoiced Frames for SO and Rate / Unvoiced Frames for SO 0 and SO WB... - Figure..- Pre-processing of the High Band to Remove Clicks... - Figure..- Energy Burst Detector... - Figure..- Generation of the High-Band Excitation Signal Figure - SO Speech Decoder Top-Level Diagram... - Figure.- Rate Generic Audio Decoding Block Diagram... - Figure.- Excitation Decoding for Rate / NELP or Rate / Special NELP... - Figure.- Comfort Noise Synthesizer Block Diagram... - Figure.0- Background Noise Estimator Block Diagram... - xx

23 GPP C.S00-D v.0 Figure.- Background Noise Estimator Block Diagram... - Figure.- Synthesis Filterbank for SO 0 and SO khz Decoding...- Figure..- TTY/TDD Processing Block Diagram... - Figure.- TTY/TDD Processing Block Diagram... - xxi

24 GPP C.S00-D v LIST OF TABLES Table..- Packet Types Supplied by Service Options,, 0, and to the Multiplex Sublayer... - Table..- Packet Types Supplied by the Multiplex Sublayer to Service Options,, 0 and... - Table...- Fraction of Packets at Rate and Rate / with Rate Reduction... - Table...- Sequence Parameters for Rate Reduction... - Table..- Valid Service Configuration Attributes for Service Option... - Table...- Sequence Parameters for Rate Reduction... - Table...- Service Option Control Message Type-Specific Fields... - Table...- Fraction of Packets at Rate and Rate / with Rate Reduction... - Table.- Service Option Group Assignments... - Table..- Valid Service Configuration Attributes for Service Option... - Table...- Service Option Encoding Rate Control Parameters Table...- Service Option Control Message Type-Specific Fields... - Table...- Service Option Control Message Type-Specific Fields... - Table..- Valid Service Configuration Attributes for Service Option Table...- Service Option 0 Encoding Rate Control Parameters... - Table...- Service Option Control Message Type-Specific Fields... - Table...- Service Option Control Message Type-Specific Fields... - Table..- Valid Service Configuration Attributes for Service Option... - Table...- Service Option Encoding Rate Control Parameters... - Table...- Service Option Control Message Type-Specific Fields Table...- Service Option Control Message Type-Specific Fields... - Table.- Bit Allocations by Packet Type... - Table.- EVRC-B Coding Schemes... - Table.- Bit Allocations by Packet Type... - Table..- Bit Allocation... - Table...- LSP Interpolation Constants... - Table...- FIR Filter Coefficients Used for Band Energy Calculations... - Table...- Threshold Scale Factors as a Function of SNR Table...- Threshold Scale Factors as a Function of SNR Table...- Threshold scale Factors as a Function of SNR... - Table...- Threshold Scale Factors as a Function of SNR... - Table...- Hangover Frames as a Function of SNR... - Table...- Hangover Frames as a Function of SNR... - Table...- Hangover Frames as a Function of SNR... - Table...- Hangover Frames as a Function of SNR... - xxii

25 GPP C.S00-D v Table...- Features Used in Mode Decision... - Table..- Mapping of Speech Mode Decision to Encoding Rate and Mode... - Table.- LSP Parameter Splits... - Table.- Number of LSP Codebooks... - Table.0.- LSPVQ Parameters for Full- Rate and SO 0 and SO WB Unvoiced Half-Rate... - Table.0.- Multiple Candidate Search in Full-Rate and SO 0 and SO WB Unvoiced Half-Rate LSPVQ. - Table.0.- LSPVQ Parameters for Half-Rate... - Table.0.- Multiple Candidate Search in Half-Rate LSPVQ... - Table.0.- LSPVQ Parameters for Quarter-Rate Unvoiced... - Table.0.- LSPVQ Parameters for Eighth-Rate SO 0 and SO WB Mode... - Table...- Adaptive Codebook Gain Quantization Table... - Table...- Positions of Individual Pulses in the Rate Algebraic Codebook... - Table...- Codeword for the Track Orders... - Table...- Positions of Individual Pulses in the Rate / Algebraic Codebook Table...- Number of Pulses Generated by Factorial Codebooks Table...- Definitions for Full-Rate CELP Factorial Codebook Table...- Dynamic Bit Allocation of Excitation Indices Table...- Pulse Positions for Half-Rate CELP Codebook Table...- Skew Table for Half-Rate CELP Codebook Table...- Pulse Positions for Half-Rate CELP Codebook... - Table.- Packet Formats... - Table.0- Packet Formatting for Full-Rate CELP... - Table.0- Packet Formatting for Full-Rate PPP... - Table.0- Packet formatting for Half-Rate CELP... - Table.0- Packet Formatting for Special Half-Rate CELP... - Table.0- Packet Formatting for Special Half-Rate PPP... - Table.0- Packet Formatting for Special Half-Rate NELP... - Table.0- Packet Formatting for Quarter-Rate NELP... - Table.0- Packet Formatting for Quarter-Rate PPP... - Table.0- Packet Formatting for Eighth-Rate Table.- Packet Formatting for Full-Rate CELP in SO Interoperable Mode Table.- Packet Formatting for Full-Rate Wideband CELP Frame...- Table.- Packet Formatting for Full-Rate Wideband MDCT Frame... - Table.- Packet Formatting for Full-Rate Narrowband MDCT Frame... - Table.- Packet Formatting for Half-Rate Wideband NELP Frame... - Table.- Packet Formatting for Half-Rate CELP in SO Interoperable Mode... - Table.- Packet Formatting for Eighth-Rate in SO Interoperable Mode... - xxiii

26 GPP C.S00-D v Table.- Packet Formatting for Eighth-Rate in Wideband Mode... - Table.- Packet Formatting for Special Half-Rate Wideband CELP Frame (for Packet Level Signaling)... - Table.-0 Packet Formatting for Special Half-Rate CELP in SO Interoperable Mode... - Table.- Packet Formatting for Special Half-Rate NELP... - Table.- Packet Formatting for Special Half-Rate Wideband and Narrowband MDCT Frame (for Packet Level Signaling)... - Table..- Received Packet Type Decoding for SO... - Table..- Received Packet Type Decoding for SO... - Table..- Received Packet Type Decoding for SO Table..- Received Packet Type Decoding for SO... - Table..- Postfilter Coefficients... - Table..- TTY Header and Character Fields... - Table..- TTY Header and Character Bit Assignment... - Table...- Baud Rate Interoperability Matrix... - Table...- TTY Bit History Buffer... - Table...- tty_dec( ) History Buffer... - Table.- Bit Allocation for Rate / Data Frames... - Table.- Illegal Codeword Value... - Table.- Data Header Description... - Table.- TTY Type Field... - Table.- Rate / Packet for TTY/TDD Baudot Code Payload... - Table.- Rate / Packet with DTMF Payload... - Table.- DTMF Data Values... - Table..- Range of Values for TTY/TDD Baudot Header Field... - Table..- Range of Values for TTY/TDD Baudot Character Field... - Table..- Range of Values for TTY/TDD Baud Rate Field... - Table...- TTY Bit History Buffer... - Table...- tty_dec() History Buffer... - Table..- Touch-Tone Telephone Keypad... - Table - LSP Quantization Table, Rate, Codebook... - Table - LSP Quantization Table, Rate, Codebook... - Table - LSP Quantization Table, Rate, Codebook... - Table - LSP Quantization Table, Rate, Codebook... - Table - LSP Quantization Table, Rate /, Codebook... - Table - LSP Quantization Table, Rate /, Codebook... - Table - LSP Quantization Table, Rate /, Codebook... - Table - LSP Quantization Table, Rate /, Codebook... - xxiv

27 GPP C.S00-D v.0 0 Table - LSP Quantization Table, Rate /, Codebook... - Table -0 Interpolation Filter Coefficients, Cutoff=0. (/)... - Table - Interpolation Filter Coefficients, Cutoff=0. (/)... - Table - Interpolation Filter Coefficients, Cutoff=0. (/)... - Table - Interpolation Filter Coefficients, Cutoff=0. (/) Table - Interpolation Filter Coefficients, Cutoff=0. (/) Table - Fixed Codebook Gain Quantization, Rate Table - Fixed Codebook Gain Quantization, Rate /... - Table - Residual Shift Interpolation Filter Coefficients... - Table - Rate / Frame Energy Quantization... - xxv

28 GPP C.S00-D v.0 This page intentionally left blank. xxvi

29 GPP C.S00-D v GENERAL. Scope Service Options,, 0, and provide two-way voice communications between the base station and the mobile station using the dynamically variable data rate speech codec algorithm described in this standard. The transmitting speech codec takes voice samples and generates an encoded speech packet for every Traffic Channel frame. The receiving station generates a speech packet from every Traffic Channel frame and supplies it to the speech codec for decoding into voice samples. Speech codecs communicate at one of four rates corresponding to the 00 bps, 00 bps, 00 bps, and 00 bps frame rates. However, Service Options and 0 do not use 00 bps frame rate. The specifications defined in Sections and of this document provide the detailed algorithmic description of the EVRC. In the case of a discrepancy between the floating point and algorithmic descriptions, the reference floating point c-code will prevail. The specifications defined in Input Signal Preprocessing (see.), Determining the Data Rate (see.), and Adaptive Postfilter (see. and.) are optional for implementations intended for varying operational environments (such as in-vehicle hands-free). Any implementations which deviate from the algorithms specified in this standard, shall meet the minimum performance requirements defined in the associated Minimum Performance Specifications.. Requirements Language Shall and shall not identify requirements to be followed strictly to conform to this document and from which no deviation is permitted. Should and should not indicate that one of several possibilities is recommended as particularly suitable, without mentioning or excluding others, that a certain course of action is preferred but not necessarily required, or that (in the negative form) a certain possibility or course of action is discouraged but not prohibited. May and need not indicate a course of action permissible within the limits of the document. Can and cannot are used for statements of possibility and capability, whether material, physical or causal.. References The following standards contain provisions which, through reference in this text, constitute provisions of this Standard. At the time of publication, the editions indicated were valid. All standards are subject to revision, and parties to agreements based on this Standard are encouraged to investigate the possibility of applying the most recent editions of the standards indicated below. 0.. Normative References N. C.S000-E, Upper Layer (Layer ) Signalling Standard for cdma000 Spread Spectrum Systems, September 00. N. C.S00-0, Discontinuous Transmission (DTX) of Speech in cdma000 Systems, December 00. N. TIA--B (R00), Mobile Station-Base Station Compatibility Standard for Wideband Spread Spectrum Cellular Systems, October 00. Common speech codec terminology also uses the term frame to represent a quantum of processing. For Service Options,, 0, and, the speech codec frame corresponds to speech sampled over 0 ms. The speech samples are processed into a packet. This packet is transmitted in a Traffic Channel frame. -

30 GPP C.S00-D v.0 0 N. C.S00-E, Medium Access Control (MAC) Standard for cdma000 Spread Spectrum Systems, September 00. N. ANSI/EIA/TIA--A, Telecommunications Telephone Terminal Equipment Transmission Requirements for Digital Wireline Telephones, October. N. ITU-T Recommendation G., Pulse Code Modulation (PCM) of Voice Frequencies, November. N. ITU-T Recommendation G., Transmission Performance Characteristics of Pulse Code Modulation, November 00. N. IEEE Standard -00, IEEE Standard Methods for Measuring Transmission Performance of Analog and Digital Telephone Sets, Handsets, and Headsets, April 00. N. IEEE Standard - (R00), Method for Determining Objective Loudness Ratings of Telephone Connections, September 00. N0. C.S000-E, Physical Layer Standard for cdma000 Spread Spectrum Systems, September 00. N. ITU-T Recommendation V., Operational and interworking requirements for DCEs operating in the text telephone mode, November 000. N. ITU-T Recommendation P., Transmission Characteristics for Wideband (0-000 Hz) Digital Handset Telephones, June 00. N. ITU-T Recommendation P., Transmission Characteristics for Wideband (0-000 Hz) Digital Hands-free Telephony Terminals, June Informative References S. C.S00-0 v.0, Enhanced Variable Rate Codec (EVRC), December. S. C.R00-G, Administration of ParameterValue Assignments for cdma000 Spread Spectrum Standards, June 00. S. ITU-T Recommendation G., Digital network echo cancellers, March 00. R. Proakis, J. G. and Manolakis, D. G., Introduction to Digital Signal Processing, Macmillan, New York,. R. Rabiner, L. R. and Schafer, R. W., Digital Processing of Speech Signals, Prentice-Hall Inc., New Jersey. R. Crochiere, R. E. and Rabiner L. R., Multirate Digital Signal Processing, Prentice-Hall Inc., New Jersey. R. Oppenheim, A. V. and Schafer, R. W., Digital Signal Processing, Prentice-Hall Inc., New Jersey. R. Dellar, J. R., Proakis, J. G., Hansen, J. H. L., Discrete-Time Processing of Speech Signals, Macmillan, New York,.Reserve. R. Alejandro, A., Acoustical and Environmental Robustness in Automatic Speech Recognition, Kluwer Academic Publishers, Boston,. R. Atal, B. S. and Schroeder, M. R., Stochastic coding of speech at very low bit rates, Proc Int. Conf. Comm., Amsterdam,, pp 0-. R. Kleijn, W. B., Kroon, P., and Nahumi, D., The RCELP Speech-Coding Algorithm, European Transactions on Telecommunications, Vol, No. Sept/Oct, pp -. R. Nahumi, D and Kleijn, W. B., An Improved kb/s RCELP Coder, IEEE Workshop on Speech Coding,. R0. U. Mittal, J. P. Ashley, and E. M. Cruz-zeno, Low Complexity Joint Optimization of Excitation Parameters in Analysis-by-Synthesis Speech Coding, Proceedings Eurospeech 00, pp. -. R. Laflamme, C., Adoul, J-P., Salami, R., Morissette, S., and Mabilleau, P., kbps wideband speech coding technique based on algebraic CELP, Proc. ICASSP, pp. -. R. Salami, R., Laflamme, C., Adoul, J-P., et.al, A Toll Quality kbits Speech coder for Personal Communication Systems (PCS), IEEE Trans. on Vehicle Technology,. R. Ashley, J.P., Cruz-Zeno, E.M., Mittal, U., Peng,W., Wideband coding of speech using a scalable pulse codebook, 000 IEEE Workshop on Speech Coding, September 000, pp -0. R. Gerson, I.A.; Jasiuk, M.A., Techniques for improving the performance of CELP-type speech coders, Selected Areas in Communications, IEEE Journal, Volume 0, Issue, June Page(s):. -

31 GPP C.S00-D v.0.. Service Option Number The variable data rate two-way voice service option using the speech codec algorithm described by this standard shall use service option numbers,, 0, and and shall be called Service Options,, 0, and respectively... Allowable Delays.... Allowable Transmitting Speech Codec Encoding Delay The transmitting speech codec shall supply a packet to the multiplex sublayer no later than 0 ms after it has obtained the last input sample for the current speech frame Allowable Receiving Speech Codec Decoding Delay The receiving decoder shall generate the first sample of speech using parameters from a packet supplied to it by the multiplex sublayer not later than ms after being supplied the packet... Special Cases.... Blanked Packets A blanked frame occurs when the transmitting station uses the entire frame for either signaling traffic or secondary traffic. The EVRC does no special encode processing during the generation of a blank packet, i.e., the generated voice packet is simply not used. The decoder, in turn, treats a blank packet in the same manner as a frame erasure..... Null Traffic Channel Data A Rate / packet with all bits set to is considered as null Traffic Channel data. This packet is declared an erased packet and handled as described in Section. For EVRC-A, if more than consecutive all-ones Rate / packets are received, the decoder s output shall be muted until a valid packet is received All Zeros Packet Rate and Rate / packets with all bits set to 0 shall be considered erased frames and shall be handled as described in Section. In addition, for EVRC-B, EVRC-WB, and EVRC-NW, Rate / and Rate / packets with all bits set to '0' shall be considered erased and shall be handled as described in Section... Terms and Numeric Information ACB. Adaptive Codebook. ACELP. Algebraic Code Excited Linear Predictive Coding, the algorithm that is used by the EVRC to generate the stochastic component of the excitation. Autocorrelation Function. A function showing the relationship of a signal with a time-shifted version of itself. Base Station. A station in the Domestic Public Radio Telecommunications Service, other than a mobile station, used for radio communications with mobile stations. CELP. See Code Excited Linear Predictive Coding. -

32 GPP C.S00-D v Codec. The combination of an encoder and decoder in series (encoder/decoder). Code Excited Linear Predictive Coding (CELP). A speech coding algorithm. CELP codecs use codebook excitation, a long-term pitch prediction filter, and a short-term formant prediction filter. Codebook. A set of vectors used by the speech codec. For each speech codec codebook subframe, one particular vector is chosen and used to excite the speech codec s filters. The codebook vector is chosen to minimize the weighted error between the original and synthesized speech after the pitch and formant synthesis filter coefficients have been determined. Decoder. Generally, a device for the translation of a signal from a digital representation into an analog format. For this standard, a device which converts speech encoded in the format specified in this standard to analog or an equivalent PCM representation. DFS. Discrete Fourier Series DFT. See Discrete Fourier Transform. Discrete Fourier Transform (DFT). A method of transforming a time domain sequence into a corresponding frequency domain sequence. Encoder. Generally, a device for the translation of a signal into a digital representation. For this standard, a device which converts speech from an analog or its equivalent PCM representation to the digital representation described in this standard. EVRC. Enhanced Variable Rate Codec, a family of source-controlled variable rate codecs including EVRC-A, EVRC-B, EVRC-WB, and EVRC-NW. EVRC-A. A narrowband Enhanced Variable Rate Codec corresponding to speech service option. EVRC-B. A narrowband Enhanced Variable Rate Codec corresponding to speech service option, capable to operate at multiple average rates or capacity operating points. EVRC-WB. A wideband Enhanced Variable Rate Codec, corresponding to speech service option 0, capable to operate at one wideband mode and two narrowband modes interoperable with EVRC-B. EVRC-NW. A narrowband-wideband Enhanced Variable Rate Codec, corresponding to speech service option, combining the wideband capability of EVRC-WB, the multiple capacity operating point capability of EVRC-B, and capability for discontinuous transmission in circuit-switch domain. FCB. Fixed Codebook. FFT. See Fast Fourier Transform. Fast Fourier Transform (FFT). An efficient implementation of the Discrete Fourier Transform. Formant. A resonant frequency of the human vocal tract causing a peak in the short-term spectrum of speech. IDFT. See Inverse Discrete Fourier Transform. IIR Filter. An infinite-duration impulse response filter is a filter for which the output, in response to an impulse input, never totally dies away. This term is usually used in reference to digital filters. Interpolation. In the speech coder context, a means of smoothing the transitions of estimated parameters from one set to another. Usually a linear function. -