ANSI TIA EIA-136-210-A-2001 TDMA Third Generation Wireless - ACELP Minimum Performance.pdf

1、 TIA/EIA STANDARD TDMA Third Generation Wireless - ACELP Minimum Performance TIA/EIA-136-210-A (Revision of TIA/EIA-136-210) APRIL 2001 TELECOMMUNICATIONS INDUSTRY ASSOCIATION The Telecommunications Industry Association represents the communications sector of ANSI/TIA/EIA-136-210-A-2001 Approved: Ap

2、ril 23, 2001 TIA/EIA-136-210-ANOTICE TIA/EIA Engineering Standards and Publications are designed to serve the public interest through eliminating misunderstandings between manufacturers and purchasers, facilitating interchangeability and improvement of products, and assisting the purchaser in select

3、ing and obtaining with minimum delay the proper product for his particular need. Existence of such Standards and Publications shall not in any respect preclude any member or nonmember of TIA/EIA from manufacturing or selling products not conforming to such Standards and Publications, nor shall the e

4、xistence of such Standards and Publications preclude their voluntary use by those other than TIA/EIA members, whether the standard is to be used either domestically or internationally. Standards and Public ations are adopted by TIA/EIA in accordance with the American National Standards Institute (AN

5、SI) patent policy. By such action, TIA/EIA does not assume any liability to any patent owner, nor does it assume any obligation whatever to parties adopting the Standard or Publication. Note: The users attention is called to the possibility that compliance with this document may require use of one o

6、r more inventions covered by patent rights. By publication of this standard, Interim Standard, or TSB, no position is taken with respect to the validity of those claims or of any patent rights in connection therewith. The patent holder(s) so far identified have, however, filed statements of willingn

7、ess to grant licenses under those rights on reasonable and nondiscriminatory terms and conditions to applicants desiring to obtain such licenses. Details may be obtained from the standards developer. This Standard does not purport to address all safety problems associated with its use or all applica

8、ble regulatory requirements. It is the responsibility of the user of this Standard to establish appropriate safety and health practices and to determine the applicability of regulatory limitations before its use. (From Standards Proposal No. 4027-210-A and 4027-210-A+1, formulated under the cognizan

9、ce of the TIA TR-45.3 Subcommittee on Time Division Digital Technology.) Published by TELECOMMUNICATIONS INDUSTRY ASSOCIATION 2001 Standards and Technology Department 2500 Wilson Boulevard Arlington, VA 22201 PRICE: Please refer to current Catalog of EIA ELECTRONIC INDUSTRIES ALLIANCE STANDARDS and

10、ENGINEERING PUBLICATIONS or call Global Engineering Documents, USA and Canada (1-800-854-7179) International (303-397-7956) All rights reserved Printed in U.S.A. PLEASE! DONT VIOLATE THE LAW! This document is copyrighted by the TIA and may not be reproduced without permission. Organizations may obta

11、in permission to reproduce a limited number of copies through entering into a license agreement. For information, contact: Global Engineering Documents 15 Inverness Way East Englewood, CO 80112-5704 or call U.S.A. and Canada 1-800-854-7179, International (303) 397-7956 TIA/EIA-136-210-Aiii1. INTRODU

12、CTION11.1 SCOPE .21.2 DEFINITIONS31.3 MODEL FOR SUBJECTIVE TEST OF THE SPEECH CODEC.62. CODEC MINIMUM STANDARDS.82.1 STRICT OBJECTIVE PERFORMANCE REQUIREMENTS .82.2 SUBJECTIVE PERFORMANCE REQUIREMENTS92.2.1 Definition .92.2.2 Description of Procedure.102.2.2.1 Test Conditions 102.2.2.2 Source Speech

13、 Material112.2.2.3 Processing of Speech Material .112.2.2.4 Randomization112.2.2.5 Presentation 122.2.2.6 Listeners .122.2.2.7 Listening Test Procedure122.2.2.8 Analysis of Results .132.2.2.9 Computational Procedure .142.2.3 Minimum Standard.162.2.4 Expected Results for Reference Conditions183. STAN

14、DARD TEST CONDITIONS203.1 STANDARD EQUIPMENT 203.1.1 Basic Equipment 203.1.2 Subjective Test Equipment .223.1.2.1 Audio path233.1.2.2 Calibration233.2 STANDARD ENVIRONMENT TEST CONDITIONS 233.3 STANDARD SOFTWARE TEST TOOLS .233.3.1 Analysis of Subjective Performance Test Results.243.3.1.1 Analysis o

15、f Variance 243.3.1.2 Newman-Keuls Comparison of Means.253.3.2 Randomization routine for subjective test253.3.2.1 Algorithm .253.3.2.2 Usage263.4 STANDARD SOURCE SPEECH FILES .303.4.1 Source Material303.4.2 Speech Files for MOS Testing303.4.3 Calibration file.323.4.4 Software Tools for Generating Spe

16、ech Files334. CNG MINIMUM STANDARDS 344.1 DELAY REQUIREMENTS.344.2 PERFORMANCE REQUIREMENTS344.2.1 Test Cases 344.2.1.1 Source Material 354.2.1.2 Truth Mark Files.364.2.1.3 Encoder Serial Output Files364.2.1.4 Reference Bit-Exact Comfort Noise Files 374.2.2 Test Procedure .384.2.2.1 Test Comfort Noi

17、se Files Generation .384.2.2.2 Reference/Test Comfort Noise File Comparisons 394.2.2.2.1 Log Spectral Distortion (LSD) 394.2.2.2.2 Log Energy Deviation (LED) 394.2.2.2.3 Calling Sequence.404.2.2.3 Statistical Analysis and Compliance Determination.404.2.2.3.1 Calling Sequence.40TIA/EIA-136-210-Aiv5.

18、CONTENTS OF SOFTWARE DISTRIBUTION. 436. TIA/EIA-136-410 ENHANCED FULL-RATE ENCODER OUTPUT FORMAT. 487. CHANGE HISTORY 518. REFERENCES 52TIA/EIA-136-210-A11. IntroductionThis standard details definitions, methods of measurement, and1minimum performance characteristics of enhanced full-rate speech2cod

19、ecs (TIA/EIA-136-410) for TIA/EIA-136 Base Stations that support3digital voice traffic based on this codec. This standard only applies to4Base Stations that employ non-bit-exact implementations of TIA/EIA-5136-410 compatible codec. It does not apply to Mobile Stations since6only bit-exact implementa

20、tions of the enhanced full-rate codec are7permitted at this end. This standard is comprised of two parts. The first8part as detailed in 2 and 3 covers enhanced full-rate speech encoding9and decoding operations in Continuous Transmission (CTX) or10Discontinuous Transmission High (DTX-high) modes of o

21、peration.11The second part as detailed in 4 covers Comfort Noise Generation12(CNG) operation of the Enhanced Full-Rate Speech decoder during13Discontinuous Transmission Low (DTX-low) mode of operation.14The first part of this standard (2 and 3) which covers normal15enhanced full-rate speech encoding

22、 and decoding operations, is largely16derived from TIA/EIA-635 which specifies the procedures to be17employed to test implementations of TIA/EIA-627 compatible full rate18speech codecs and to verify that they meet recommended minimum19performance standards.20This standard consists of this document a

23、nd software containing speech21source material, calibration source material, software data analysis tools22including a spreadsheet tool for data analysis. The software distribution23also contains specifically for CNG reference comfort noise data files,24truth mark files, encoder serial output data f

24、iles, and a set of C source25files for a software tool to produce and to evaluate the performance26metrics. A detailed description of contents and formats of the software27distribution is given in 5 of this document. The software distribution28shall accompany this standard, for example, on the same

25、CD-ROM,29from the same Internet site from where this document was obtained.30Compatibility, as used in connection with this standard and TIA/EIA-31136-410, is understood to mean any Mobile Station is able to place and32receive calls in any cellular system. Conversely, all cellular systems are33able

26、to place and receive calls with any Mobile Station. To assure34compatibility, it is essential that speech processing parameters be35specified.36The enhanced full rate speech codec is intended to be used at both37TIA/EIA-136 Mobile Stations and at compatible Base Stations in the38Cellular Service. Th

27、is standard is meant to define the recommended39minimum performance requirements of TIA/EIA-136-410 compatible40non-bit-exact implementations of the enhanced full-rate codecs at the41Base Station end. It does not apply to bit-exact implementations of the42enhanced full-rate codec. Since bit-exact im

28、plementation is mandatory43at the Mobile Station end, this standard does not apply to Mobile44Stations. By Base Station end, we also include implementations in45TIA/EIA-136-210-A2which codecs are placed at a Mobile Switching Center or elsewhere at1the fixed end of the cellular system.2Although the b

29、asic purpose of cellular telecommunications has been3voice communication, evolving usages (for example, data) may allow4the omission of some of the features specified herein provided that5system compatibility is not compromised.61.1 Scope7This document defines the minimum performance specifications

30、for8non-bit-exact implementations of TIA/EIA-136-410 speech codecs9operating in CTX (DTX-high) and CNG (DTX-low) operation under10channel error free conditions. This speech codec is the Algebraic11Codebook Excited Linear Predictive coding (ACELP) system described12in TIA/EIA-136-410 document, i.e.,

31、TDMA Cellular/PCS - Radio13Interface - Enhanced Full-Rate Voice Codec, which is accompanied by14a soft copy consisting of fixed point ANSI C code and a number of test15sequences. The ACELP speech codec is used to digitally encode the16speech signal for transmission at a data rate of 7400 bps or 148

32、bits17every 20 ms.18The TIA/EIA-136-410 standard includes a bit-exact specification of the19ACELP speech codec using a fixed point ANSI C code. Six sets of test20sequences are provided along with this standard. These test sequences21are designed to provide for maximum coverage of the fixed point ANS

33、I22C code. There are separate test sequences for testing the encoder and23for testing the decoder portions of the TIA/EIA-136-410 standard24speech codec. A bit-exact implementation of this standard on a fixed25point device would have to pass all six test sequences for the encoder26and the six test s

34、equences for the decoder in order to comply with the27TIA/EIA-136-410 standard. These test sequences are described in28Annex A of TIA/EIA-136-410 standard. As mentioned before,29implementations of the TIA/EIA-136-410 standard at the Mobile30Stations must pass all these test sequences since they are

35、required to be31bit-exact.32At the Base Station end, non-bit-exact implementations of the TIA/EIA-33136-410 standard are permitted. For such implementations, it may not34be possible to generate test vectors that can guarantee compatibility35with the standard by using them as input to the speech enco

36、der or36decoder and by examining their output for exact replication of the37corresponding reference vector. This document describes a series of38tests which are used to test conformance to the specification. These39tests do not necessarily ensure that the codec operates satisfactorily40under all pos

37、sible input signals and it is still a requirement of the41manufacturer to ensure that his implementation operates in a consistent42manner. The purpose of these tests is to test for minimum performance43levels and the manufacturer should attempt to provide the highest44performance possible.45TIA/EIA-

38、136-210-A3The testing of any non-bit-exact implementation of the codec operating1in CTX or DTX-high is based upon two classes of procedures:2Objective tests, which are based upon actual measurements of 3the speech codec function.4Subjective tests, which are based upon listening tests to judge 5overa

39、ll speech quality.6Minimum performance specification for CNG is based on objective7tests. These tests are defined by the test database, a method for8generating individual test data files, the test requirements on candidate9CNGs, a method to collect performance metrics, and the means to test10those m

40、etrics so as to validate conformance to the specification.11The main purpose of the testing is to ensure adequate performance12between any Base Stations encoder and any Mobile Stations decoder13and vice-versa. Since the Mobile Station is restricted to have bit-exact14implementations of the TIA/EIA-1

41、36-410 standard only, it suffices to15demonstrate that minimum performance standards have been met when16inter-operating with any bit-exact implementation. For this purpose, the17fixed point ANSI C code provided as a part of the TIA/EIA-136-41018standard can be treated as the reference or master cod

42、er.19The restriction that Mobile Stations can only have bit-exact20implementations of the TIA/EIA-136-410 standard is the primary21reason that allows this standard to ensure minimum performance22standards between any pairing of Base Stations and Mobile Stations;23otherwise, we can guarantee this onl

43、y by testing all combinations of24encoder/decoder pairings in both Base Stations and Mobile Stations.25With the number of equipment manufacturers expected to supply26equipment this could easily become a prohibitive task.27By convention in this document, the Courier font is used to indicate28C langua

44、ge and other software constructs such as file and variable29names. The diacritical mark, , is used to indicate section numbers.301.2 Definitions31ACELP Algebraic Code Excited Linear Predictive coding. A form of CELP that utilizes32an algebraic codebook structure.33BER Bit Error Rate - Equals (# of r

45、eceived bit errors)/(Total # of transmitted bits).34BFI Bad Frame Indicator for a speech frame35BFI_CN Bad Frame Indicator for a comfort noise frame36CELP Code Excited Linear Predictive Coding. These techniques use codebooks to37vector quantize the excitation (residual) signal of a Linear Predictive

46、 Codec38(LPC).39TIA/EIA-136-210-A4CNG Comfort Noise Generator, Comfort Noise Generation or Comfort Noise1Synthesis used with the TIA/EIA-136-410 enhanced full-rate vocoder operating2in discontinuous transmission mode.3Codec The combination of an encoder and decoder in series (encoder/decoder).4Coder

47、 Same as “encoder“.5Comfort Noise Parameter Generation. This refers to the state of the TIA/EIA-136-410 enhanced full-6rate encoder during DTX-low mode of operation in which the encoder models7the background noise at the encoder input by a set of parameters for Comfort8Noise Generation at the far en

48、d decoder.9CTX Continuous Transmission. This refers to the state in which mobile stations10transmit into the voice traffic channel TDMA frames containing TIA/EIA-136-11410 enhanced full-rate encoded data.12dBA A-weighted sound pressure level expressed in dB obtained by the use of a13metering charact

49、eristic and the weighting A specified in ANSI S1.4-1983, and14the addendum ANSI S1.4a-1985.15dBPa Sound level with respect to 1 Pascal, 20 log10(Pressure / 1 Pa).16dBm Power relative to 1 milliWatt.17dBm0 ITU-T Recommendation G.711 specifies a theoretical load capacity with a full18scale sine wave to be +3.17 dBm0 for -law PCM coding. In this standard,19+3.17 dBm0 is also defined to be the level of a full amplitude sine wave in the2016 bit twos complement notation used for the data files.21dBov Sound level in decibels with respect to 16-bit overload (the maximum value in