《turbo乘积码译码算法研究(1)》由会员分享，可在线阅读，更多相关《turbo乘积码译码算法研究(1)（62页珍藏版）》请在金锄头文库上搜索。
1、国内图书分类号：TN911.22国际图书分类号：621.394.67工学硕士学位论文Turbo 乘积码译码算法研究硕 士 研究生：孙丽楠导申 请师：学 位：王学东教授工学硕士学 科、专 业：信息与通信工程所 在答 辩单 位：日 期：电子与信息技术研究院2007 年 7 月授予学位单位：哈尔滨工业大学Classified Index: TN911.22U.D.C: 621.394.67Dissertation for the Masters Degree in EngineeringSTUDIES ON DECODING ALGORITHMOF TURBO PRODUCT CODECandida
2、te:Supervisor:Academic Degree Applied for:Specialty:Affiliation:Date of Defence:Degree-Conferring-Institution:Sun LinanProf. Wang XuedongMaster of EngineeringInformation and CommunicationEngineeringSchool of Electronics andInformation TechnologyJuly, 2007Harbin Institute of Technology哈尔滨工业大学工学硕士学位论文
3、摘要提高通信的可靠性一直是通信系统设计所追求的最主要目标之一，纠错码是提高信息传输可靠性的一种重要手段，Turbo码是一种性能卓越的前向纠错码。自1993年Turbo码提出以来，众多学者发表了大量相关研究论文。Turbo乘积码(TPC)作为一种分组的Turbo码，把码块的行和列都加以编码，使得行和列都有纠检错的能力。与Turbo卷积码(TCC)相比，TPC实现复杂度低，时延小，性能相近，在 105比特误码率下不具有误码平层效应。可以说，TPC是未来通信系统中一种先进的前向纠错信道编码方式，有广阔的应用前景。本文首先介绍了 TPC的提出与发展过程，并对国内外研究现状及前景进行了阐述。描述了 TP
4、C的基本结构，分析了二维 TPC的纠错能力，讨论了译码原理，比较了几种译码算法的优缺点，重点分析了 Cyclic-2 PML算法及 Chase II算法的基本原理。在对译码算法分析研究的基础上，设计了分别以 Cyclic-2 PML算法和Chase II算法作为单元译码器的 TPC软输入软输出迭代译码器的流程和方案，并在 MATLAB环境下实现了基于这两种译码算法的 TPC译码器。对Cyclic-2 PML算法重点设计了邻近码字的生成方法及差值度量的计算结构。对 Chase II算法主要设计了可信度的排序和软信息的计算流程。对两种译码器的复杂度进行了分析和比较。最后在 MATLAB平台上构建了
5、 AWGN信道下的误码性能仿真系统，并对不同码块长度、迭代次数、硬判决 /软判决、量化比特数等条件下的TPC进行了误码性能的仿真，分析了以上因素对误码性能的影响。在不同子码构成 TPC的情况下，比较了由上述两种译码算法构造的 TPC译码器的性能。关键词Turbo乘积码；迭代译码；Cyclic-2 PML算法；Chase II算法- I -哈尔滨工业大学工学硕士学位论文AbstractOne of the main intentions of the design of communication system is toenhance the reliability of communicat
6、ion. Error-correcting code is an importantmeasure to enhance the reliability of information transmission. Turbo Codes is akind of forward error correction codes with excellent performance. Since it wasintroduced in 1993, many research thesis on Turbo decoding have been published.As a kind of block T
7、urbo codes, Turbo product codes encode each row andcolumon of an information block to make all the rows and columons detect andcorrect errors. Compared with Turbo convolutional codes(TCC), TPC offerslower complexity, less delaly, almost same performance and does not suffer fromthe error floor associ
8、ated with Turbo codes below BER of 105 . Therefore, TPCis well-suited to be implemented as an advanced FEC channel encodingtechnique for future communication systems and very attractive for futureapplications.The first objective of this paper is the introduction of the produce,improvement, internati
9、onal actuality and development tendency of TPC. Then wedescribe the structure of TPC, analyze the error correcting capability of 2D-TPC,discuss the decoding algorithms, compare the performances of several decodingalgorithms, and particularly describe decoding theories of Cyclic-2 PML andChase II alg
10、orithms.The soft input and soft output iterative decoder is designed using Cyclic-2PML and Chase II algorithm as the sub-decoder respectively, and implementedunder MATLAB environment, based on the research of decoding algorithms. Thecreation of nearby code words and computation of difference metric
11、areimportante to the design of Cyclic-2 PML algrothim, and the key steps of thedesign of Chase II are the reorder of the code letters by their reliabilities, and thecomputation of the soft imformation. Here we also analyze and compare thecomplexity of the two decoders.At last, we build simulation sy
12、stem based on AWGN channel, and simulateusing MATLAB software under different conditions such as code lengths,iterative times, hard/soft decision and quantification bits. We analyze bit errorrate performance which is influnced by those factors. Finally, we compare the- II -哈尔滨工业大学工学硕士学位论文performance of the two decoders mentioned before, based on the condition thatTPC is composed by different kind of sub-codewords.KeywordsTurbo product codes, iterative decoding, Cyclic-2 PML algorithm,Chase II algorithm- III-哈尔滨工业大学工学硕士学位论文目录摘要 . IAbstract . II第 1章绪论 .11.1课题背景 .
