《ldpc码的编译码算法研究及其量化分析论文论文》由会员分享，可在线阅读，更多相关《ldpc码的编译码算法研究及其量化分析论文论文（76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：621.39学校代码：10213 密级：公开工学硕士学位论文LDPC 码的编译码算法研究及其量化分析硕 士 研 究 生：李丹导师：张乃通教授申 请 学 位：工学硕士学科：信息与通信工程所 在 单 位：哈尔滨工业大学答 辩 日 期： 2009 年 12 月授予学位单位：哈尔滨工业大学Classified Index: TN911.22U.D.C: 621.39Dissertation for the Master Degree of EngineeringResearch On LDPC Encoding And Decoding Algorith
2、ms And Its Quantitative analysisCandidate： Supervisor：Academic Degree Applied for：Speciality：Affiliation：Date of Defence：Degree Conferring Institution：Li Dan Prof.Zhang NaitongMaster of EngineeringCommunication EngineeringHarbin Institute of TechnologyDecember, 2009Harbin Institute of Technology哈尔滨工
3、业大学工学硕士学位论文摘要低密度奇偶校验码（LDPC）相对于 Turbo 码的优异的译码性能和具有高译码吞吐量的可能使其成为 IEEE802.16e 标准的几种信道编码之一。IEEE802.16e 标准中的 LDPC 码是通过基于基础矩阵循环移位得到的一种结构性的 LDPC 码，这种特性使得其编码器和译码器的设计都相对简单，可以有更高的并行度和译码吞吐量。基于对 IEEE802.16e 标准和传统的 Efficient 编码算法的深入研究，提出了两种简化 Efficient 编码算法。在深刻理解消息传递算法的基础上，对概率测度的 BP译码算法、对数似然比测度的 BP 译码算法、校验似然比的 B
4、P 译码算法和最小和算法及其修正算法进行了推导，并对上述几种算法的性能进行了比较。采用理论分析与计算机仿真相结合的方法，研究了 LDPC 码的量化译码及其三种不同的硬件实现结构：全并行结构、串行结构和部分并行结构。仿真结果表明：当选取偏移量 0.3，修正最小和算法在瀑布区的误比特性能和 Log-BP 算法基本相同，而在平层区则有效的降低了 Log-BP 算法的错误平层约一个量级，而其复杂度却远小于Log-BP 算法的译码复杂度；串行译码算法的收敛速度约是并行译码算法的两倍，但其译码速率比较慢，为了解决这个矛盾而提出了部分并行的译码思想，使得硬件实现成为可能；对应三种不同的调度结构，给出其相应的
5、量化译码方案，这些方案能大大降低译码复杂度，同时量化译码性能非常接近连续译码；综合考虑，部分并行的修正最小和算法是今后研究的主要方向。关键词：LDPC码；编码；译码；串行调度；并行调度；串并结合调度；I哈尔滨工业大学工学硕士学位论文AbstractLow Density Parity-Check (LDPC) code has better decoding performance andhigher decoding throughput comparing to Turbo code. Its excellent performance makesLDPC code become an o
6、ptional channel code of IEEE802.16e standard. LDPC code inIEEE802.16e standard is an architecture-aware LDPC code and is created by circularright shift based on some base matrixes. These features make LDPC codes encoder anddecoder based on IEEE802.16e standard much simpler and have higher parallel d
7、egreeand decode throughput.Based on the research of IEEE802.16e standard and traditional encoding algorithmdeeply, two kinds of predigested encoding algorithms are proposed. Some decodingalgorithms including Belief Propagation (BP) decoding algorithm in probability domain,Log-domain and likelihood r
8、atio domain, Min-Sum decoding algorithm and modifiedMin-Sum algorithms are proved on the base of understanding massage passing algorithm.At the same time, the performances of these algorithms are compared. Furthermore, bytheoretical analysis and simulation, quantization decoding of LDPC codes and th
9、reedifferent hardware implementation architectures (full parallel architecture, partly parallelarchitecture and full serial architecture) are studied. The simulation results show that ifthe offset is reasonable ( 0.3 ), the bit error performance of modified Min-Sumalgorithm in waterfall area is almo
10、st the same as that of Log-BP algorithm and it canreduce the error floor of the Log-BP algorithm by almost one order of magnitude at highsignal-to-noise ratio (SNR), but its decoder complexity is far less than that of Log-BPalgorithm. The convergence speed of serial decoding is about two times faste
11、r than thatof the parallel decoding algorithm, but its decoding speed is slower. To resolve theconflict, partly parallel decoding algorithm is presented which makes hardwareimplementation possible. Corresponding to the three different hardware architectures,their quantization decoding schemes are pr
12、oposed which can greatly reduce the decodingcomplexity with a little performance loss. In a word, the partly parallel modifiedMin-Sum algorithm is the main investigative direction.KEY WORDS: LDPC codes, Encode, Decode, Serial schedule, Parallel schedule,Partly parallel schedule, QuantizationII哈尔滨工业大学工学硕士学位论文目录摘要 .- 21 -Abstract .II第 1 章 绪 论 . 11.1 信道编码概述. 11.2 信道模型和信道容量. 21.2.1 几种常见的信道模型 . 21.2.2 信道容量 . 31.3 LDPC 码的研究背景及研究现状 . 41.4 论文的研究内容. 6第 2 章 LDPC 码的编码方法 . 72.1 LDPC 码简介 . 72.1.1 LDPC 码的定义 .
