非规则LDPC码度分布优化算法研究  

Optimization of Degree Distribution Pairs of Irregular LDPC Codes

作　　者：马琦[1] 王海滨[1,2] 陈曾平[1] 

机构地区：[1]国防科技大学ATR实验室,湖南长沙410073 [2]海军飞行学院教研部,辽宁葫芦岛125001

出　　处：《计算机仿真》2009年第4期171-176,共6页Computer Simulation

基　　金：总装备部武器装备创新基金(6130524,7130524)

摘　　要：由于差分进化和微粒群优化都能实现非规则LDPC码最优度分布序列的搜索,因此有必要分析两种方法的特点和性能以便在特定应用环境中选择合适的方法。针对非规则LDPC码度分布的优化,介绍了差分进化和微粒群优化两种进化方式的极值搜索算法,分析了基于和积算法及其简化算法的密度进化理论,提出了可变步长的门限搜索方法。通过仿真,从算法复杂度和性能两方面比较了几种极值搜索算法和门限计算方法。这对促进LDPC码的理论分析和应用研究具有重要的指导意义,提出的可变步长门限搜索方法可大幅提高门限搜索速度。The Differential Evolution algorithm and the Particle Swarm Optimization algorithm both can be used to search the highest threshold of LDPC code, so the performance of the methods should be analyzed in order to select appropriate method under given conditions. In this paper, a method of optimization of irregular LDPC codes＇ dega＇ee distribution is discussed. The Differential Evolution algorithm and the Particle Swarm Optimization algorithm are ex- plained firstly, and the density evolution theory based on sum - product decoding algorithm and its simplfied algo- rithm is analyzed, then a threshold searching method with variable step is provided. With the simulation, the extremum searching methods are compared with the threshold searching methods in this paper, and some useful results are given at the end. This paper is helpful to the improvement of the theory and the application of LDPC codes and the proposed method can speed up the process of threshold searching.

关 键 词：低密度奇偶校验码 度分布对 密度进化 

分 类 号：TN911.22[电子电信—通信与信息系统]