基于不同SISO算法的TC编译码器仿真及其性能分析  

Simulation and performance Analysis of SISO Algorithm-based TC Codec

作　　者：徐友云[1] 吴灏[1] 宋文涛[1] 罗汉文[1] 

机构地区：[1]上海交通大学

出　　处：《计算机与网络》1999年第Z1期44-45,共2页Computer & Network

摘　　要：软输入软输出(SISO)译码器既是构成Turbo Codes编译码器的核心部件,又是导致设计成本和译码延时的主要因素之一。文章介绍两种用于递归卷积码译码的改进的软输入软输出最大后验概率译码算法MAP和SOVA,并对这两种算法改进前后的复杂度和相应的译码延时进行对比分析;根据基于这两种算法的TurboCodes编译码器的计算机仿真结果,就算法的计算量、实现成本、延时开销和纠错性能等问题进行了综合分析,对进一步简化算法的可能性做了探讨;最后对Turbo Codes编译码器的工程实现给出探索性建议。Soft - input soft - output (SISO) decoder is the kernel of the Turbo Codec, but also is one of the key facts of resulting in device cost and decoding latency. This paper presents two modified SISO maximum a posteriori decoding algorithms for recursive systematic convolutional codes, which are called MAP and SOVA. Then comparisons on the complexity and decoding latency of two algorithms before and after modified are given. According to the results of the computer simulations of the turbo codecs with the two algorithms, a comprehensive analysis about the amount of computation, the implementing cost, the decoding latency and the error - correcting performance of the two algorithms is followed. Meanwhile it discusses the probability of the further simplifying for the algorithms. Finally, an exploratory suggestion on the engineering implementation of turbo codecs is put forward.

关 键 词：TURBO CODES 软输入软输出算法 译码延时 纠错性能 数字通信 

分 类 号：TN764[电子电信—电路与系统]