基于SOVA的固定时延咬尾卷积码译码算法

The Decoding Algorithm of Tailbiting Convolutional Code Based on SOVA

针对咬尾卷积码最大似然译码算法复杂度过高,循环维特比算法及其低复杂度改进算法译码延迟不固定的缺点,提出了一种基于软输出维特比译码(SOVA)的咬尾卷积码译码算法,算法在降低译码复杂度的同时使译码算法保持固定的译码延迟.算法的主要思想是:在正式译码之前,采用经过修改的SOVA算法确定编码寄存器的初始状态,进而把咬尾卷积码的译码算法转化为普通卷积码的译码算法.仿真结果表明,在误比特率性能上,该算法比较接近最大似然译码算法,并且优于循环维特比译码算法.

The complexity of maximum likelihood decoding algorithm for Tailbiting convolutional codes is too high.The delay of circular Viterbi algorithm and its improved algorithm of low complexity for Tailbiting convolutional codes is not fixed.In this paper,we propose a decoding algorithm based on the soft-output Viterbi-algorithm（SOVA）,this algorithm reduces the decoding complexity and has a fixed decoding delay.The main idea of the algorithm is to use a modified version of the SOVA to determine the initial state of the encoding register before the formal decoding,and then transform the decoding algorithm of the Tailbiting convolutional codes into the decoding algorithm of the common convolutional code.Simulation results are close to the performance of the maximum-likelihood decoding,and better than the circular Viterbi algorithm.