极化码序列连续删除译码算法的改进设计

极化码连续删除译码算法性能和传统的LDPC码存在一定差距。序列连续删除算法(SCL)的提出极大地改善译码性能,是极化码推向实际应用中的重要一步。但是该算法复杂度较高,延迟大。改进的序列连续删除(SCL)译码算法是基于改善极化码码长受限的情况,文中描述SCL算法是通过码树上的搜索序列路径来表示译码过程。改进的算法通过减少译码算法在码树上的序列路径来降低时间和空间复杂度。通过仿真表明,改进的算法有效地降低了译码的复杂度同时在性能上也接近最大似然(ML)译码算法。

Modified Successive Cancellation List Decoding Algorithm for Polar Codes

The decoding performance of polar codes successive cancellation (SC) algorithm could hardly match with that of traditional LDPC. The proposal of successive cancellation list (SCL) greatly improves the decoding performance,and thus is an important step to put the polar codes into practical application, and however, this algorithm is of high complexity and long latency.Modified SCL decoding algorithm aims to improve the finite-length performance of polar codes.This paper describes the decoding process with searching sequence of paths on code tree. This proposed algorithm can reduce the time and space complexity by reducing unnecessary path searching operations. Simulations indicate that the modified SCL decoding algorithm decreses the decoding complexity while approaching the performance of maximum likelihood (ML) decoding algorithm.