低复杂度的Turbo乘积码迭代译码研究

Research of low complexity iterative decoding for Turbo product codes

Turbo乘积码(简称TPC码)的传统迭代译码算法寻找竞争码字难、软信息存储量大。针对这些问题提出一种低复杂度的迭代译码算法,并得出新的译码器迭代结构。该算法在Chase迭代SISO译码的基础上,采用无需寻找竞争码字的相关运算来简化软输出信息的计算,同时用前一个迭代译码单元的软输入信息替换传统算法中信道原始接收信息,然后与当前迭代译码单元的软输出信息直接进行线性叠加后作为下一个迭代译码单元的软输入,从而简化了软输入信息的计算和系统存储量。仿真结果验证了该算法的可行性和有效性。

For the traditional iterative decoding algorithm of Turbo Product Codes（TPC）,it＇s hard to search competitive codeword and it needs so many memory capacities of soft information.This paper presents a very low complexity iterative decoding algorithm to solve these problems,and obtains a new iterative decoder structure.It is based on the iterative SISO Chase algorithm,and a correlation operation is used to simplify the calculation of soft output information.Thus there is no need to find competitive codeword.Meanwhile,let the soft inputs for the previous decoding unit replace the traditional received information from channel,and then obtain soft inputs for the next iterative decoding unit by a linear combination of soft outputs for the current iterative decoding unit.Thereby the calculation of soft input information and memory capacities are simplified.Simulation results show the feasibility and effectiveness of the algorithm.