系统极化码的置信传播译码性能分析

摘要

Belief propagation( BP) algorithm can generate soft information for systematic polar codes( SPC) to make decoding decisions and the soft information also can be used as extrinsic information in turbo deco-ding. This paper first provides a detailed description of BP algorithm based on the structure of channel po-larization. Then,it introduces two decision methods,one of which is by exploiting the soft information while the other one is by using polar encoding. The decoding performance under these two methods are compared by simulation,and the simulation results demonstrate that a better bit error rate( BER) performance can be obtained if the soft information is used,while the frame error rate( FER) performance is slightly improved as well in high signal-to-noise ratio( SNR) regime.%置信传播( BP)算法可以为系统极化码提供软信息作为判决依据，也可以为系统极化码在级联迭代译码中提供交换软信息。在详细描述基于信道极化结构的置信传播算法基础上，比较了系统极化码在软信息判决方法和极化编码判决方法下错误率性能的差异。仿真结果表明，软信息判决方法可以提高系统极化码的误比特率，在高信噪比下误帧率方面也略有提高。