摘要
针对Turbo乘积码(TPC:Turbo Product Code)距香农极限性能差距较大的问题,将非顺序(NS:NonSequential)译码推广到软输入软输出(SISO:Soft-Input/Soft-Output)Turbo乘积码译码器中,以提高误码率性能。该算法根据决定码字的可靠度选择更可靠的行或列译码,跳过低于可靠度门限的行或列,以避免迭代过程中引入额外错误而降低译码性能。仿真结果表明,对于以扩展汉明码(64,57,4)为子码的TPC码,NS-SISO译码算法的误码率仅为标准SISO算法的一半,可见新算法在误码率性能方面有明显提升。
In order to reduce the gap between the TPC(Turbo Product Code) and Shannon's limit performance,we generalized the Non-sequential decoding algorithm for SISO(Soft-Input/Soft-Output) decoding of TPC.The proposed algorithm only decodes the reliable component codes to avoid introducing extra errors.The simulation results show that the error rate of the NS-SISO decoding algorithm is only half of the standard SISO algorithm when Hamming code(64,57,4) is used as component code of TPC.It can be seen that the new algorithm has a visible error performance gain.
出处
《吉林大学学报(信息科学版)》
CAS
2018年第1期20-25,共6页
Journal of Jilin University(Information Science Edition)
基金
国家自然科学基金资助项目(61671173)
关键词
TURBO乘积码
迭代译码
非顺序译码
信道编码
turbo product codes(TPC)
iterative decodes
non-sequential decodes
channel coding