基于MRF的高性能二进制LDPC码译码算法

二进制LDPC码译码改进算法主要是提升硬判决性能或者降低软判决计算复杂度。本文应用高斯-马尔可夫随机场(Markov Random Field,MRF)模型实现信源参数估计,对信道译码端接收的比特序列进行对数似然比修正,在译码时加入信源的残留冗余信息来增加译码器的纠错能力。信源估计修正系数自适应可变,是由误码率参数调控。在计算复杂度不变的情况下,基于MRF的LDPC码译码算法有效提高了译码性能,降低误比特率

High-performance Decoding Algorithm for LDPC Code Based on MRF

LDPC code decoding algorithm is mainly to enhance the performance of a hard decision or soft decision to reduce the computational complexity. We Use Gauss - Markov random field model to achieve the source parameter estimation, the bit sequence channel decoder receives the logarithmic likelihood ratio correction, adding the source of information during the decoding residual redundancy to increase the decoder the error correction capability. Source estimated correction factor adaptive variable, is regulated by the BER parameter. In the case of the computational complexity of the same, based on the MRF LDPC code decoding algorithm can effectively improve the performance of the LDPC code decoding algorithm to reduce the bit error rate