低复杂度的TPC自适应译码算法

A low-complexity adaptive decoding algorithm for Turbo product code

为降低TPC译码算法复杂度,提出一种新的低复杂度自适应译码算法。新算法适用于子码为扩展汉明码的TPC码,以不估计SNR的自适应译码算法为基本框架结构,利用码字可靠性特征,引入1个可用于外信息计算的简单公式,在自适应减小不可靠比特数的同时,降低外信息的计算量。研究结果表明:对于扩展汉明码(64,57,4)为子码的TPC码,新算法相比于原自适应译码算法在误码率为10-5时Eb/N0仅降低了0.05 d B,复杂度却降低约1/3,可见新算法在性能和复杂度方面实现了很好的平衡和折中。

A new efficient adaptive decoding algorithm was proposed to reduce the complexity of Turbo product code （TPC） decoding algorithm. The new algorithm was suitable for the TPCs formed by extended Hamming codes. The adaptive decoding algorithm without estimating SNR was the basic framework of the proposed algorithm and a simple formula for calculating the extrinsic information was introduced after analyzing codeword reliability. In the proposed algorithm, the least reliable bits could be reduced adaptively; meanwhile, computation of the extrinsic information should be simplified. The results show that when the extended Hamming code （64,5714） is the subcode of TPC and the bit error rate is 10-5, the performance loss of the proposed algorithm is only 0.05 dB compared with the adaptive decoding algorithm without estimating SNR. But the complexity is decreased by about 1/3. So the proposed algorithm is an excellent balance and compromise between complexity and performance.