基于LDPC的联合调度译码算法的研究

Research on joint scheduling decoding algorithm based on LDPC

针对低密度奇偶校验码(LDPC)动态调度中存在的贪婪性以及非动态调度中译码性能低的问题,提出基于动态调度与非动态调度的联合优化算法(ELNWRBP)。该算法合理分配动态调度算法(NWRBP)和非动态调度算法(LBP)权重占比。NWRBP算法提高译码性能,而LBP算法缓解了动态调度中的“贪婪性”问题。两种调度算法联合使用,在译码性能和译码效率之间达到一种平衡。仿真结果和数据表明:在码字为(1296,1/2)、误码率为3×10^(-5)的条件下,相比于NWRBP算法和LBP算法的性能分别提高了约0.15 dB和0.45 dB。

In view of the greed in dynamic scheduling of low density parity check(LDPC)and the low decoding performance in non⁃dynamic scheduling,a joint optimization algorithm ELNWRBP(enhanced layered⁃NWRBP)based on dynamic scheduling and non⁃dynamic scheduling is proposed.In this algorithm,the weights of dynamic scheduling algorithm NWRBP(node⁃wise residual BP)and non⁃dynamic scheduling algorithm LBP(layered BP)are allocated reasonably.The NWRBP algorithm is to improve the decoding performance,while LBP algorithm is to alleviate the greed in dynamic scheduling.The combination of the two scheduling algorithms achieves a balance between decoding performance and decoding efficiency.The simulation results and data show that,in comparison with the NWRBP algorithm and the LBP algorithm,the performance of the proposed algorithm is improved by about 0.15 dB and 0.45 dB,respectively,under the condition that the code word is(1296,12)and the bit error rate(BER)is 3×10^(-5).