To reduce decoding computational complexity of nonbinary Low-Density Parity-Check (LDPC) codes, a weighted symbol-flipping decoding algorithm based on a new criterion is proposed. The flipped symbol is determined according to the symbol flipping function and the reliabilities of the received bits in the algorithm. The decoding procedure would be stopped in advance by analyzing the trend of the number of unsatisfied checks. The simulation results show that the new algorithm can tremendously reduces the average number of required iterations with negligible performance degradation compared to the symbol-flipping decoding algorithm. Thus it achieves an appealing tradeoff between performance and complexity.%为了降低多进制低密度奇偶校验(Low-Density Parity-Check,LDPC)码译码算法的复杂度,该文提出了基于新停止准则的符号翻转译码算法.该算法根据翻转函数和接收比特可靠性度量来确定对应的翻转符号,通过分析不满足校验方程个数的变化趋势来提前终止迭代.仿真结果表明,新算法在保持原有符号翻转译码算法误码性能不变的情况下,极大地减少了译码迭代次数,取得了译码性能和复杂度的折衷.
展开▼