一种基于遗传算法的RSC码盲识别方法

针对目前递归系统卷积(RSC)码盲识别算法容错性差、计算量大的问题,提出了基于遗传算法的RSC多项式参数盲识别算法。首先根据RSC码特殊的编码结构,构建了基于遗传算法的识别模型,将结果向量的码重作为适应度函数,然后推导出了不同误码率条件下平均码重的理论值,实现了算法中最优门限的获得。该算法容错性能较好,并且最大计算量只与初始种群的规模、遗传代数的上限以及输出路数成正比。最后仿真验证表明,理论推导的码重分布情况能够与仿真结果较好地吻合,并且在误码率高达0.06的情况下,各种寄存器个数下的RSC码参数识别率接近于0.9。

Blind identification of RSC code based on genetic algorithm

To address the problems of poor performance and heavy computation in blind identification of Recursive System-atic Convolutional (RSC) code ,a new algorithm for blind identification of RSC polynomial parameters is proposed based on the genetic algorithm .Considering the special structure of RSC code ,the identification model is constructed based on the genetic algorithm .The weight of the result vector is used as fitness function ,and the theoretical value of the average code weight is derived at different Bit Error Rates ,as the results .The optimal threshold is then obtained .The performance of the proposed algorithm is good ,and the maximum amount of calculation is only proportional to the initial population size ,genetic generations ,and paths of outputs .The simulation results show that the theoretical derivation of the code weight is in good a-greement with the simulation results ,and the recognition rate of the RSC code is close to 0 .9 at different number of registers when the Bit Error Rate is up to 0 .06 .