基于多项式遍历和码字相关的Turbo码盲识别

摘要

针对1/n码率Turbo码的盲识别问题,提出了一种基于多项式遍历和码字相关的检测识别方法.该方法首先对码字序列进行分组,利用分组验证的方法对数据中的卷积码进行快速检测,进而识别出Turbo码的码率;然后,利用欧几里得算法对Turbo码的分量编码器a的参数进行识别,进而通过对编码器b的遍历,恢复出伪随机交织序列;最后,通过与信息序列的码字相关,验证编码器b的参数是否正确,同时利用相关谱峰值所在位置识别交织器参数,实现了Turbo码的全部参数估计.仿真实验验证了该算法的有效性.对算法的误码适应能力进行的仿真分析表明,该方法能够在较高误码率条件下实现Turbo码的检测与识别.%For blind recognition of 1/n rate Turbo code,an algorithm based on polynomial traverse and codes correlation is proposed.Firstly,the convolutional code is detected based on grouping check method,and the code rate of Turbo code is identified.Then the parameters of RSC(a) are estimated based on Euclidean algorithm,and the interleavered sequence is recovered by traverse of the RSC(b).The parameters of RSC(b) are validated based on the first order statistical test,and finally the parameters of interleaver are estimated by the correlation spectrum.The validity of algorithm is verified by the simulation results.Case studies are presented to illustrate that the method can recognize the Turbo code in a high noisy environment.