RS码编码参数的盲识别

针对现有的RS(Reed-Solomon)码盲识别计算复杂度较大的问题,提出了一种新的识别方法.首先统计不同码长分组时的码重分布,并定义与理论码重分布之间的相似度系数,通过计算找出最相似的一组即对应正确的码长;然后建立二元假设,并确定判决门限对码根进行判定;通过遍历域内所有的本原多项式,找出完整的连续码根分布,进而完成生成多项式的识别.仿真结果表明,所提方法的计算量较其他方法明显减少,并能有效完成码长和生成多项式的识别,在误码率小于10-3时,对常用RS码的识别率能达到90％以上.

Blind parameter recognition of RS codes

As the existing methods for Reed-Solomon (RS) codes recognition are very complicated,a new method is proposed.Firstly,code weight distribution under different block length is counted.By defining similarity coefficient to the theoretical code weight,the most similar one is found,which corresponds to the correct code length.Then binary hypothesis test is established with a decision threshold to find code roots.By traversing all the primitive polynomials in the field,the complete distribution of continuous code roots is found,with which generator polynomial is calculated.Simulation results show that the computation cost of this method is significantly less than that of other methods,and it can effectively complete the recognition of code length and generator polynomial which has a recognition probability of 90％ for the commonly used RS codes when the bit error rate is less than 10-3.