线性盲信道辨识算法中存在的问题及解决方法

The problem of linear blind channel identification algorithms and the resolvent

传统的盲信道辨识与均衡技术大多采用高阶统计的方法 ,这种方法有一些弊端。后来开始研究基于二阶统计的方法 ,这是一个较大的突破。基于二阶统计的盲信道辨识算法大致有以下五种 :线性预测算法LPA[1] [2 ] ,外积分解算法OP DA[3] [4 ] ,多步线性预测算法MSLP[5] ,最小均方平滑算法LSS[6 ] 和约束最小输出能量算法CMOE[7] 。但是这些算法的仿真结果显示信道是不能被辨识的 ,该文将通过分析这些算法 ,指出导致辨识失败的原因 ;并在数字通信系统的背景下 ,给出解决方法。

Traditionally, the blind channel identification/equalization techniques based on higher order statistics, are known to suffer from many drawbacks. Later, some methods using the second order statistics, such as linear prediction algorithm (LPA) [1][2] , outer product decomposition algorithm (OPDA) [3][4] , multi step linear prediction algorithm (MSLP) [5] , least square smoothing algorithm (LSS) [6] , and constrained minimum output energy algorithm (CMOE) [7] , are developed to improve blind channel identification. However, the simulate results show that these methods could not identify the blind channel because there still needs other condition, which couldn′t be got for knowing nothing about the channel in the whole blind condition. In this paper, the reasons of these methods in the failure of channel identification are analyzed and the resolvent in the digital communication system are provided.