一种新型的卷积混和盲信号分离算法  

A New Algorithm For Blind Source Separation of Convolutive Mixtures

作　　者：柳艳红[1] 魏学业[1] 王大虎[1] 

机构地区：[1]北京交通大学电子信息工程学院,北京100044

出　　处：《信号处理》2007年第6期801-805,共5页Journal of Signal Processing

摘　　要：本文提出了一种新的卷积混合盲信号分离算法。首先将卷积混合模型进行等价简化,再用理论推导论证了卷积混合盲分离问题等价于最优化Wiener滤波器问题,给出了一种去相关分离准则函数。最后,对两个源信号卷积混合的情况,用标准自适应LMS算法使分离准则函数达到最小值,从而得到了两个信号混合的分离算法;然后,推广到多个源信号混合得到了相应分离算法。通过计算机仿真试验验证了本算法的有效性和可行性。A new algorithm for blind source separation of convolutive mixtures is proposed. The equivalent model of the convolutive mixtures is obtained in order to make analysis and procession of the problem well-suited. Through theoretical deduction, the separation of convolutive mixtures is proved to be equivalent to optimum Wiener filtering problem and a new decorrelation criterion for separation of convolutive mixtures is proposed. We propose an algorithm for the separation of the two convolutive mixtures by the means of the standard LMS algorithm and then another algorithm for multi-source convolutive mixtures. The numerical computer simulations are presented ,to illustrate the validity and feasibility of our algorithm.

关 键 词：盲信号分离 卷积混和 最优滤波器 去相关 LMS算法 

分 类 号：TN911.7[电子电信—通信与信息系统] TP391.4[电子电信—信息与通信工程]