盲分离问题的可分性理论(英文)  

Separability Theory for Blind Signal Separation

作　　者：章晋龙[1] 谢胜利[1] 何昭水[1] 

机构地区：[1]华南理工大学无线电与自动控制研究所,广州510640

出　　处：《自动化学报》2004年第3期337-344,共8页Acta Automatica Sinica

基　　金：Supported by National Natural Science Foundation of P. R. China(60274006);Trans-Centuries Training Program Foundation for the Talents by the Ministry of Education,National Natural Science Foundation of P. R. China for Excellent Youth(60325310);the Natu

摘　　要：对盲信号分离的基本理论问题进行了探讨,给出了从混叠信号中分离出源信号的充要条件,并回答了源信号能提取的数量问题.同时,还给出了基于该理论的盲分离算法,该方法适应于病态混叠情形下的提取.且就几种不同的情况进行了计算机仿真.仿真结果表明了所提理论的正确性及相应算法的有效性.The basic theory of blind signal separation is discussed. Not only the sufficiency-neces-sity condition is given for the extraction of single source but also the problem how many signals can be extracted is answered. Moreover, the blind separation algorithm based on this theory is proposed. The method can also be applied to blind extraction in the case of morbid mixture. At last, simulation is done in different mixed conditions, and the results show that the separation theory is correct and the algorithm is valid.

关 键 词：盲信号分离 基本理论 病态混叠 四阶累计量 罚函数 

分 类 号：TN911[电子电信—通信与信息系统]