基于智能单粒子与信号变化度的盲源分离算法  

Blind source separation algorithm based on intelligent single particle and signal variability

作　　者：陈雷[1] 张立毅[1,2] 郭艳菊[3] 李锵[2] 刘婷[1,2] 

机构地区：[1]天津商业大学信息工程学院,天津300134 [2]天津大学电子信息工程学院,天津300072 [3]河北工业大学信息工程学院,天津300401

出　　处：《电路与系统学报》2012年第4期89-94,共6页Journal of Circuits and Systems

基　　金：国家自然科学基金资助项目(60802049);天津市高校科技发展基金资助项目(20080710)

摘　　要：提出了一种新的基于智能单粒子优化的有序盲源分离算法。采用信号变化度作为分离的目标函数,利用球坐标变换方法对分离向量进行变换,使用智能单粒子优化算法对目标函数进行求解,通过去相关方法从混合信号中去除已分离出的源信号成分,最终实现按照信号变化度降序分离出各源信号。仿真结果表明,本算法能够有效实现对源信号的有序分离,且分离精度很高。A novel sequential blind source separation algorithm based on intelligent single particle optimization is proposed. The signal variability is used as the objective function and the separation vector is transformed using the spherical coordinates transform method. The intelligent single particle optimization algorithm is used for solving the objective function. The source signal component separated is wiped off using decorrelation method and source signal could be separated out respectively according to the decreasing order of the signal variability. Simulation results show that the separation algorithm proposed can realize sequential separation of source signal efficiently and the separation precision is very high.

关 键 词：盲源分离 信号变化度 智能单粒子 有序分离 

分 类 号：TN911.7[电子电信—通信与信息系统]