Natural gradient-based recursive least-squares algorithm for adaptive blind source separation  

作　　者：ZHUXiaolong ZHANGXianda YEJimin 

机构地区：[1]KeyLaboratoryforRadarSignalProcessing,XidianUniversity,Xi'an710071,China [2]DepartmentofAutomation,StateKeyLaboratoryofIntelligentTechnologyandSystems,TsinghuaUniversity,Beijing100084,China

出　　处：《Science in China(Series F)》2004年第1期55-65,共11页中国科学（F辑英文版）

摘　　要：This paper focuses on the problem of adaptive blind source separation (BSS). First, a recursive least-squares (RLS) whitening algorithm is proposed. By combining it with a natural gradient-based RLS algorithm for nonlinear principle component analysis (PCA), and using reasonable approximations, a novel RLS algorithm which can achieve BSS without additional pre-whitening of the observed mixtures is obtained. Analyses of the equilibrium points show that both of the RLS whitening algorithm and the natural gradient-based RLS algorithm for BSS have the desired convergence properties. It is also proved that the combined new RLS algorithm for BSS is equivariant and has the property of keeping the separating matrix from becoming singular. Finally, the effectiveness of the proposed algorithm is verified by extensive simulation results.This paper focuses on the problem of adaptive blind source separation (BSS). First, a recursive least-squares (RLS) whitening algorithm is proposed. By combining it with a natural gradient-based RLS algorithm for nonlinear principle component analysis (PCA), and using reasonable approximations, a novel RLS algorithm which can achieve BSS without additional pre-whitening of the observed mixtures is obtained. Analyses of the equilibrium points show that both of the RLS whitening algorithm and the natural gradient-based RLS algorithm for BSS have the desired convergence properties. It is also proved that the combined new RLS algorithm for BSS is equivariant and has the property of keeping the separating matrix from becoming singular. Finally, the effectiveness of the proposed algorithm is verified by extensive simulation results.

关 键 词：blind source separation natural gradient recursive least-squares pre-whitening. 

分 类 号：TN957.52[电子电信—信号与信息处理]