Research On Blind Sources Separation Algorithm In Non-stationary Environment

Blind source separation(BSS)is a process of recovering the source signals only by observed mixed signals but otherwise unobserved source signals from their mixtures without any prior knowledge of the channel.BBS has become the basic and important issue in signal processing and communication areas and takes a key role in the field of speech signal processing,image processing,wireless communication,and so on.This thesis studies around this hot topic.And the study focused on the algorithm of blind source separation in non-stationary environments.This thesis introduces the researching background and domestic and international development of blind source separation,and summaries the three basic models of blind source separation,discusses the solvability and the uncertainty of blind source separation,describes the basic knowledge of blind source separation algorithm,including signal processing,objective function,optimize algorithm and performance index.It also introduces the concept and classification of non-stationary environments in blind source separation.Aiming at some problems of blind source separation algorithm of non-stationary environments,especially researching for the mixed matrix on time-varying or mutation,this thesis presents two kinds of improved algorithm for blind source separation.One is to overcome the disadvantages of EASI algorithm,the thesis presents a adaptive variable step symbol EASI algorithm based on reference separation system,The new algorithm combines the EASI algorithm with the sign operation which is derived from a generalized dynamic separation model,and the variable step-size which is based on reference separation system.The sign operator improves the performance of the algorithm and lead to much faster convergence speed,and the variable step-size balances the contradiction between convergence speed and the error in steady state.The other one is to overcome the shortcoming of non-holonomic natural gradient algorithm,the thesis presents a adaptive constraint variable step algorithm based on non-holonomic natural gradient algorithm.At the same time,an adaptive means is adopted to the constrain factor.It could do the adjustment adaptively according to the separation situation,and the variable step-size which is based on the gradient of cost function is also applied to the proposed algorithm.It balances the contradiction between convergence speed and the error in steady state.Computer simulation results show that the approved algorithms performance is superior to the usual algorithm.