Recoverability Analysis Of Blind Source Separation Based On Sparse Representation

Blind signal separation is a new technique of signal processing aimed at retrieving a system' s unknown information from its output only, which can be used in array processing and data analysis. There are many potential application, especially in wireless communication, medical signal processing, image processing and radar signal processing. In recent years, blind signal separation has been an attractive trend in the area of academiao The main contributions of this thesis are as follows:1. This paper presents modified K-means BSS algorithm , and the simulation shows that the algorithm not only keeps down the characteristic in separating the blind sources as conventional K-means BSS algorithm but also makes up the lack of conventional K-means BSS algorithm.2. This paper discusses recoverability of underdetermined blind source separation based on a two-stage sparse representation approach, within the stochastic framework. Blind source separation based on sparse representation usually supposes that source signals are sparse sufficiently. This paper estimates recoverability probability when source signals are not sparse sufficiently, which not only reflects the relationship between the recoverability and sparseness of sources but also indicates efficacy of the two-stage sparse representation approach for solving BSS.