Research On The Noisy Blind Signal Separation Based On Sparse Component Analysis

Blind Source Separation (BSS), also called blind signal separation, is to recover the unknown independent sources from several observed mixed signals according to the signal's statistical characteristic without any prior knowledge of the sources and channels. Generally, to simplify the research work, BSS methods mostly concentrate on the over-complete case or complete case, that is, the number of observed signals is not less than that of source signals. However, in the real communications, we always meet the under-determined case, that is, the number of observed signals is less than that of source signals. So it is of great significance to resolve problem of signal separation in the under-determined case.Independent component analysis is a new blind separation technique of BSS. However, most of these ICA algorithms ignore the under-determined case. Sparse representation of signals has received a great deal of attention in recent years. For the signal in time domain is usually not sparse, so an effective sparse representation of the signal is necessary. Sparse component analysis is a method of signal processing based on sparse representation. SCA estimate the sparse signals from observations which is different from ICA. Sparse representation has been widely applied in underdetermined BSS.Currently, most of speech separation algorithms consider the noise-free case. But in the real communications, signals can inevitably be interfered by the background noise. So it is of great value and significance to resolve problem of signal separation in the noisy environment. Presently some scholars are devoting themselves to the research of noisy BSS algorithms, but the achievement is little.In this dissertation, we analyze and summarize the previous work of BSS. A research on blind separation in under-determined case and in noisy environment was made, and then present several valid resolving methods mainly based on the linear program:1. Work of summarize and implement to the two-stage approach of under-determined BSS was done.2. In the second stage of two-stage method the source signals are estimated by combined with SCA and linear program, which was proposed to handle the observed noisy signals separation. Computer simulation results exhibit good separating performance of the proposed approach.