Research Of The Adaptive Blind Source Separation Algorithm Based On Negentropy

Blind signal separation (BSS) is a new powerful technique in modern signal processing. Nowadays, it has been wildly applied in different fields, such as speech signal processing, image processing, multi-user communication, array processing and biomedical signal proeessing.The main method to solve the problem of BSS is independent component analysis (ICA).The purpose of ICA is to reach a separate matrix which allows the separated signals to be statistically independent each other. This article studies the basic theory of independent component analysis and introduces the pretreatment, separation criteria and optimization algorithms of independent component analysis.The nongaussian-maximization criterion is one of independence metrology criterion which blind source separation usually uses. The negentropy is a robust and optimal measure of non-Gaussian. This article focuses on the negentropy criterion and study deeply the essence of negentropy criterion for blind source separation. Any super-Gaussian distribution and sub-Gaussian distribution are generated by the generalized Gaussian distribution. The simulation results confirm the extreme characteristics of negentropy when the source signals possess different distribution.Taking the negentropy criterion as the cost function, one robust auto-adapted algorithm is obtained. And a new adaptive step-size algorithm based on the distance between separating matrix and the best separating matrix is proposed. Computer simulations confirm the theoretical analysis and show the algorithm’s superiority on convergence speed and steady-state error.