Research On Blind Source Separation Based On Particle Swarm Optimization And Fish Swarm Algorithm

Blind source separation is an focal issue that separates or extracts the source signals only using the mixed signals in the case of no any prior knowledges. It is widely applied to audio signal processing, communication signal processing, noise decreasing, medical signal processing, earthquake signal processing and image signal processing etc.The major contribution of this paper is summarized as follow:(1) In this paper, the basic model and theory of blind source separation are systematically introduced, the main algorithms of blind source separation are systematically summarized, and their virtues and defects are analyzed. How to select the cost functions and how to optimize them in the independent component analysis are mainly analyzed and studied.(2 ) The particle swarm optimization and improved particle swarm...