Research On Blind Source Separation In Communication System

During the past decade, blind source separation (BSS), also called independent component analysis (ICA) has become an active branch of signal processing. It can recover independent source signals from their observed mixtures without knowing the distribution type of the source signals and the mixing coefficients.In this paper, after an introduction to the development history of BSS, a simple mathematic description was given, including the mathematical model of BSS, the assumptions made about BSS problems and the mathematical theory and methods commonly used in BSS. Then, some algorithms and applications of BSS were investigated deeply, and many more efficient methods and further improvements for the existed methods were given. The main works in this dissertation can be introduced as follows: Main problems of the research of Independent Component Analysis (ICA) are discussed. Several classical cost functions of BSS algorithms based on ICA and their derivations are introduced. This paper unifies them in the information-theoretic framework.Several algorithms aiming at complex signal are summarized, and their performances are compared using simulation. The simulation of separating mixed signal with same frequency is actualized with Cross-Cumulant Zero-Forcing algorithms. For signals in time variant mixing model, the problem of separating simple real time mixing signals with single sample is solved by the generalized EASI algorithm, and the problem of separating real time complex mixing signal is solved by batch program algorithm in the form of sliding window. The noisy BSS is primarily studied. With the noise-free algorithms and the adaptive bias removal method, a modified noisy EASI algorithms is proposed.