A stable ICA algorithm based on exponent density and Gaussian parametric density mixture models

Independent Component Analysis (ICA) is an effective method to solve the problem of Blind Source Separation (BSS). In this paper, a new algorithm is proposed to separate signals mixtured by sub-Gaussian, super-Gaussian, symmetric and asymmetric sources. Alternative score functions in the algorithm are derived by using exponent density model and Gaussian parametric density mixture model. The score functions are self-adaptive through estimating the high-order moments of original signals. Moreover, a stability condition for the proposed algorithm is given to guarantee separating the true solution. Simulations are presented to illustrate the performance and effectiveness of the proposed algorithm.