A stable ICA algorithm based on exponent density and Gaussian parametric density mixture models
Kefeng Wang, Xu Xu, Chonghui Guo · 2011
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.