Title :
An approach for nonlinear blind source separation of signals with noise using neural networks and higher-order cumulants
Author :
Zhang, Nuo ; Zhang, Xiaowei ; Lu, Jianming ; Yahagi, Takashi
Author_Institution :
Graduate Sch. of Sci. & Technol., Chiba Univ., Japan
Abstract :
We propose a robust approach for blind source separation when observations are contaminated with Gaussian noise and nonlinear distortion. A radial basis function network (RBFN) is employed to estimate the inverse of the nonlinear mixing matrix. We utilize an novel cost function which consists of mutual information and higher-order cumulants of signals. Compared with moments, higher-order cumulants can provide a clearer form and more information of signals. Thus, the proposed method has not only the capacity of recovering the nonlinearly mixed signals, but also removing high-level Gaussian noise from transmitted signals. Through simulation and analysis of artificially synthesized signals, we illustrate the efficacy of this approach.
Keywords :
Gaussian noise; blind source separation; higher order statistics; matrix inversion; nonlinear distortion; parameter estimation; radial basis function networks; signal denoising; Gaussian noise; cost function; higher-order cumulants; mutual information; neural networks; nonlinear blind source separation; nonlinear distortion; nonlinear mixing matrix inversion; radial basis function network; synthesized signals; Analytical models; Blind source separation; Cost function; Gaussian noise; Mutual information; Neural networks; Noise robustness; Nonlinear distortion; Radial basis function networks; Signal analysis;
Conference_Titel :
Circuits and Systems, 2005. ISCAS 2005. IEEE International Symposium on
Print_ISBN :
0-7803-8834-8
DOI :
10.1109/ISCAS.2005.1465938