Title :
BSS algorithm by diffusing nonparameteric density estimator
Author :
Peng Li ; Li, Peng
Author_Institution :
Dept. of Math., North China Univ. of Water Resources & Electr. Power, Zhengzhou, China
Abstract :
Nonparametric diffusion mixing estimator (DME) based blind signal separation (BSS) algorithm is proposed under the framework of natural gradient optimization method. In order to improve the performance of signal separation by BSS, the probability distributions of source signals must be described as accurately as possible. In this paper, we use the new data-driven bandwidth selection method based MDE to estimate the probability distributions of sources, which can improve the performance of fixed-width kernel density estimator (FKDE). The MDE is inspired via a Langevin diffusion process. As a result, the proposed algorithm has a wider application and do not need to assume the parametric nonlinear functions as them. The effectiveness of the proposed algorithm has been confirmed by simulation experiments.
Keywords :
bandwidth allocation; blind source separation; gradient methods; nonparametric statistics; optimisation; parameter estimation; statistical distributions; BSS algorithm; Langevin diffusion process; MDE; blind signal separation; data-driven bandwidth selection method; fixed-width kernel density estimator; gradient optimization method; nonparametric diffusion mixing estimator; probability distributions; source signals; Adaptation model; Artificial neural networks; Kernel; Symmetric matrices; Variable speed drives; Blind Source Separation(BSS); Diffusion Mixing estimator(DME); Independent Component Analysis(ICA); fixed-width kernel density estimator(FKDE);
Conference_Titel :
Computer Science and Information Technology (ICCSIT), 2010 3rd IEEE International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-5537-9
DOI :
10.1109/ICCSIT.2010.5563624
