Paper Details: Downloads: 512
Serial Number: P1120826001
Title: Blind Source Separation: A comparison between Temporal Predictability and Adaptive Self-Normalized Radial Basis Function (ASN-RBF) Neural Network Models
Authors: D.Malathi and N.Gunasekaran
Abstract: In this paper, an Adaptive Self-Normalized Radial Basis Function (ASN-RBF) neural network is proposed for separating the individual source signals from an artificially mixed signal. This method is based on a generative model, with a Radial Basis Function (RBF) Function neural network to model the nonlinearity from the latent variables to the observations. The Gaussian function is used to approximate the inverse of the nonlinear mixing matrix which is assumed to exist and able to be approximated. The network is trained by the gradient descent optimization algorithm with fixed centers to update the parameters in the generative model. This proposed method is well-suited for nonlinear data analysis problems and theoretically interesting. Minimum three signals are considered for simulation. Simulation results show the feasibility of the proposed method. The performance of the proposed network is compared with the Jone’s BSS algorithm in terms of scaling parameter and it is illustrated with computer simulated experiments.
Keywords: Blind source separation, Adaptive self-normalized radial basis function neural network, Gradient descent optimization algorithm
Journal/Conference: International Journal of Artificial Intelligence and Machine Learning
Volume:
Issue:
Submission Date: 6/24/2008 12:00:00 AM
Review Date: 7/25/2008 12:00:00 AM
Publishing Date: 8/29/2008 12:00:00 AM
Article Downloads: 512
Download:

Facebook