An ensemble learning algorithm for blind signal separation problem

Li, Yan and Wen, Peng (2005) An ensemble learning algorithm for blind signal separation problem. In: 2005 International Conference on Computational Intelligence for Modelling, Control and Automation (CIMCA 2005), 28- 30 Nov 2005, Vienna, Austria .

[img]
Preview
PDF (Published version)
Li_Wen_CIMCA2005_PV.pdf

Download (424Kb)

Abstract

The framework in Bayesian learning algorithms is based on the assumptions that the quantities of interest are governed by probability distributions, and that optimal decisions can be made by reasoning about these probabilities together with the data. In this paper, a Bayesian ensemble learning approach based on enhanced least square backpropagation (LSB) neural network training algorithm is proposed for blind signal separation problem. The method uses a three layer neural network with an enhanced LSB training algorithm to model the unknown blind mixing system. Ensemble learning is applied to estimate the parametric approximation of the posterior probability density function (pdf). The Kullback- Leibler information divergence is used as the cost function in the paper. The experimental results on both artificial data and real recordings demonstrate that the proposed algorithm can separate blind signals very well


Statistics for USQ ePrint 8164
Statistics for this ePrint Item
Item Type: Conference or Workshop Item (Commonwealth Reporting Category E) (Paper)
Refereed: Yes
Item Status: Live Archive
Additional Information: doi: 10.1109/CIMCA.2005.1631425. Published version deposited in accordance with the copyright policy of the publisher.
Depositing User: Ms Leslie Blay
Faculty / Department / School: Historic - Faculty of Engineering and Surveying - Department of Electrical, Electronic and Computer Engineering
Date Deposited: 21 May 2010 09:30
Last Modified: 02 Jul 2013 23:55
Uncontrolled Keywords: backpropagation; belief networks; blind source separation; least squares approximations; neural nets; statistical distributions; Bayesian ensemble learning; Kullback-Leibler information divergence; LSB neural network training algorithm; blind signal separation problem; enhanced least square backpropagation; parametric approximation; posterior probability density function; probability distribution
Fields of Research (FOR2008): 09 Engineering > 0906 Electrical and Electronic Engineering > 090609 Signal Processing
08 Information and Computing Sciences > 0802 Computation Theory and Mathematics > 080201 Analysis of Algorithms and Complexity
08 Information and Computing Sciences > 0801 Artificial Intelligence and Image Processing > 080108 Neural, Evolutionary and Fuzzy Computation
Socio-Economic Objective (SEO2008): E Expanding Knowledge > 97 Expanding Knowledge > 970109 Expanding Knowledge in Engineering
URI: http://eprints.usq.edu.au/id/eprint/8164

Actions (login required)

View Item Archive Repository Staff Only