Browse by author
Lookup NU author(s): Pei Gao, Dr Wai Lok Woo, Emeritus Professor Satnam Dlay
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
Independent component analysis (ICA) is to recover a set of statistically independent components from the output of the mixing system. However, most algorithms could achieve excellent performance only under linear assumption, which may not accord with practical situations. In this paper, a mathematical model of the nonlinear mixing system stemmed from the theory of functional analysis is investigated, achieving the formulation of a stable unique inverse solution. In addition, the optimum framework for the nonlinear demixer can be obtained directly from the derived mixing model. Furthermore, a novel method to acquire the desired solution is proposed using the Weierstrass approach, which displays high flexibility of nonlinearity. Finally, simulation results are presented to verify the proposed scheme. (6 References).
Author(s): Gao P, Woo WL, Dlay SS
Publication type: Conference Proceedings (inc. Abstract)
Publication status: Published
Conference Name: Communication Systems, Networks and Digital Signal Processing (CSNDSP)
Year of Conference: 2004
Pages: 588-591
Notes: Dlay SS Newcastle upon Tyne, UK. Communication Systems, Networks and Digital Signal Processing. CSNDSP 2004. Fourth International Symposium. Newcastle upon Tyne, UK. 20-22 July 2004.