Blind separation of mixed kurtosis signals using local exponential nonlinearities

Muhammad Tufail, Masahide Abe, Masayuki Kawamata

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper we propose exponential type nonlinearities in order to blindly separate instantaneous mixtures of signals with symmetric probability distributions. These nonlinear functions are applied only in a certain range around zero in order to ensure the stability of the separating algorithm. The proposed truncated nonlinearities neutralize the effect of outliers while the higher order terms inherently present in the exponential function result in fast convergence especially for signals with bounded support. By varying the truncation threshold, signals with both sub-Gaussian and super-Gaussian probability distributions can be separated. Furthermore, when the sources consist of signals with mixed kurtosis signs we propose to estimate the characteristic function online in order to classify the signals as sub-Gaussian or super-Gaussian and consequently choose an adequate value of the truncation threshold. Finally, some computer simulations are presented to demonstrate the superior performance of the proposed idea.

Original languageEnglish
Title of host publication2005 IEEE International 48th Midwest Symposium on Circuits and Systems, MWSCAS 2005
Pages39-42
Number of pages4
DOIs
Publication statusPublished - 2005 Dec 1
Event2005 IEEE International 48th Midwest Symposium on Circuits and Systems, MWSCAS 2005 - Cincinnati, OH, United States
Duration: 2005 Aug 72005 Aug 10

Publication series

NameMidwest Symposium on Circuits and Systems
Volume2005
ISSN (Print)1548-3746

Other

Other2005 IEEE International 48th Midwest Symposium on Circuits and Systems, MWSCAS 2005
CountryUnited States
CityCincinnati, OH
Period05/8/705/8/10

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Blind separation of mixed kurtosis signals using local exponential nonlinearities'. Together they form a unique fingerprint.

Cite this