SSP'05 -- Proceedings

	IEEE/SP 13th workshop on Statistical Signal Processing July, 17-20, 2005 - Bordeaux - France

Information regarding the paper

Title

Negentropy Analysis of Surface Electromyogram Signal

Author(s)

Kianoush Nazarpour	Department of Electrical Engineering, Tarbiat Modarres University
Ahmad R. Sharafat	Department of Electrical Engineering, Tarbiat Modarres University
Seyyed M. P. Firoozabadi	Department of Medical Physics, Tarbiat Modarres University

Get the paper in PDF format

To obtain Acrobat Reader (version 5 minimum required) necessary to his read.

Abstract

This study deals with measuring the non-Gaussianity in Surface Electromyogram signal (sEMG). The signal was obtained from biceps brachii muscle during elbow flexion at four different levels of Maximum Voluntary Contraction (MVC). Typically the sEMG generated from constant-force, constant angle, non-fatiguing contractions is modelled as a stochastic process, and its probability density function (pdf) is assumed to be Gaussian. Results of utilizing negentropy for characterizing the non-Gaussianity of sEMG signal indicate that its pdf is clearly non-Gaussian during light contractions (below 30% of MVC) and it tends to a Gaussian process at higher force levels. The results validate the application of Higher Order Statistics (HOS) based methods in sEMG signal processing at low levels of MVC