Power Spectral Density Estimation of EMG Signals Using Parametric and Non-Parametric

Approach

Akankshya Shradhanjali, Subhra Chowdhury and Neelesh Kumar*

Biomedical Instrumentation Unit, CSIO -CSIR, Sector 30-C, Chandigarh, India 160030.

*Corresponding Author’s E-mail: neel5278@gmail.com

Abstract 05 April 2013

Abstract

EMG is classified on the basis of RMS amplitude values and frequency analysis by using non-parametric and parametric approach. The variation of EMG spectra is used to compare the estimation methods in terms of frequency resolution as well as in determining the spectral components. Spectral analysis of EMG signals are done on four major lower limb muscles during isometric contraction using three methods i.e. Welch, Yule walker and Burg. Yule walker showed superior performance among the three methods as it performs well for long continuous signals.

Keywords: Spectral Estimation, Power Spectral Density, Parametric Approach, Non- Parametric Approach, Frequency Resolution, RMS value, Amplitude.