Fractal analysis of sEMG signal under varying load conditions

Abstract

This paper reflects the studies undertaken to understand the complexity of the Electromyography (EMG) signal during biceps muscle contraction (flexion), by estimation of Fractal Dimension using Hurst's Rescaled Range (R/S) analysis. The measurements were done using surface electrodes with necessary signal acquisition circuitry and signal data is collected using sound card of PC through Matlab for data analysis purpose. The signal is also subjected to Frequency Domain techniques to estimate the power generated by muscle under different loading conditions. The Fractal Dimension, as obtained here, is found to vary under different loading conditions. The study of non-linear behavior of the EMG signal under varying loading conditions can lead to significant findings, pertaining to time based monitoring of muscle rehabilitation and endurance in a quantitative way.