Performance Comparison of EMG Signal Analysis for Manual Lifting using Spectrogram

Abstract

Electromyography (EMG) signal is non-stationary signal and highly complex time and frequency characteristics. Fast-Fourier transform common technique in signal processing involving EMG signal. However, this technique has a limitation to provide the time-frequency information for EMG signals. This paper presents the analysis of EMG signal of the variable lifting height and mass of load between the four subjects selected in manual lifting by using spectrogram. Spectrogram is one of the time-frequency representation (TFR) that represents the threedimensional of the signal with respect to time and frequency in magnitude presentations. The manual lifting tasks is based on manual lifting of 5 kg and 10 kg load that performed by the right biceps brachii at lifting height of 75 cm and 140 cm. Four from ten healthy volunteers in fresh condition is selected into this comparison of subject performance tasks with their raw data collections. The raw data of EMG signals were then analyzed using MATLAB 2011 to obtain the voltage in time and frequency information. This study obtained the mean instantaneous RMS Voltage (Vrms(t)) to visualize the strength of the subjects produced during the manual lifting tasks. Results of this study evince the physical details of the subjects would able to effect the performance of the lifting. Higher lifting height and the number of contraction, the better performance of the subjects. It concluded that the application of spectrogram is able to providing the performance of the subjects by time-frequency information for EMG signals.