Statistical Modeling of EMG Signal

Naufel B. Mohammed, Atta A. Saeed, Avan M. Ahmed

The aim of this research is to measure and build a statistical model of EMG signals . A linear regression method was applied as a statistical modeling method ,the common types of linear regression models was explored. The electromyography (EMG) was measured from the two hands of a person as a way to perform noise reduction with the use of XOR logical operation facilities . To measure EMG signals, the research used six OLIMEX  EMG shield , controlled  by Arduino 328 control board , a new classification and modeling of EMG signals of 5 movements of an arm are presented. One of the six channels  used for the EMG  measurements was used as a  reference channel , while the remaining as a measuring channels . The resultant  EMG of each channel was considered as an independent variable (emg)  for the linear regression model and the movement angle (degree) as a dependent variable for the model.