Performance of channel selection used for multi-class EEG signal classification of motor imagery

Abstract

The brain computer interface (BCI) is a system which involves communicating and controlling the machine with the help of brain signal (l’électroencéphalographie EEG), can be used to help people with physical disabilities regain their motor ability. In this paper we investigate the classification of mental tasks based on EEG data for Brain Computer Interfaces, classification of 4 imaginary motor activities (left hand, right hand, foot, tongue) whith the BCI competition III data set IIIa. Performance comparisons will be made between different Linear Discriminant Analysis (LDA), Support Vector Machine (SVM), K-Nearest Neighbors (KNN) algorithms of classification using time_frequency characteristics. This article also shows the influence of choice, number and position of electrodes for each subject (channel selection) were investigated to provide an improvement for the classification accuracy of the algorithm. Results show that using one subset of the channels with positions varied from subject to subject; gave good classification results by comparing it with other research results an average accuracy of 86.06% was observed among all 3 subjects.