Classification of ElectroCorticoGraphy Signals Reduced by Wavelet Transform

Abstract

Studies to solve the mystery of how the human brain works is receiving considerable attention in recent years. Analysis of the signals produced in brain is also within the scope. In this study, classification of ECoG (Electrocorticography) signals which produced in brain is performed. The data used in this study were obtained from data set no 1 which had been used on BCI Competition III. The first part, to decrease the processing load, the number of channels are reduced by eliminating channels (electrodes) which have low separation success. Than it was obtained Wavelet coefficients by Discrete Wavelet Transform (DWT) and determined classification features from Wavelet Coefficents. These features are tested by KNN (K Nearest Neighbors), SVM (Support Vector Machine) and LDA (Linear Discriminate Analysis) classification methods. It's obtained that 94% success in classification by using KNN.