Journal Homepage: www.ijcst.org

Abstract – Brain–Computer Interfaces (BCIs) provide a non-muscular channel to communicate with the outside world by means of brain activity. A crucial step for efficient BCI operation is brain signal processing methods. Most BCI systems for humans use scalp recorded electroencephalographic activity, whereas Electrocorticography (ECoG) is a minimally-invasive alternative to electroencephalogram (EEG), providing higher and superior signal characteristics allowing rapid user training and faster communication. Its efficiency is based on brain signal processing methods that classify brain signal patterns in different tasks accurately. Artifacts in raw brain signal make it necessary to pre-process signals for feature extraction. This paper presents a BCI system pre-processing and extracting features from ECoG signals through the use of Symlet Wavelets. Signal classification is done using Support Vector Machine (SVM) with Radial Basis Function (RBF). Index Terms – Brain–computer interfaces (BCIs)