SSVEP Signal Detection for BCI Application

Abstract

Steady State Visually Evoked Potential (SSVEP) is one of the most popularly used signals in Brain Computer Interface (BCI) applications. A new method to detect SSVEP signals of three different frequencies (6Hz, 8Hz and 15Hz) has been proposed. This method uses Fast Walsh Hadamard Transform (FWHT) for feature extraction and Naive Bayes Classifier (NBC) for feature classification. The algorithms used in the proposed method FWHT and NBC consumes vey less memory and also makes the method less computationally complex. The proposed method also uses less execution time making it suitable for real time BCI application.