Brain-computer interface design based on slow cortical potentials using matlab/simulink

Abstract

A brain-computer interface (BCI) is a communication system that translates brain-activity into commands for a computer or other electronic devices. In other words, a BCI allows users to act on their environment by using only brain-activity, without using peripheral nerves and muscles. The feature selection and classification methods of EEG signals are two key points of the BCI system. In this paper, we designed a simulation model based on slow cortical potentials (SCPs) using Matlab/Simulink. The simulation model mainly had three parts: preprocessing, feature selection, and classification. We obtained significant improvement on classification accuracy for data set Ia of BCI competition 2003, which was a typical BCI data. The classification accuracy rates were 70.90% (190/268) and 89.76% (263/293) for train and test data sets respectively. The result also showed that training could improve the classification accuracy since the test data were taken later in the recording sessions than the training data.