Fast and Accurate Classifier-Based Brain-Computer Interface System Using Single Channel EEG Data

Abstract

In brain-computer interface system (BCIs), direct communication between humans and computers is performed by analyzing neural signals and transforming them into digital signals. In the continuation of our previous works, in this paper, we advanced the proposed BCI system that was based on the gaze on rotating vanes. The speed of communication and convenience of the user are very important factors in BCI systems. Therefore, in this paper, for the convenience of the user, a single EEG channel was used. Also, for to increase the speed of the system, we tried to the classification of 0.5 sec epochs with a partial least squares regression (PLSR) as a fast classifier. In addition, we computed the information transfer rate (ITR) that has proved our proposed BCI system is fast and accurate. This system could be used in real-time implementations due to having high classification rate, speed and convenience of the user.