Development of new brain computer interface based on EEG and EMG

Abstract

It is widely accepted in the BCI (brain computer interface) research community that neurological phenomena are the only source of control in BCI system. However, asynchronous control is an important issue for BCIs working in real-life settings, where it should determine from brain signals not only the desired commands but also when the user wants to input it or stop it. This paper proposes a new mouse control system using a combined classification method from EEG (electroencephalogram) and EMG signals to solve the problem. There should be defined at least four commands, up, down, left, and right, for the mouse control. However, stop command is added in our system for the asynchronous control, especially to off the commands. These commands are derived from a combined classification method from source of control by using the mono-polar of 6 channel EEG signals and EMG signal with masseter muscle. EEG signals are used for the distinctions from left and right or up and down depending on EMG signal where it is used as a toggle operation for which turns on or off the commands and reverses the commands. From the experimental simulation result, we get at least 88% accuracy.