PROJECT TITLE :

An Adaptive Motion-Onset VEP-Based Brain-Computer Interface

ABSTRACT:

Motion-onset visual evoked potential (mVEP) has been recently proposed for EEG-based brain-computer interface (BCI) system. It's a scalp potential of visual motion response, and typically composed of three components: P1, N2, and P2. Typically several repetitions are needed to increase the signal-to-noise ratio (SNR) of mVEP, however additional repetitions will price more time therefore lower the efficiency. Considering the fluctuation of subject's state across time, the adaptive repetitions primarily based on the topic's real-time signal quality is important for increasing the Communication potency of mVEP-based mostly BCI. During this paper, the amplitudes of the three parts of mVEP are proposed to create a dynamic stopping criteria in keeping with the practical information transfer rate (PITR) from the coaching knowledge. Throughout on-line check, the repeated stimulus stopped once the predefined threshold was exceeded by the $64000-time signals and then another circle of stimulus newly began. Analysis tests showed that the proposed dynamic stopping strategy could considerably improve the communication potency of mVEP-primarily based BCI that the common PITR increases from 14.five bit/min of the ancient fixed repetition technique to 20.8 bit/min. The improvement has great price in real-life BCI applications because the communication efficiency is terribly necessary.