Research On Brain-computer Interface Method Based On Steady-state Visual Evoked Potentials

Brain computer interface(BCI)technology is an emerging human-computer interaction technology that does not rely on external muscles and peripheral nervous system,opening an additional pathway for information exchange and control between human brain and machine.This technology provides not only the possibility for physically challengef patients to achieve hardware control over the outside world,but also a completely new way of information exchange for the ordinary person.Recently,BCI based on Steady-State Visual Evoked Potential(SSVEP)has received a lot of attention from scholars because of its high signal-to-noise ratio and high information transfer rate.As the application scenario of SSVEP-BCI continues are expanding,the study of effective SSVEP identification algorithm becomes more and more important,which is also the main content of this study.In this paper,a stimulation platform capable of evoking SSVEP signals is firstly built,and the SSVEP signals are evoked from the subjects using a sinusoidally encoded stimulus interface.Then,the collected data are used to carry out the study of SSVEP recognition algorithm and control platform construction.The following research work was conducted in this paper.(1)A real-time online SSVEP-BCI system is constructed using the SSVEP stimulation platform,which enables real-time target stimulation recognition and sends the recognition results to the receiving end via TCP/IP protocol.Currently,the functions that can be realized are the walking control of NAO robot and the recognition output of characters.(2)An individual signal mixture template multivariate synchronization index(IST-MSI)algorithm is proposed,which incorporated individual training template and individual harmonic sensitivity coefficient into the standard MSI algorithm.Specifically,the proposed method enlarged the frequency-domain power spectrum of the fundamental frequency and its harmonics to reduce the redundant information in the individual training template.The synchronization index values at non-target frequency identified by MSI algorithm are significantly reduced through unequal ratio scaling of harmonic sensitivity coefficient and mixing coefficient,thereby improving the SSVEP recognition.The experimental results showed that under the signal length of 1.2 seconds,the average classification accuracy of IST-MSI algorithm reached 84.3% in six target frequencies,which was 5.8% higher than that of standard MSI algorithm.(3)Filter bank-driven multivariate synchronization index(Filter Bank MSI,FBMSI)is proposed.The algorithm is designed to decompose the EEG signal into 9 subbands by using 9 bandpass filters with different frequency bands.The synchronization index of each sub-band signal and the reference signal are calculated separately,and the fundamental frequency component and harmonic component are combined through the parameter search of the weight vector,so as to improve the recognition accuracy of SSVEP.The offline data set of nine subjects was used for parameter seeking optimization,and online experiments were conducted for validation.The online experimental results show that the FBMSI algorithm achieves an average classification accuracy of 83.56% with six instructions for a signal length of 1 second,which is 12.26% better than the standard MSI algorithm.