Performance And Application Of Brain Computer Interface Based On Steady State Visual Evoked Potentials

With the rapid development of computer science,EEG research has entered a new fast-developing stage.So far the study of EEG is mainly based on brain-computer Interface(Brain Computer Interface,BCI)technology which establishes a connection for transmitting information between human brain and peripheral so as to control the external devices directly.Among BCI systems,the SSVEP system based on the steady-state Visual Evoked Potential(Steady State Visual Evoked Potential,SSVEP)is the most widely used.Based on the basis of previous studies,this thesis uses the BCI technology to extract the electroencephalogram(electroencephalogram,EEG),analyse the characteristic frequency of SSVEP,and realize the real-time control of mobile phone music play system.First of all,collecting SSVEP signals which is stimulated by black and white squares through Neuroscan system.SSVEP is the rhythmicity of EEG signals induced by continuous visual stimulation.Nowdays,many methods can be used to study SSVEP,this thesis take advantages Canonical Correlation Analysis(Canonical Correlation Analysis)to extract performance indicators of SSVEP response frequency,analyze the effects of channel number,data length,area of the brain and other properties of pattern on the accuracy of characteristic frequency,and draw the quantitative conclusions to lay foundation for the control of music player system.Second,according to the SSVEP of black and white squares,chessboard stimulation,transverse stripe stimulation and vertical stripe stimulation collected respectively.This thesis make use of the Improved 2 Dimension Empirical Mode Decomposition(2D-EEMD)algorithm to preanalyse the SSVEP signals and compare the differences between the four stimulations by Fourier.Under simulation of a single square,SSVEP response maintains macro consistent in the whole area of the brain while other simulations,SSVEP response is obviously regional that main response of frequencies is different in different areas(frontal lobe,temporal lobe,parietal lobe,occipital lobe).Finally,the mobile phone music player system as an external device is designed.The extraction of SSVEP's characteristic frequency of black and white stimulus based on CCA is taken as instruction transmitting to the mobile client,controling music player App(App named ssvepControl)to complete the "play/pause","keep","the before song",and "the next song" operations real-time.The BCI system can achieve real-time operation of 5 seconds,and the accuracy rate more than 95%.