Optimized Design And Practice Of P300-based Brain-Computer Interface

As a new method of human-computer interaction,the brain—computer interface(BCI)draw much attention these years.Due to its high accuracy and lack of training,ERP-based BCI has been the hotspot among researchers.Moreover,the P300 speller is one of the highest classification accuracy and bits rate experimental model.However,due to the intensification of adjacent stimuli and double-flash problem,the further improvement of the P300 speller was limited.Recent years,the researches on the improvement of accuracy mainly focused on two aspects,that is paradigm and algorithm.However,few studies have combined the two kinds of researches.In this paper,firstly,a P300 speller was achieved by using EEG acquisition equipment and Matlab.Secondly,the research findings of both paradigm and algorithm was combined,to explore the possibility of further improving the classification accuracy.As a result,it has been found that applying the face paradigm and the combined classifier method at the same time can further improve the accuracy of the speller.It is found that the combined classifier method can also improve the classification accuracy of the face paradigm,which proves the possibility of combining various research results.Finally,by analyzing the superimposed waveforms of event-related potentials under the two paradigms,it is concluded from the perspective of cognitive science that the neighboring stimuli interfere with participants' attentiveness and that the adjacent stimuli at the lower and right sides interfere with the subjects more greatly.This provides a theoretical basis for further improving the classification accuracy of the speller.