Research on steady state visual evoked potential based on FBCCA

Steady state visual evoked potential (SSVEP) is widely used in brain computer interface systems because of its steady signal, low number of channels and high recognition rate. In this paper, SSVEP was elicited by sinusoidal stimulation, and Filter Bank canonical correlation analysis (FBCCA) was used to solve the problem that the amplitude and energy of the nonlinear and non-stationary signals at each frequency band varied with time. The experimental results show that the correct recognition rate is 97.5%. And the mean ITR are 152bits/min. The method can effectively improve the identification and control of SSVEP EEG signals. At the same time, the stimulus paradigm displayed on the computer screen reduced the visual fatigue of the subjects and improved the comfort level of the subjects.