Study On Recognition Method Of Brain-computer Interface Based On SSVEP Frequency And Blocking Features

Steady State Visual Evoked Potential means that when the human body is subjected to a fixed frequency flicker or changing pattern of visual stimulation,the potential activity of the cerebral cortex is modulated to produce a continuous response(at the fundamental or octave of the stimulus frequency)that is related to the stimulation frequency.Brain computer interfaces(BCIs)based on steady state visual evoked potentials have become a popular direction in BCI field because of its less training time and high information transmission rate.However,due to the limitations of human physiological factors and hardware technology,the frequency range of visual stimulation that can induce high-quality SSVEP signals is limited,which also limits the expansion of the number of control commands of traditional frequency-encoded SSVEP-BCI.To some extent,these problems hinder the further development of this type of BCI system.In response to the above problems,this paper studies the new BCI paradigm based on SSVEP frequency response and blocking response.This paper focus on the following aspects:1.Research on locking response law of SSVEP.This article is the first to study the blocking response characteristics of SSVEP under autonomous eyes closed and explore the blocking response law of SSVEP under different stimuli and the optimal Parameters of stimulus source.In the study,firstly,time-frequency analysis was used to obtain the time-frequency diagram of EEG signals and the results of time-frequency analysis verified that subjects can effectively block SSVEP response by autonomously closing their eyes.In addition,the research found that the blocking response phenomenon under different visual stimuli is different,so the SSVEP blocking response law under different stimuli was further studied.This paper designs a blocking experiment to collect EEG signals under visual stimuli of different frequencies,colors,brightness,and sizes.Frequency band energy and blocking rate were used as evaluation indicators to conduct a further analysis of the factors affecting the blocking response of SSVEP.Finally,we determined the best stimulation parameters to induce SSVEP blocking response.2.Research on frequency and blocking recognition algorithm.This article first introduces the frequency identification method based on filter bank canonical correlation analysis(FBCCA).On the basis of frequency identification,it is proposed to use the correlation coefficients after FBCCA subband fusion as the blocking recognition features and then two methods for blocking recognition were proposed.The first is blocking recognition and blocking based on the correlation coefficient threshold duration calculation method.The second is the blocking recognition based on multi-class SVM and the blocking duration calculation method based on classification status.3.A new BCI paradigm based on SSVEP frequency response and blocking response is proposed.The key issue of this paradigm is the effective identification of blocking and the accurate calculation of blocking duration.Therefore,based on the optimal stimulus parameters,this article designed a blocking experiment with multiple switching between open and closed eyes under two stimulus frequencies.The stimulation frequencies selected in the experiment were 8 Hz and 15 Hz form the low and middle frequency bands and both blocking duration were 6s.Through the blocking experiment,the experimental data in different states were obtained.This verifies the feasibility and practicability of the new paradigm proposed in this paper and lays the foundation for the future online BCI system based on SSVEP frequency and time-frequency joint coding of blocking response.