The Research On Key Technologies Of Brain Computer Interface Based On Blocking SSVEP And P300 Hybrid Paradigm

The brain-computer interface is a device that converts an electroencephalogram generated in a human brain into a digital signal by an electroencephalogram acquisition device and is analyzed by a computer and output through an output device(such as a computer monitor and a speaker).The brain-computer interface is a special way of communicating with external devices only through the brain,not through the human muscles and peripheral nerves.Communication through the brain-computer interface does not depend on human body language and normal language,thus opening up a way for people with disabilities who cannot speak and become inactive to communicate with others,such as people with amyotrophic lateral sclerosis.Communication with the others may reawaken their enthusiasm for life.The brain-computer interface is a device that converts an electroencephalogram generated in a human brain into a digital signal by an electroencephalogram acquisition device and is analyzed by a computer and output through an output device(such as a computer monitor and a speaker).In order to achieve the purpose of communicating with others.Therefore,the performance of the brain-computer interface determines the fluency of communication with the outside world.Nowadays,the research of brain-computer interface is mostly single paradigm brain-computer interface.Although the single paradigm brain-computer interface has the advantages of simple design,it may not be as accurate as the hybrid paradigm brain-computer interface in recognition,so the research of hybrid paradigm brain-computer interface has become an important research direction of today's brain-computer interface.In this paper,the data of the traditional Oddball paradigm and the single SSVEP paradigm are collected for comparison with the data in the mixed paradigm.Because the SSVEP signal and the P300 signal pass through the same visual path before being generated in the brain,in order to improve the speed of the paradigm,this paper decided to use the SSVEP blocking signal.The character of this paper is designed as a matrix of 5×8.Under the single SSVEP,the character string is firstly flashed from left to right according to the frequencies of 8 Hz,9 Hz,10 Hz,11 Hz,12 Hz,13 Hz,14 Hz and 15 Hz,then the character line is from top to bottom.It flashes at a frequency of 8 Hz,9 Hz,10 Hz,11 Hz,and 12 Hz in sequence.The mixing paradigm is that the character string is flashed at the frequencies of 8 Hz,9 Hz,10 Hz,11 Hz,12 Hz,13 Hz,14 Hz,and 15 Hz from left to right,at the same time the character line randomly appears in a flicker-free state,generating the SSVEP-B signal and then The P300 component is triggered in the brain.Use SSVEP and P300 to determine the column and row where the character is located.In the process of acquisition,the EEG signal will inevitably collect some interference signals,including power frequency interference and eye electricity.In this paper,a VMD-based method for removing artifacts is designed.The experimental results show that the method can effectively remove the EOG signals that were accidentally acquired during EEG signal acquisition.The feature classification of EEG signals for removing eye electricity artifacts is carried out.The SSVEP signal is selected by CCA method.The P300 signal is identified by discrete wavelet decomposition and multi-feature fusion of time-domain energy entropy,and use the support vector machine to classify the features.Through the traditional P300 paradigm,single SSVEP paradigm and mixed paradigm experimental data identification accuracy and information transmission rate comparison analysis,P300 component and SSVEP component in the hybrid paradigm compared with the traditional P300 paradigm and single SSVEP paradigm,recognition accuracy and information transmission rate There are obvious improvements.