Research On Portable Brain Control System Integrating Target Frequency Recognition And Focus State Detection

Brain-computer interface(BCI)based on Steady-state visual evoked potentials(SSVEP)have the advantages of high information transmission rate,short training time and high output commands.This makes it one of the most frequently used EEG patterns.However,in SSVEP-based asynchronous BCI systems,a key problem is how to distinguish the control state from the idle state.In addition,in the BCI system based on SSVEP,the visual stimulator is mostly LCD,which limits the activity range of the subjects and reduces the portability of the system.In view of the above problems,this paper proposes a fusion Attention detection and frequency recognition method based on weighted Dempster-Shafer theory(ADFR-DS)method to study the SSVEP-based asynchronous BCI system with fusion attention detection.On this basis,build a portable brain control system based on decision fusion,and evaluate the performance of the portable brain control system based on decision fusion.By selecting LED lights as visual stimulators,to improve the portability of BCI systems based on SSVEPs and promote their development in education,military,medical and other fields.In order to combine attention detection and frequency recognition,an ADFR-DS method is proposed to process Electroencephalogram(EEG)data from different channels simultaneously: Individualized frequency band based optimized complex network(IFBOCN)algorithm processes EEG signals from the prefrontal region to detect attention.While Ensemble task-related component analysis(e TRCA)algorithm processes EEG signals from the occipital region for frequency recognition,the ADFR-DS method fuses the classification results of the two algorithms at the decision level to generate the final output of the BCI system.Offline experimental results show that ADFR-DS is superior to e TRCA algorithm in terms of true positive rate,true negative rate,accuracy rate(ACC)and information transmission rate(ITR).Specifically,ADFR-DS increases the average ACC of e TRCA from 62.71% to 69.3%.And increased the average ITR from 184.28 bits/min to 216bits/min(data length 0.3s).In terms of experimental design,a portable asynchronous brain control system based on decision fusion is built.LED lights were used instead of LCD as visual stimulator to improve the portability of the BCI system.In order to evaluate the performance of the portable brain control system based on decision fusion,an off-line experiment on the effect of different stimulus distances on decision fusion algorithm and an asynchronous online brain control system experiment of intelligent vehicle are designed to evaluate the performance of ADFR-DS algorithm.The off-line experimental results show that the performance of ADFR-DS algorithm is higher than that of Filter bank canonical correlation analysis(FBCCA)algorithm under different stimulus distances,and the performance stability of ADFR-DS algorithm is better than that of FBCCA algorithm when the stimulus distance increases to 2 or 3 meters.Specifically,when the stimulator is 2 meters away,the average performance of ADFR-DS is 10% higher than that of FBCCA algorithm,and the performance of ADFR-DS algorithm decreases by 5% compared with 1 meter,while the performance of FBCCA algorithm decreases by 10%±1%.Online experiment results show that the average number of false triggers of ADFR-DS is reduced by 1.2 times and the average number of collisions is reduced by 1.6 times compared with eTRCA.