可穿戴式干电极脑机接口系统设计

An embedded wearable dry electrode brain-computer interface system is proposed to solve the problem of low portability of existing electroencephalogram (EEG) devices. The system first collects EEG signals through dry electrodes with a 24-bit analog-to-digital conversion chip and then uses an FIR digital filtering to perform 3 to 35 Hz bandpass filtering. At last, an embedded processor is used for EEG recognition. In terms of the recognition algorithm, EEG signals are truncated to eliminate the fixed delay of visual stimulation and the group delay caused by the FIR filtering; then the Pearson correlation coefficient method is used for online recognition, and the effects of stimulation duration on the correct rate and information transmission rate are analyzed. Experimental results show that the average signal-to-noise ratio of the collected signals in this system is 74.86 dB, and the common mode rejection ratio at 50 Hz is -132.57 dB. The average recognition time of the correlation coefficient method is 0.13 s, and the average accuracy of the online four-target steady-state visual evoked potential experiments is 69.54%. Compared with the portable BCI system using the standard CCA algorithm, the average recognition time is shortened by 0.27 s, and the average accuracy increases by 10%. These results show that the proposed system provides new ideas for applications of the dry electrodes BCI system.