Analysis Of EEG Signal Based On Motor Imagery And Design Of Brain-Computer Interface System

Brain-computer interface technology has achieved the direct control of human brain to external devices.From the original intention of solving the life ability of severely disabled people to many fields such as entertainment,transportation and medical treatment,the application of brain-computer interface technology has become more and more extensive.Studies have shown that the correct interpretation of EEG signals is a key step in accurate communication between the brain and external devices.Therefore,it is necessary to identify EEG signals by relevant algorithms.This paper analyzes the spontaneous EEG signals based on motor imagery,and realizes the recognition and classification of two kinds of signals.And a set of intelligent car system based on brain-computer interface is developed which realizes the control of the intelligent car by EEG signals.The main contents are:EEG signals are filtered through an elliptic filter to retain the effective signals within the range of 8~30HZ that can reflect the changes of brain consciousness and effectively reduce the interference of high and low frequencies.By comparing the advantages and disadvantages of commonly used algorithms for extracting EEG signal features,this paper combines the common spatial pattern method and AR model method to obtain new feature values,and classifies them by support vector machine classifier and linear discriminant classifier.The experimental results show that the classification accuracy of the two classifiers is better than that of the single feature extraction method.In order to further prove the feasibility of feature extraction method,the EEG signals of two subjects in the third session of brain-computer interface competition are classified.The classification results show that the improved combined feature extraction method in this paper can significantly improve the classification accuracy of EEG signals.A brain-computer interface intelligent car system based on TGAM module is designed.The hardware part takes the Arduino Nano development board as the core,and adopts the modular design to build the hardware circuit such as signal acquisition module,signaltransmission module,motor drive control module and power supply module.Meanwhile,the software design of the intelligent car system is realized in the Arduino IDE compilation environment.After many experiments,it is proved that the intelligent car system can realize the real-time control of the car by EEG signals.