Classification of motor imagery EEG signals based on wavelet transform and sample entropy

Modeling and learning of brain activity patterns represent a huge challenge to the brain-computer interface(BCI) based on electroencephalography(EEG). Many existing methods based on EEG signal frequency domain feature extraction can not achieve high classification accuracy rate requirements[1]. This paper presents a method of feature extraction of EEG based on wavelet transform and sample entropy. The dynamic changes of the sample entropy of the left and right hand motion imaginary EEG signal and its neurophysiological significance were analyzed. Finally, Fisher's linear discriminant was used to classify left-right hand motion imaginary EEG. Experiments shows that the average correct rate of classification was 88.9%.