An EEG Signal Analysis Method Based on Common Spatial Pattern and Support Vector Machine Algorithm

To solve the problem of EEG signal on feature extraction and low recognition rate, we analyze the characteristics of EEG signal and propose an EEG signal analysis method based on common spatial pattern (CSP) and support vector machine (SVM) algorithms. In this method, we first preprocess the EEG signal by wavelet packet transform (WPT) and extract the specific frequency band signal in the EEG signal. Then the features of the EEG signal are extracted by constructing pair-wise CSP (PW-CSP) filters. Finally, the recognition of the EEG signal was realized by the support vector machine algorithm which optimizes parameter selection by particle swarm optimization (PSO), and the 2008 BCI competition 2A data set was used to verify the classification effect of the algorithm. The experimental results indicated that the classification accuracy of the method may reach up to 93.07%, and the average accuracy is higher than the results of the top two in the 2008 brain-computer interface (BCI) competition. The EEG signal features extracted by this method can reflect the signal characteristics of EEG well, and improve the accuracy of classification, which may assist in the development and application of BCI.