EEG pattern recognition based on dual-tree complex wavelet transform and particle swarm optimization

Aiming at the issue of motor imagery electroencephalography (EEG) pattern recognition in the research of brain-computer interface (BCI), a novel method based on dual-tree complex wavelet transform (DTCWT) and particle swarm optimization (PSO) was proposed. The advantage of DTCWT over discrete wavelet transform (DWT) was discussed in depth and the ERD/ERS phenomenon was verified at first. Then, the signal component related to sensory motor rhythms was extracted based on dual tree complex wavelet decomposition and reconstruction. Afterward, PSO algorithm was implemented to search the optimal time interval automatically for feature extraction. Finally, average energy, root mean square and signal variance were extracted as features and linear discriminant analysis (LDA) was applied for classification. The results show that the proposed method can find the relatively optimal time interval for feature extraction automatically and the maximum classification accuracy is 90%, which is better than the BCI competition winner.