Feature extraction and classification of EEG for imagery movement based on mu/beta rhythms

Classification of electroencephalogram(EEG) is a crucial issue for EEG-based brain computer interface(BCI) system. The paper presents a method for EEG classification, where property of event-related desynchronization/synchronization(ERD/ERS) of mu/beta rhythms, The mu/beta rhythms are obtained after filtering and wavelet packet transform. The energy feature is formed by the squared amplitude of the preprocessed data, and then be classified by the function “classifiy” attached by matlab7.0.This is an extension of our previous work on the use of ERD/ERS of mu/beta rhythms for EEG classification. Numerical experiments with imagery movement data set in 2003 BCI competition, confirm the useful behavior of the property for EEG classification, and well verify the property in turn.