In the movement imaginary experiment, the characteristic of EEG (Electroencephalograph) mainly concerned on the Mu(μ) rhythm and Beta(P) rhythm(8–25Hz), and we can distinguish the movement imaginary of the left-right hand by the characteristic of the signal of this band. Adopt the wireless EEG augmentation equipment with 8–32 channel which can wear, by using FIR band-pass filter and independent component analysis(ICA) to filter and denoise the original date; then off-line analyze the date which were filtered and denoised by the method of commom spatial pattern(CSP), extract the feature of the signal; Finally, the characteristics is classified by support vector machine (SVM) optimized by genetic algorithm. The results of the extraction process on Matlab showed that the average recognition rate about the movement imagery of left-right hand is above 95%. The experiment proved that the above mentioned method is feasible.