脑机接口系统中EEG 信号特征提取与分类Feature Ex

脑机接口(BCI)技术是近年来国际上的研究热点之一，它通常利用脑电(EEG)来实
现无动作的人机交互，运动想象是其中一种重要BCI 实验范式，利用第二届国际脑机接口竞赛中的一组实验数据为处理对象，将数据经公共空间模式滤波、小波时频分解、然后采用T 加权提取最后特征，并利用支持向量机进行分类器设计。实验结果表明，该算法效果较好，最终识别正确率达到89.3％。
关键词：运动想象(MI)；公共空间模式(CSP)；支持向量机(SVM)
Abstract: Brain-computer interface (BCI) is one kind of techniques on which more and more interest has been put by the international researchers recently. By using electroencephalogram, BCI establishes the interfaces between human and computer，in which motor imagery is an  important experimental paradigm. In this paper, the dataset from BCI Competition II is analyzed using the algorithm designed through common spatial pattern filter, continuous wavelet transform decomposes , then to use the T-weighted extraction final characteristic, and using support vector machines to design the classifier. By means of the above approaches, better classification results are obtained. The result on testing set revealed an accuracy of 89.3% for classification.
Key words: brain-computer interface (BCI); electroencephalogram (EEG); motor imagery (MI);
common spatial pattern (CSP); support vector machines (SVM)