Abstract: Identification and classification technology plays an important part in study of the BCI system.Presents an algorithm for classifying single-trial electroencephalogram（EEG）.It combines common spatial subspace decomposition with support vector machine to extract features from multichannel EEG.This algorithm is applied to the data set IV of “BCI Competition 2003” with a classification accuracy of 89% on the test set.

Key words: electroencephalogram, Brain-Computer Interface（BCI）, Support Vector Machine（SVM）, Common Spatial Subspace Decomposition（CSSD）

摘要： 在脑-机接口的研究中分类识别技术占有重要地位。介绍了一种方法，用于对单次信号的分类。这种方法主要思想是将共空域子空间分解和支持向量机相结合，以便提取信号特征。该方法被用于“BCI Competition 2003”第IV数据包上，分类正确率达89%。

关键词: 脑电信号, 脑－机接口, 支持向量机, 共空域子空间分解