关键词: 脑电信号, 独立分量分析, 自回归模型, BP神经网络

Abstract: 【Abstract】 Electroencephalogram, or EEG, signals are an important source of information for study of underlying brain processes. The communication based on EEG between human brain and computer is a new modality of human-computer interaction. In this article, EEG signals of different mental tasks are preprocessed using Independent Component Anlysis(ICA), AR model is used to extract the feature, and Back-Propagation(BP)network as the classifier. The result prove that method can improve the accuracy of classifying different mental tasks and get better discrimination accuracy.

Key words: EEG, Independent Component Anlysis (ICA), AR model, BP network