摘要
脑电信号(EEG)是研究脑活动的一种重要的信息来源,基于脑电信号的人与计算机的通信已成为一种新的人机接口方式。在此主要通过时域回归方法对BCIⅡ竞赛数据进行EEG信号去噪预处理,运用6阶AR参数提取脑电特征作为神经网络的输入,最后用Matlab 7.0进行仿真,得到分类正确率为90%。实验表明,该方法可以达到很好的分类效果。
Electroencephalogram（EEG）signal is an important information source of investigating the brain action.The communication based on EEG between human brain and computer becomes a new modality of human-computer interaction.Through time-domain regression method for EEG denoising pretreatment of BCI Ⅱ match data,AR model coefficient is extracted as feature vector,and the mental tasks based on BP network is classified.The simulation was performed by means of Matlab 7.0.And the classification correctness of 90% was achieved.The experiments show that this method can get a good result of classification.
出处
《现代电子技术》
2010年第10期27-29,共3页
Modern Electronics Technique
基金
陕西省教育厅专项科研计划项目(09JK433)
渭南师范学院研究生基金资助项目(10YKZ009)
