基于小波包能量的脑电信号特征提取方法  

EEG feature extraction method based on wavelet packet energy

作　　者：徐宝国[1] 宋爱国[1] 王爱民[1] 

机构地区：[1]东南大学仪器科学与工程学院,南京210096

出　　处：《东南大学学报（自然科学版）》2010年第6期1203-1206,共4页Journal of Southeast University：Natural Science Edition

基　　金：国家高技术研究发展计划(863计划)资助项目(2009AA01Z311;2008AA04Z0202);国家自然科学基金资助项目(60775057)

摘　　要：在脑机接口研究中,针对运动想象脑电信号的特征抽取,提出了一种基于小波包变换和AR模型的特征提取方法.该方法首先利用小波包变换对大脑C3和C4处采集的2路运动想象脑电信号进行3层分解,抽取小波系数的能量特征;然后,利用Burg算法提取脑电信号的5阶AR模型系数;最后,将这2类特征组合,使用基于马氏距离的线性判别分类器对左右手运动想象脑电模式进行分类,正确率达到91.43%.该方法提取的特征向量较好地反应了运动想象脑电信号的事件相关去同步和事件相关同步的变化时程,为BCI研究中脑电信号的模式识别提供了新的思路.此外,该方法的识别率高,复杂性低,适合应用于在线脑机接口中.In the study of brain-computer interface（BCI）,a novel method of extracting motor imagery electroencephalography（EEG） features based on the wavelet packet transform（WPT） and the autoregressive（AR） model is proposed.First,the EEG signals sampled from the C3 and C4 positions of the brain are decomposed to three levels by the WPT,and the features of the wavelet coefficient energy are computed.Then,the fifth-order AR coefficients of the EEG signals are estimated by the Burg＇s algorithm.Finally,by combining the two kinds of features,the combination features are used as the input vectors for linear discriminant analysis（LDA） classifier based on the mahalanobis distance to classify the pattern of the left and the right hand motor imagery EEG signals,and the recognition rate is 91.43%.The experimental results show that the eigenvector extracted by the proposed method can effectively reflect the event-related desynchronization（ERD） and the event-related synchronization（ERS） time course changes of the motor imagery EEG.This method provides a new idea for the EEG pattern recognition in BCI research.In addition,this method has a high recognition rate and low complexity.It is suitable for the application in online BCI systems.

关 键 词：脑机接口 运动想象 小波包变换 

分 类 号：R318[医药卫生—生物医学工程]