1.Measurement Technology and Instrumentation Key Lab of Hebei Province, Yanshan University,Qinhuangdao, Hebei 066004, China
2.School of Civil Engineering, Hebei University of Technology, Tianjin 300401, China
3.Audio-Visual Machinery Research Institute, Qinhuangdao,Hebei 066000, China
Abstract：P300 is widely used in the field of brain cognition and BCI, however, its intensity is so weak that it is easy to suffer from interferences of environment and artifacts of blink, ECG, and EMG and normally submerged in EEG. In order to separate P300 from interferences speedily and efficiently, the feature of P300 in time, frequency and spatial domain are analyzed. A method is proposed based on BSS to extract P300 feature, combining coherence average, wavelet transformation and BSS. A new technique is described aiming at selecting ICs corresponding to P300 after BSS to multi-lead EEG. Three BSS algorithms, Informax, FastICA and AMUSE are compared in the performance of P300 feature extraction. Experimental results verify that the proposed method has an obvious improvement in P300 feature extraction compared with the other methods using temporal and frequency characteristics of P300 only.
［1］张作生, 王峰.参数模型法提取单次诱发脑电［J］.中国生物医学工程报,1991,(4):228-233.
［2］罗志增,李亚飞,等.脑电信号的混沌分析和小波包变换特征提取算法［J］.仪器仪表学报,2011,32(001):33-39.
［3］韦保林,罗晓曙,等.EEG信号的径向基函数神经网络预测［J］.中国生物医学工程学报,2004,22(6):488-492.
［4］Jie X, Cao R, Li L. Emotion recognition based on the sample entropy of EEG［J］.Biomed Mater Eng, 2014, 24（1）: 1185-1192.
［5］黄日辉. 思维脑电及P300脑电的特征提取与识别［D］. 江门:五邑大学, 2008.
2015, Vol. 36 
