基于小波分解与支持向量机的P300识别算法_杨立才.pdfVIP

26 6 Vol.26 No.6 2007 12 Chinese Journal of Biomedical Engineering December 2007 P300 1* 1 1 2 杨立才 李金亮 姚玉翠 李光林 1( , 250061) 2(, , 60608) : P300 , P300 , Span ; P300 BCI Competition 2003P300 ,,, ,5 100% : -; P300; ; P300 Detection Algorithm Based on Wavelet Decomposition and Support Vector Machines 1* 1 1 2 YANGL-i Cai LIJin-Liang YAO Yu-Cui LI Guang-Lin 1(Dep artment of Biomedical Engineering, School of Control Science Engineering , Shandong University, Ji. nan 250061) 2 (The University of Illinois at Chicago, Chicago Illinois 60608, USA) Abstract: In order to overcomethe shortcomings of support vector machines in terms of low training and detection speed for P300detection, a new algorithm based onwavelet decomposition andsupport vector machineswas proposed inthis paper. Using wavelet decomposition and span estimation method, we implemented the feature extraction of EEG signals and the fast choiceof the optimal parameters of support vector machines; and realized the detection of P300 component with support vector machines, which have a good classification performance. The algorithm was testedwith aP300 dataset from the BCI competition 2003. The results showed that the algorithm designed in this paperwas superior to thetraditional support vector machines algorithms intermsofthetraining and detection speed. Usingthis algorithm, we achieved an accuracy of 100% in P300 detectionwithin fiverepetitions. Key words:brain-computer interf