7th IEEE International Conference on Computer and Information Technology (CIT 2007)
Development of P300 Detection Algorithm for Brain Computer Interface in Single Trial
Aizu-Wakamatsu City, Fukushima, Japan
October 16-October 19
ISBN: 0-7695-2983-6
This paper proposed a combinational method to estimate and discriminate the P300 component. The Kalman filter and the wavelet transformation were used to estimate the P300 component and PCA and three discrimination meth- ods; linear discrimination analysis, Mahalanobis distance, and Bayes' theorem, were used to discriminate the esti- mated P300 from single trial. The results show the effec- tiveness of the combinational technique in discrimination. The combination of wavelet transformation and a linear discrimination analysis can perform discrimination in the highest accuracy and the fastest processing speed. KEYWORD: ERP, P300, Kalman filter, Wavelet transfor- mation, Linear discrimination analysis, Mahalanobis dis- tance, Bayes' theorem
Citation:
Hiroyuki Ishita, Motoki Sakai, Ji?nei Watanabe, Wenxi Chen, Daming Wei, "Development of P300 Detection Algorithm for Brain Computer Interface in Single Trial," cit, pp.1100-1105, 7th IEEE International Conference on Computer and Information Technology (CIT 2007), 2007