A new approach to detect P300 in a single trial based on PCA and SVM classifier

Single trial detection of P300 signal is one of the trending areas of Brain Computer Interface (BCI) research. We propose a new method with a high level of accuracy to detect P300 signals in a single trial. Features were obtained with a new technique making use of the wavelet coefficients. Reduced feature dimension was achieved using Principal Component Analysis (PCA). Support Vector Machine (SVM) was used as the classifier. The proposed method has achieved an accuracy of 98.47% for Subject A and 95.06% for Subject B. Thus a high degree of accuracy was obtained.