Comparative analysis of signal processing in brain computer interface

Brain computer interface (BCI) systems utilise Electroencephalography (EEG) to translate specific human thinking activities into control commands. An essential part of any BCI is a pattern recognition system. In this paper, a number of different features and classifiers are compared in terms of classification accuracy and computation time. Two typical features are studied: autoregressive (AR) and spectrum components along with three different classifiers; the K-nearest neighbor, linear discriminant analysis (LDA) and Bayesian statistical classifiers. The results showed that all classifiers achieved very high accuracies and short computation times.