Application of New Approaches for the Feature Extraction and Classification of EEG Signal Processing in Brain Research

This paper describes some new approaches for the feature extraction of SSVEP signal in EEG signal processing. One of them is Canonical Correlation Analysis (CCA) and another one is CWT along with ANN. Basically CCA is applied to analyze the frequency components of SSVEP in EEG. The essence of this method is to extract narrowband frequency components of SSVEP in EEG. The CWT offers a valuable tool for the analysis of signals as it provides precise location in terms of time of high frequency component. The selections of the mother wavelet having high correlation with the signal provide a more accurate timefrequency analysis. ANNs are considered to be good classifier due to their inherent features as robustness, adaptive learning, and generalization ability and self-organization capability.