Feature Selection using Relative Wavelet Energy for Brain-Computer Interface Design

Abstract

The critical issues in brain-computer interface (BCI) research is how to translate a person's intention into brain signals for controlling computer program or wheelchair. In this paper, we used a new method: relative wavelet energy (RWE) for feature selection in BCIs design and linear discriminant analysis (LDA) and support vector machine (SVM) were utilized to classify the pattern of left and right hand movement imagery. Its performance was evaluated by mutual information (MI) using the data set Mb of BCI Competition III. This technology provides another useful way to EEG feature selection in BCIs research.