Optimization of a set of wavelets for classification of imaginary movement-related cortical potentials

Abstract

We have proposed an algorithm for optimization of a set of wavelets associated to multi-channel recordings. The method has been applied to the problem of classifying MRCPs in the context of BCIs. The preliminary results show that optimizing the wavelets may substantially improve the classification performance on a test set. These results must be confirmed in future studies on larger subject and signal samples.