Comparison of LDA and RBF-NN in EEG features classification for motor imagery [articol]

Abstract

This paper presents an approach that uses self-organizing fuzzy neural network based time series prediction to extract the EEG features in time domain. EEG signals from two electrodes placed on the scalp over the motor cortex are predicted by a single fuzzy neural network. Features derived from the mean squared error of the predictions and from the mean squared of the predicted signals are extracted from EEG data within a sliding window using two auto-organizing fuzzy neural networks with multi inputs and a single output. The features are classified by linear discriminant analysis and radial-basis function neural network.