Spatial filtering optimisation in motor imagery EEG-based BCI

Common spatial pattern (CSP) is becoming a standard way to combine linearly multi-channel EEG data in order to increase discrimination between two motor imagery tasks. We demonstrate in this article that the use of robust estimates allow improving the quality of CSP decomposition and CSP-based BCI. Furthermore, an evolutionary algorithm (EA)-type for electrode subset selection is proposed. It is shown that CSP with the obtained subset electrode yield comparable results with the ones obtained with CSP over large multi-channel recordings.