Phase detection of multi-channel SSVEPs via complex sparse spatial weighting

Abstract

A brain-computer interface (BCI) based on steady-state visual evoked potentials (SSVEP) is one of the most practical BCI, because of high recognition accuracies and short time training. Phase of SSVEPs can be potentially applicable for generating device commands. However, the effective method of estimating the phase of SSVEPs has not yet been established, especially, in the case of using multi-channel electroencephalogram (EEG). In this paper, we propose a novel method for estimating the phase of SSVEPs from multi-channel EEG, which uses complex sparse spatial weighting. We conducted experiments with the phase-coded SSVEPs based BCI for evaluating performance of our proposed method. As a result, our proposed method showed higher recognition accuracies than conventional methods in all six subjects.