Comparative Analysis to Identify Efficient Method for SSVEP Based BCI

The new developments of medicine and information technology can help severely paralyzed people to restore the link with their environment through the Brain Computer Interface (BCI). In fact, a BCI can replace the movements of human organs by actions of a machine controlled by thoughts. Several studies on BCIs show that systems based on the steady state visual evoked potential (SSVEP) signal allow to reach the best performances. In the present work, we aim to evaluate and compare signal processing methods used in the conception of SSVEP based BCIs. Three methods from the literature were examined which are the Minimum Energy Combination (MEC), the Canonical Correlation Analysis (CCA) and the Multivariate synchronization index (MSI). Results show that the most efficient method is CCA.