Multi-phase cycle coding for SSVEP based brain-computer interfaces

BACKGROUND: Brain-computer interfaces (BCIs) based on Steady State Visual Evoked Potential (SSVEP) have attracted more and more attentions for their short time response and high information transfer rate (ITR). The use of a high stimulation frequency (from 30 Hz to 40 Hz) is more comfortable for users and can avoid the amplitude-frequency problem, but the number of available phases for stimulation source is limited. To circumvent this deficiency, a novel protocol named Multi-Phase Cycle Coding (MPCC) for SSVEP-based BCIs was proposed in the present study. METHODS: In MPCC, each target is coded by a block word that includes a series of cyclic codewords, and each block word is corresponding to a certain flickering visual stimulus, which is a combination of multiple phases from an available phase set and flickers at single frequency. The methods of generating block code and extracting phase were presented and experiments were performed to investigate the feasibility of MPCC. RESULTS: The optimal stimulation frequency was subject-specific, and the optimal duration was longer than 0.5 s. The BCI system with MPCC could achieve average discrimination accuracy 93.51 ± 5.62% and information transfer rate 33.77 ± 8.67%. CONCLUSIONS: The MPCC has the error correction ability, can effectively increase the encoded targets and improve the performance of the system. Therefore, the MPCC is promising for practical BCIs.