Filter Bank-Driven Multivariate Synchronization Index for Training-Free SSVEP BCI

In recent years, multivariate synchronization index (MSI) algorithm, as a novel frequency detection method, has attracted increasing attentions in the study of brain-computer interfaces (BCIs) based on steady state visual evoked potential (SSVEP). However, MSI algorithm is hard to fully exploit SSVEP-related harmonic components electroencephalogram (EEG), which limits application BCI systems. this paper, we propose filter bank-driven (FBMSI) overcome limitation and further improve accuracy SSVEP recognition. We evaluate efficacy FBMSI method by developing 6-command SSVEP-NAO robot system with extensive experimental analyses. An offline first performed EEG collected from nine subjects investigate effects varying parameters model performance. Offline results show that proposed achieved stable improvement effect. conduct an online experiment six assess developed real-time application. The yields promising average 83.56% using data length even only one second, was 12.26% higher than standard algorithm. These confirmed effectiveness recognition demonstrated its development improved