An SSVEP Recognition Method by Combining Individual Template with CCA

Abstract

Steady-state visual evoked potential (SSVEP) based Brain--computer interface (BCI) is one of the non-invasive BCI systems that has received great attention over the years. It requires shorter training and provides relatively higher information transfer rate. Various approaches have been developed to recognize SSVEP. One popular approach is based on canonical correlation analysis (CCA). The standard CCA uses artificial reference templates of pure sine and cosine signals with frequencies corresponding to the frequencies of the visual stimuli. These reference templates may not optimally reflect the natural SSVEP features hidden in the recorded EEG signal. Thus, they may not produce high performance. This paper introduced a new approach called ITCCA (Individual Template Canonical Correlation Analysis). ITCCA optimized the reference signal by incorporating features of the subject's EEG data into the artificial reference template. The performance of this approach was compared with that of the standard CCA. The experiment showed that ITCCA obtained better results than the standard CCA method.