Table 1.
Different methods used for SSVEP recognition in BCI
| Method | Concept | Training Requirement | Reference | The Number of EEG Channels Used in the Study |
|---|---|---|---|---|
| PSDA | Significant peaks at the frequencies of the stimuli are detected from Power Spectral Density of the user’s EEG signal within a time window | _ | (Ming et al., 2002) | 2 |
| CCA | A method for exploring the relationship between two multivariate sets of vectors | _ | (Lin et al., 2007) | 8 |
| MCCA | It uses the optimal reference signals after adjustment, with increased computational time | Yes | (Yu Zhang et al., 2011b) | 8 |
| L1MCCA | This method is an extension of the CCA for reference signal optimization | Yes | (Yu Zhang et al., 2013) | 8 |
| LASSO | It assumes that SSVEPs are standard linear regression models of stimulation signals | _ | (Yu Zhang et al., 2012) | 3 |
| MsetCCA | An extension of CCA to recognize multiple linear transforms to optimize signal references with EEG signals | Yes | (Yangsong Zhang et al., 2014) | 8 |
| CFA | A method to exploit the latent common features shared by a set of EEG signals experiments as the improvement reference | Yes | (Yu Zhang et al., 2015) | 8 |
| MLR | Multivariate Linear Regression is implemented to exploit the distinguished SSVEP components | Yes | (H. Wang et al., 2016) | 8 |