Review Article
Review on EEG-Based Authentication Technology
Table 4
Deep learning methods for EEG-based authentication.
| Research | Tasks | Number of subjects | Number of electrodes | Deep learning model | Accuracy (%) |
| Sun et al. [72] | Resting state | 109 | 16 | LSTM | 99.58 | Mao et al. [50] | ERP | 100 | 64 | CNN | 97.00 | Wang et al. [24] | Resting state | 109 | 64 | GCNN | 99.98 | Das et al. [37] | MI | 40 | 17 | CNN | 99.30 | Wilaiprasitporn et al. [71] | ERP with emotion | | | CNN + RNN | 99.90 | Chen et al. [58] | ERP | 100 | | GSLT-CNN | 97 | RSVP | 10 | | 99 | ERP with emotion | 32 | | 99 | Multiple data sets | 157 | | 96 |
| Zhang et al. [26] | Resting state | 8 | 14 | Attention-based RNN | 98.20 | MI | 8 | 64 | 99.89 |
| Wu et al. [34] | RSVP with eye blinking | 10 | 16 | CNN | 97.60 | Kumar et al. [29] | VEP | 58 | 14 | LSTM | 97.57 | Wang et al. [73] | SSVEP | 10 | 8 | CNN | 99.73 |
|
|