Table 13.
Comparison with state-of-the-art EEG-based biometric systems.
| Study | Approach | Reported accuracy | Paradigm |
|---|---|---|---|
| Roy et al.8 | CNN-based deep learning | >95% | Task-related EEG |
| Zhang et al.21 | CNN + RNN hybrid | 96–97% | Resting state |
| Thomas et al.3 | Multimodal (EEG + ECG) |
98% |
Event-related potentials |
| Arnau et al.24 | Brain Encoding Dataset (BED) release | 88 | Resting, cognitive, and VEP/VEPC |
| This work | Lenient PREP + MFCC + XGBoost | 98.0% | VEPC 10 Hz |
Bold indicates the best/ maximum reported identification accuracy among the compared state-of-the-art methods.
