How time window influences biometrics performance: an EEG-based fingerprints connectivity study

TL;DR

This study examines how the duration of EEG signal segments affects the accuracy of biometric identification based on brain connectivity, highlighting the importance of optimal time window selection.

Contribution

It demonstrates the significant impact of time window length on EEG biometric performance and identifies an optimal window size for connectivity-based recognition.

Findings

Performance improves with longer time windows up to a point.

EEG connectivity features are promising for biometric identification.

Extending beyond the optimal window does not enhance performance.

Abstract

EEG-based biometric represents a relatively recent research field that aims to recognize individuals based on their recorded brain activity by means of electroencephalography (EEG). Among the numerous features that have been proposed, connectivity-based approaches represent one of the more promising methods tested so far. In this paper, we investigate how the performance of an EEG biometric system varies with respect to different time windows to understand if it is possible to define the optimal duration of EEG signal that can be used to extract those distinctive features. Overall, the results have shown a pronounced effect of the time window on the biometric performance measured in terms of EER (equal error rate) and AUC (area under the curve), with an evident increase of the biometric performance with an increase of the time window. In conclusion, we want to highlight that EEG connectivity has the potential to represent an optimal candidate as EEG fingerprint and that, in this context, it is very important to define a sufficient time window able to collect the subject specific features. Moreover, our preliminary results show that extending the window size beyond a certain maximum does not improve biometric systems' performance.