A3 Refereed book chapter or chapter in a compilation book
EEG-based secure authentication mechanism using discrete wavelet transform and ensemble machine learning methods
Authors: Subasi, Abdulhamit; Qaisar, Saeed Mian; Sarirete, Akila
Editors: Subasi, Abdulhamit; Qaisar, Saeed Mian, Nisar, Humaira
Publisher: Elsevier
Publication year: 2025
Book title : Artificial Intelligence and Multimodal Signal Processing in Human-Machine Interaction
Journal name in source: Artificial Intelligence and Multimodal Signal Processing in Human-Machine Interaction
Series title: Artificial Intelligence Applications in Healthcare and Medicine
First page : 283
Last page: 302
ISBN: 978-0-443-29150-0
eISBN: 978-0-443-29151-7
DOI: https://doi.org/10.1016/B978-0-443-29150-0.00004-4
Web address : https://doi.org/10.1016/B978-0-443-29150-0.00004-4
In recent years, there has been growing interest in using electroencephalography (EEG) signals for secure authentication due to their unique characteristics and potential applications in security systems. While brain signals have long been studied in clinical contexts, their utilization as biometric identifiers in automated recognition systems has only recently garnered attention within the scientific community. Brain signals offer distinct advantages that are not present in conventional biometrics such as face, iris, and fingerprints, including compliance with privacy regulations, resilience to spoofing attempts, continuous identification capabilities, inherent liveness detection, and universality, making them an appealing option. Nonetheless, several challenges must be addressed, including understanding the uniqueness and stability of brain responses, developing stimulation protocols, and managing the noninvasiveness of the acquisition process. This chapter proposes an EEG-based secure authentication mechanism employing Discrete Wavelet Transform (DWT) in conjunction with ensemble machine learning methods. EEG signals, known for their unique characteristics and resistance to spoofing attacks, offer promising potential for secure authentication systems. The discrete wavelet transform is utilized for feature extraction from EEG signals, capturing their intricate temporal and spectral patterns. Ensemble machine learning methods are then employed to effectively classify EEG signals and authenticate users. The proposed approach aims to enhance the accuracy and robustness of EEG-based biometric identification systems by integrating multiple classifiers and leveraging the complementary strengths of each method. Experimental results demonstrate the effectiveness of the proposed approach in accurately identifying individuals based on their EEG signals, highlighting its potential for applications in security, access control, and authentication systems. This research contributes to the advancement of biometric identification technology and underscores the promise of EEG signals as a reliable secure authentication.
