Electroencephalography (EEG) is an emerging factor for biometric authentication due to quantifiable brainwave signaling behaviors and data patterns. Using EEG technology has been demonstrated as an approach to authenticate users to secure computing systems, but presents a challenge when users are impaired or inebriated due to alcohol consumption. The influential behavior of alcohol presents a bias into EEG measurements thus, leading to invalid authentication attempts. In this presentation, we provide discussion on the use of EEG measurements as a biometric authentication factor, and express techniques to authorize inebriated–or more specifically–drunk users. Our approach utilizes machine learning algorithms to automatically factor users’ brainwave behaviors for both normal and drunken states. We evaluate our approach using an EEG dataset as preliminary work and validate our findings with real world experiments using a 5-channel EEG wireless headset. Our experimental evaluation provides preliminary work and demonstrates how EEG measurements provide feasibility as a biometric authentication factor during scenarios when a user is impaired.
Tommy Chin is a Security Researcher at Grimm (SMFS, Inc.) and holds numerous academic degrees, certificates, and published works. His current interests focus towards networking, tracking systems, moving target defense, machine learning, and neural networks. He is also an alumnus of Rochester Institute of Technology and a member of SPARSA and RC3.
Peter Muller is a Graduate Student at Rochester Institute of Technology, where he spends his time scripting and networking. He has interests in machine learning, embedded systems, and reverse engineering, touching anything from neural networks to Nintendo 64 processors. He is also a member of SPARSA.