Brain waves for biometric user recognition

Abstract

Brain signals are being investigated within the medical field for more than a century to study brain diseases like epilepsy, spinal cord injuries, Alzheimer, Parkinson, schizophrenia, and stroke among the others. They are also used as the basis of brain computer interface and brain machine interface with assistance, rehabilitative and entertainment applications. Despite the broad interest in clinical applications, the use of brain signals sensed by means of electroencephalogram has been only recently investigated by the scientific community as a biometric characteristic, and the use of brain waves for the purpose of automatic people recognition is only at an embryonic stage. However, brain signals present some peculiarities, not shared by the most commonly used biometrics, like face, iris, and fingerprints, concerning secretness, privacy compliance, robustness against spoofing attacks, possibility to perform continuous identification, intrinsic liveness detection, and universality. Moreover since 70s there is evidence that EEG signals are genetically influenced and that they carry personality correlates. These peculiarities make the use of brain signals appealing. On the other hand there are many challenges related to the use of brain signals which need to be properly addressed in order to deploy biometric systems based on brain activity in real life applications. Among these challenges, the definition of the brain response elicitation protocol and the convenience of the acquisition process should be addressed, to cite a few. In this work of doctoral thesis the aforementioned issues are further developed. The first chapters provide a comprehensive and critical review of state of the art methods for electroencephalogram based automatic user recognition that have been proposed in literature so far, also pointing out neurophysiological evidences related to the performed claims. The methods implemented and the experiments carried on within the present work are reported and detailed in the next chapters, together with the results obtained from the analysis of data.