Cognitive biometric cryptosystems a case study on EEG

Abstract

In this paper we propose several cryptosystems for the protection of templates extracted from cognitive biometrics. The proposed architectures exploit different possibilities for combining the information made continuously available during the recognition phase. Specifically, we focus on electroencephalog-raphy (EEG) signals as considered biometrics. Brain waves are in fact one of the most emerging cognitive modalities to be used for people recognition. An extensive set of experimental tests, performed on a large database comprising recordings taken from 40 healthy subjects during two separate recording sessions, is carried out to evaluate the recognition rates and security levels achievable with several system configurations.