EEG personal recognition based on ‘qualified majority’ over signal patches

Electroencephalography (EEG)-based personal recognition in realistic contexts is still a matter of research, with the following issues to be clarified: (1) duration signal length, called ‘epoch’, which must very short for practical purposes and (2) contribution EEG sub-bands. These two aspects are connected because shorter epoch’s duration, lower low-frequency sub-bands while enhancing high-frequency However, it well known that former characterises inner brain activity resting or unconscious states. could no use wild, where subject conscious not condition put himself resting-state-like condition. Furthermore, latter may concur much better process, characterising normal when awake. This study aims at clarifying problems mentioned above by proposing novel architecture based on extremely fragments ‘patches’, subdividing each epoch. Patches individually classified. A ‘qualified majority’ classified patches allows taking final decision. It shown experiments this approach can adopted clarifies sub-bands’ role implemented vitro but similar conceivable wild.