eduzhai > Applied Sciences > Engineering >

MusicID A Brainwave-based User Authentication System for Internet of Things

  • Save

... pages left unread,continue reading

Document pages: 8 pages

Abstract: We propose MusicID, an authentication solution for smart devices that usesmusic-induced brainwave patterns as a behavioral biometric modality. Weexperimentally evaluate MusicID using data collected from real users whilstthey are listening to two forms of music; a popular English song andindividual s favorite song. We show that an accuracy over 98 for useridentification and an accuracy over 97 for user verification can be achievedby using data collected from a 4-electrode commodity brainwave headset. Wefurther show that a single electrode is able to provide an accuracy ofapproximately 85 and the use of two electrodes provides an accuracy ofapproximately 95 . As already shown by commodity brain-sensing headsets formeditation applications, we believe including dry EEG electrodes insmart-headsets is feasible and MusicID has the potential of providing an entrypoint and continuous authentication framework for upcoming surge ofsmart-devices mainly driven by Augmented Reality (AR) Virtual Reality (VR)applications.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×