Authors:
Gerardo Canfora
1
;
Paolo di Notte
2
;
Francesco Mercaldo
3
and
Corrado Aaron Visaggio
1
Affiliations:
1
University of Sannio, Italy
;
2
Reply, Italy
;
3
University of Sannio and Centro Regionale Information Communication Technology - CeRICT scrl, Italy
Keyword(s):
Continuous Authentication, Silent Authentication, Security, Behavioral Models, Android.
Related
Ontology
Subjects/Areas/Topics:
Biometrics Security and Privacy
;
Identification, Authentication and Non-Repudiation
;
Information and Systems Security
;
Security and Privacy in Mobile Systems
Abstract:
Due to the increasing pervasiveness of mobile technologies, sensitive user information is often stored on
mobile devices. Nowadays, mobile devices do not continuously verify the identity of the user while sensitive
activities are performed. This enables attackers full access to sensitive data and applications on the device, if
they obtain the password or grab the device after login. In order to mitigate this risk, we propose a continuous
and silent monitoring process based on a set of features: orientation, touch and cell tower. The underlying
assumption is that the features are representative of smartphone owner behaviour and this is the reason why
the features can be useful to discriminate the owner by an impostor. Results show that our system, modeling
the user behavior of 21 volunteer participants, obtains encouraging results, since we measured a precision in
distinguishing an impostor from the owner between 99% and 100%.