A novel identification/verification model using smartphone's sensors and user behavior

Smartphones are increasingly entering people's life; every person in the house carry one or two smartphones (Android, iPhone, Tab…). They use explicit authentication, which is inefficient; once the smartphone is stolen, a thief can steal personal information stored on the phone and can access all services that might have the password stored. In addition, elderly and physically impaired users need to have their medical profile secured and easily accessed without password limitation. For this reason, smartphone sensors are good candidates for providing an implicit authentication. This work introduces a new perspective of context-based user authentication: users can be authenticated implicitly using data captured by sensors of the smartphone and user behavior; these data are used in the creation of a unique profile for each user. The proposed model is supposed to be as secure as traditional authentication methods.