Non-intrusive and Transparent Authentication on Smart Phones

Introduction

This work aims to contribute to the field of non-intrusive and transparent authentication on smart phones by defining an implicit authentication model consisting of a set of distinguishable recurring features extracted from a combination of different sources of inbuilt sensors which have not yet been previously combined for this purpose. The research goals of this work are (1) define a robust methodology for accurate and transparent sensor data collection (2) identify sets of distinguishable and recurring features to define an implicit authentication model and (3) evaluate the usability and security threats of this authentication model so that a smart phone could be trained for a brief period of time after which it will be capable of authenticating users in a non-intrusive and transparent manner.