Behavioral biometric authentication on smartphone using keystroke dynamics

Abstract

Over the past few years, smartphones have revolutionized our life. Authentication plays important role as all private and business data are stored in smartphone. Password authentication is not secure technique as password can be easily stolen by hackers. Hence, need to use advance technology to increase the security of authentication. Keystroke Dynamics is behavioral based authentication technique. It identifies the user identity through their typing behavior to check user is genuine or imposter. In this paper author is presenting a KDSP system for Smart devices to increase the security of authentication at login time. On the basis of timing feature set, typing speed, flight time and error rate of keystroke dynamics author is checking the identity of the user. Proposed system works on all the Android smartphones and tablets. All experimental work was performed on 268 subjects to capture keystroke data on server. In Statistical analysis, Equal-error-rate (EER) is measured 7.415% of the 60 subjects for False acceptance rate (FAR) and 208 subjects for False rejection rate (FRR).