HMOG: New Behavioral Biometric Features for Continuous Authentication of Smartphone Users

摘要

We introduce Hand Movement, Orientation, and Grasp (HMOG), a set ofbehavioral features to continuously authenticate smartphone users. HMOGfeatures unobtrusively capture subtle micro-movement and orientation dynamicsresulting from how a user grasps, holds, and taps on the smartphone. Weevaluated authentication and biometric key generation (BKG) performance of HMOGfeatures on data collected from 100 subjects typing on a virtual keyboard. Datawas collected under two conditions: sitting and walking. We achievedauthentication EERs as low as 7.16% (walking) and 10.05% (sitting) when wecombined HMOG, tap, and keystroke features. We performed experiments toinvestigate why HMOG features perform well during walking. Our results suggestthat this is due to the ability of HMOG features to capture distinctive bodymovements caused by walking, in addition to the hand-movement dynamics fromtaps. With BKG, we achieved EERs of 15.1% using HMOG combined with taps. Incomparison, BKG using tap, key hold, and swipe features had EERs between 25.7%and 34.2%. We also analyzed the energy consumption of HMOG feature extractionand computation. Our analysis shows that HMOG features extracted at 16Hz sensorsampling rate incurred a minor overhead of 7.9% without sacrificingauthentication accuracy. Two points distinguish our work from currentliterature: 1) we present the results of a comprehensive evaluation of threetypes of features (HMOG, keystroke, and tap) and their combinations under thesame experimental conditions, and 2) we analyze the features from threeperspectives (authentication, BKG, and energy consumption on smartphones).