CLAPAUTH: A Gesture-Based User-Friendly Authentication Scheme to Access a Secure Infrastructure

In this paper we propose a gesture-based user-friendly smartwatch-based user authentication scheme called CLAPAUTH to authenticate the users to gain physical access to a secure infrastructure. In CLAPAUTH users are authenticated by performing clapping actions, while wearing their smartwatch in one hand. CLAPAUTH, while users perform clapping gestures, profiles them by collecting data from their smartwatches' built-in accelerometer and gyroscope sensors. We have evaluated the proposed scheme on a publicly available dataset by using state-of-the-art n-class machine learning classifiers, namely Random Forest (RF), Artificial Neural Network (ANN), and K-Nearest Neighbors (KNN). KNN outperformed other two classifiers and attained 93.3% TAR at the cost of 0.22% FAR. CLAPAUTH could be widely accepted as it utilizes users' familiarity with a common action, such as clapping, and users are not required to remember any secret code or gesture.