A Continuous Identity Authentication Scheme Based on Physiological and Behavioral Characteristics.

摘要: Wearable devices have flourished over the past ten years providing great advantages to people and, recently, they also been used for identity authentication. Most of authentication methods adopt a one-time manner which cannot provide continuous certification. To address this issue, we present two-step method based on an own-built fingertip sensor device can capture motion data (e.g., acceleration and angular velocity) physiological photoplethysmography (PPG) signal) simultaneously. When is worn user’s fingertip, it will automatically recognize whether wearer legitimate user or not. More specifically, multisensor collected analyzed extract representative intensive features. Then, human activity recognition applied as first step enhance practicability system. After correctly discriminating state, one-class machine learning algorithm second step. wears device, process carried at set intervals. Analyses were conducted using from 40 individuals across various operational scenarios. Extensive experiments executed examine effectiveness proposed approach, achieved average accuracy rate 98.5% F1-score 86.67%. Our results suggest that scheme provides feasible practical solution