Continuous Trust Without Compromise: Privacy-Preserving Behavioral Biometrics in Zero-Trust Authentication
Wafa Hussain Fadaaq (Department of IT and Computer Science, Stardom University, Istanbul), Wisam Hazim Gwad, Shahab Wahhab Kareem
Abstract
This paper promotes a framework that combines behavioral biometrics, federated learning, and Zero Trust as foundations of continuous authentication. Behavioral modalities like keystroke dynamics, mouse movements, gestures and motion signals give dynamic identity traits difficult to forge. Federated learning guarantees privacy protection because raw biometric data are only stored on user devices. The proposed framework achieves an Equal Error Rate (EER) of only 7%, lower False Acceptance and Rejection Rates, and real-time latency of less than 200 ms.