CAREER: Advancing Fairness in Biometric Systems: Towards Security and Privacy Enhancement

Over the past two decades, the use of biometric recognition such as fingerprint or facial recognition, has grown significantly and found applications worldwide due to its ability to provide accurate authentication and convenient user experiences. However, ongoing issues related to privacy and security have emerged, raising concerns within both the scientific community and the general public.

Balancing these concerns without compromising system security and performance is a significant challenge. The goal of this NSF CAREER research project is to investigate biometric system security vulnerabilities and develop template data protection mechanisms to enhance the safety and privacy of these systems. The proposed effort also includes educational and outreach activities: Youth Cybersecurity Research (YCR), a program to match interested high school students with faculty with shared research interests; Datathon, an annual full-day focused on the intersection of cybersecurity and biometrics; a student context at a major Biometrics conference; and development of new course materials related to biometric security, for use in both existing and new courses.

The goal of this project is to develop a comprehensive and standardized framework for analyzing both hardware and software attacks on biometric systems while addressing issues of fairness and bias. The research focuses on two main thrusts: investigating vulnerabilities and developing robust countermeasures. This project aims to establish a research infrastructure that identifies unexplored vulnerabilities of biometric systems, such as side-channel and hardware fault injection attacks, as well as leakage through bias sources.

It also seeks to develop a general evaluation methodology for side-channel attacks in biometrics, including metrics, protocols, and result-reporting procedures. Additional avenues of exploration include developing scalable and universal approaches to enhance existing anti-spoofing biometric techniques and to deepen the understanding of ethical considerations in biometric security.

This exploration can lead to advancements in biometric security and contribute to the knowledge base surrounding the protection of sensitive biometric information.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.