Privacy–Security Trade-Offs in Biometric Security Systems—Part I: Single Use Case

Abstract

This is the second part of a two-part paper on the information theoretic study of biometric security systems. In this paper, the performance of reusable biometric security systems, in which the same biometric information is reused in multiple locations, is analyzed. The scenario in which the subsystems are jointly designed is first considered. An outer bound on the achievable trade-off between the privacy leakage of the biometric measurements and rates of keys generated at the subsystems is derived. A scheme that achieves the derived outer bound is then presented. Next, an incremental design approach is studied, in which the biometric measurements are reused while keeping the existing system intact. An achievable privacy-security trade-off region for this design approach is derived. It is shown that under certain conditions, the incremental design approach can achieve the performance of the joint design approach. Finally, examples are given to illustrate the results derived.