Inference-Based Similarity Search in Randomized Montgomery Domains for Privacy-Preserving Biometric Identification

TL;DR

This paper introduces a privacy-preserving similarity search method for biometric identification that uses obfuscated distance measures and Montgomery multiplication, achieving high accuracy with significantly reduced computational costs.

Contribution

It presents a novel inference-based framework utilizing randomized Montgomery domains to enhance privacy and efficiency in biometric similarity search.

Findings

Achieves near-optimal search accuracy

Reduces computational cost by orders of magnitude

Provides strong privacy guarantees against adversarial analysis

Abstract

Similarity search is essential to many important applications and often involves searching at scale on high-dimensional data based on their similarity to a query. In biometric applications, recent vulnerability studies have shown that adversarial machine learning can compromise biometric recognition systems by exploiting the biometric similarity information. Existing methods for biometric privacy protection are in general based on pairwise matching of secured biometric templates and have inherent limitations in search efficiency and scalability. In this paper, we propose an inference-based framework for privacy-preserving similarity search in Hamming space. Our approach builds on an obfuscated distance measure that can conceal Hamming distance in a dynamic interval. Such a mechanism enables us to systematically design statistically reliable methods for retrieving most likely candidates without knowing the exact distance values. We further propose to apply Montgomery multiplication for generating search indexes that can withstand adversarial similarity analysis, and show that information leakage in randomized Montgomery domains can be made negligibly small. Our experiments on public biometric datasets demonstrate that the inference-based approach can achieve a search accuracy close to the best performance possible with secure computation methods, but the associated cost is reduced by orders of magnitude compared to cryptographic primitives.