Privacy preserving biometric authentication based on fully homomorphic encryption, blockchain, and IPFS data storage

Privacy preserving biometric authentication based on fully homomorphic encryption, blockchain, and IPFS data storage

연구 분야: Cryptography

학회: Multimedia Tools and Applications

The widespread adoption of biometric authentication systems introduces privacy concerns due to the reliance on centralized storage and processing of sensitive data. This paper proposes a novel privacy-preserving approach that addresses these limitations. Our system leverages Fully Homomorphic Encryption (FHE) to ensure the confidentiality of biometric templates stored in the decentralized InterPlanetary File System (IPFS). Additionally, blockchain technology guarantees the integrity of both biometric templates and matching results. Complex distance computations are performed securely off-chain and retrieved through a blockchain Oracle. This integrated approach mitigates security vulnerabilities like template modification, channel interception, and override comparator. To evaluate the effectiveness and robustness of our approach, we conducted experiments on various fingerprint databases. The results are promising: we achieved preserved accuracy (0% Equal Error Rate difference). Additionally, the encrypted distance calculation using FHE is efficient, taking only 0.00144seconds, and enhances privacy for biometric data. Notably, our approach eliminates the need for a trusted central server, addressing a key limitation of existing methods.