Chaotic construction of cryptographic keys based on biometric data
Ihsen Nakouri, Mohamed Hamdi, Tai-hoon Kim · 2016
The protection of sensitive data stored on cloud platforms has been intensively addressed in the literature. One of the most promising solutions is the generation of biometric-based cryptographic keys that can be used for authentication, encryption, and biometric template protection. Multiple important challenges are faced when dealing with such applications is the invariance of cryptographic keys to the modifications that occur in the biometric features. Fuzzy extractors have been proposed in the literature as efficient tools that cope with this issue by processing the noisy biometric data in order to produce a random cipher key. In this paper, we extend this concept through the introduction of a new construction scheme that relies on the use of chaotic maps in order to enhance the sensitivity of the key generation process to the biometric input. Throughout analytical and experimental performance evaluation, we demonstrate that the proposed scheme outperforms existing solutions especially in terms of entropy loss.