Methods of reducing bio-cryptographic algorithms computational complexity

Abstract

Traditional biometric systems have always been regarded with skepticism, due to the uniqueness and universality features of human biometrics - once stolen or compromised, they cannot be safely replaced. For this reason, common efforts of Cryptography and biometric systems researchers gave birth to Bio-cryptography, a domain which proposes to overcome the main security issues associated with biometrics usage, by generating or securing existing cryptographic keys, using biometric data. The main advantages brought by Bio-cryptography are represented by eliminating the need for storing biometric templates and the flexibility of resulted bio-cryptograms. On the other hand, one major drawback concerning bio-cryptographic algorithms is their high consumption of computational resources, caused by the various mathematical processing and large amounts of data used. This paper proposes two methods of reducing bio-cryptographic algorithms computational complexity: the usage of error-correcting codes in the secret key recovery stage and the shared biocryptograms approach for the entire biometric encryption process.