Development of an Improved Access Control System Using a Hybrid Authentication

Abstract

Data security is indispensable in our society, as human activities are increasingly becoming computerized and are transferred through complex communication networks across the globe. The existing methods used in securing data include cryptography and biometrics. Significant amounts of research have gone into the integration of biometrics with cryptography with the potential to augment access control and secure transmission of sensitive biometric data, it is expected that a robust security system should provide a 100% safety which the world is still far from. On this note, this paper developed an improved access control system using a combination of cryptography and bimodal biometric with emphasis to enhance the techniques such as Advanced Encryption Standard based Fast Fourier Transform (AES-FFT) for encryption and decryption and using Gravitational Search Algorithm (GSA) to elect an optimal feature selection. Five hundred and seventy (570) face and iris images were acquired from iris and face dataset to evaluate recognition accuracy, false positive rate, sensitivity, specificity and average recognition rate time. The developed AES-FFT technique ensured scalable encrypting capacity, good imperceptibility and security performance against various attacks therefore, this vindicate AES-FFT as a better technique in bimodal bio-cryptography with optimal computationally efficiency than existing AES in terms of accuracy and time.