| Abstract: | This paper presents a new efficient and lightweight approach for enhancing the security of biometric models, namely, fingerprint templates, against possible attacks. The proposed design is based on Vernam stream cipher in which the key generator is designed in the hardware manner. The designed cryptosystem consists of using multi‐scroll chaotic system that is characterized by a large key space and can be generated N×N grid multi‐scroll attractors, with a good behavior of chaotic dynamic. The hardware approach is carried out through describing Euler method by VHDL. Field‐programmable gate array (FPGA) experimental results validate the developed architecture while still providing a good compromise between hardware resources and performance. Indeed, security analysis also shows that the designed encryption algorithm is robust against statistical, brute force, and entropy attacks. Therefore, it can be considered as a lightweight security solution, which could be very useful in many embedded applications namely securing biometric authentication systems. |
