Design Of Key Protection Circuit For Smart Card Based On Fingerprint Fuzzy Vault

We are living in an era of information development, face a lot of information security issues. The information encryption helps us to solve the security issues of transmission, but the key's security of encryption is a problem. Jules and Sudan proposed a scheme of fingerprint encryption based on Fuzzy vault, which mixed and encrypted the fingerprint features and the key, on the one hand, protecting the security of fingerprint and key, on the other hand, it makes the key recovery more convenient.This paper's aim is the protection of key in smart card, improves the problem of traditional fuzzy vault for smart card. Firstly, the way of pure software achieves the encryption system, the speed of compute is slow, traditional Clancy's chaff point generation algorithm is time-consuming. In view of the this problem, this paper designed the system based on the method of software and hardware, the way of hardware accelerated designed the most time-consuming chaff point and feature point matching, thereby reduced the running time of the system. Secondly, the method of fingerprint matching based on local information or geometric hash require a large storage capacity, smart card don't has so much storage capacity. This paper improved the algorithm of matching for smart card, made the calculation time, circuit area and storage meet the needs of application.Finally, the circuit area which based on the SMIC0.13?m technology is evaluated to13389 gates, and memory capacity is 40 K. At the same time, evaluating resource consumption as 361 slices based on FPGA of XilinxK7,the circuit of polynomial calculation designed two schemes, and chosen the scheme of smaller area. Increasing the module of address coding at chaff generation circuit, so that the operating speed is improved about 2times, is improved about 3 times more than the recent literature. Also this paper improved the algorithm of matching and the alignment of feature points at decoding stage, compared it with the algorithm before improved, the false rejection rate reduced about 2%,the speed increased 1 time, hardware accelerated speed increased about 10 times more than pure software.