New Fuzzy Vault Scheme Based On Biological Features And Research On Three-factor Identity Authentication Scheme

With the decreasing of the cost of biometric devices and the development of biometric technology,biometrics have been playing an increasingly important role in people's daily lives.Biometric authentication is one of the most important applications of biometrics.Handwritten signatures in financial business,fingerprint recognition embedded in intelligent mobile phones,face recognition in access control systems,and so on are the practical applications of biometrics in identity authentication.However,it is not good for practical applications to use biometric authentication only.In order to enhance the security of an authentication system,it is imperative to combine traditional identity authentication technology and current biometric authentication technology.Nowadays,a new important application of biometrics is to keep a cryptographic key secure.Their structures of the standard cryptosystem such as AES,RSA and so on are open,hence,the security of a cryptosystem does not depend on the cryptosystem itself,but on the security of its employed key.Biometric encryption can bind the key and biometric security together,which not only ensures tsecurity of the key,but also ensures the security of the biometric features.Fuzzy vault scheme is a kind of key binding scheme,which is based on the combination of biometrics and cryptography.The wide applications of biological characteristics have brought about the security and privacy of biometric features,and the "fuzziness" of biological features has a natural contradiction with the precise cryptosystem.The main work of this thesis is as follows:1.This thesis briefly introduces the research and applications of biometric identification technology,biometric encryption technology and remote authentication scheme based on biometric features.2.After careful analysis on the biological template protection scheme based on Fuzzy vault,we proposes an improved dual authentication scheme based on CRC cyclic redundancy check and the discrete logarithm problem.this scheme adds a parameter set(p,g,gK)in the fuzzy vault.Among the parameter set,p is the selected prime number,g is a generator of the finite multiplication group Fp*,K is the secret data to be preserved in the fuzzy vault.Since it is difficult to ask k when it is known(p,g,gk),the parameter group(p,g,gk)will not reveal the information of the secret data k.Therefore,the scheme of this paper not only solves the CRC collision problem in CRC based fuzzy vault,but also makes it a kind of biometric authentication scheme which can resist the blend substitution attack.3.Combined with the traditional identity authentication technology and biometric authentication technology,we proposes a three-factor remote user authentication protocol with error tolerance.This is a three-factor authentication protocol using "password + smart card + biometrics"which is an improved protocol based on Wen's scheme.The secure sketch algorithm and the elliptic curve version of the Diffie-Hellman algorithm applied in our scheme can effectively overcome the shortcoming existed in Wen's scheme,such as the biometrics recognition error,session key being exposed to the registration center and other defects.At the same time,our protocol keep the advantages of the original protocol in overcoming these defects.