Secure Biometric Authentication Technology Based On Homomorphic Encryption Or Piecewise Logistic MAP

In the era of big data,data is wealth.Secure identity authentication is an important guarantee for users to enjoy proprietary information and services.In recent years,biometric identity authentication technology has been rapidly developed and widely used due to its safety and convenience.Its development has also brought new impacts and challenges to the data privacy security in the traditional service models.But biometric-based identification has also exposed some security risks.Due to these reasons,it is of great practical significance to study secure biometric authentication.The key to secure biometrics is the security of biometric templates.Solving the security concerns of user's biometrics stored in the third-party servers and studying the privacy preserving biometric authentication technology are the focus of this paper.Based on this fact,we have proposed a secure biometric authentication scheme applying homomorphic encryption and a revocable biometric template authentication scheme based on chaotic mapping,respectively.These schemes change the way of the traditional identity authentications which directly stored the biometric templates in plaintext,our proposed schemes enable the storage and matching of biometrics in the encryption domain,and improved the security of the biometric identity authentication in open network.The main research work of this paper is as follows1.The key technologies for secure biometric authentication that is biometric encryption and biometric transformation.From the perspective of cryptography to the strong protection of data,homomorphic encryption and chaos theory are studied,respectively.The characteristics of each technique and its applicability to biometric protection are analyzed.2.A secure biometric authentication scheme based on homomorphic encryption technology is proposed.The scheme combines the homomorphic encryption technology based on ring learning with errors with an improved message coding scheme.Firstly,the extracted biometric data is packaged by the message coding scheme,then the packaged biometrics are encrypted by homomorphic encryption,and the encrypted data is stored as a template.The matching of the biometric information to be authenticated with the template is performed in the encryption domain.The scheme uses the Hamming distance to perform similarity matching.The calculation of the Hamming distance requires only one multiplication,which greatly reduces the matching time.The scheme adds random information to protect the intermediate data transmission between the client and the server.The experimental simulation and system security analysis show that the scheme can achieve biometric authentication in the encryption domain with both small amount of calculation and short time-consuming,and it can resist internal server attacks,replaying attacks,and stealing intermediate data attacks.It has good applicability to complex network environments.3.A revocable biometric template authentication scheme based on chaotic map is proposed.The scheme uses an improved piecewise Logistic chaotic mapping system to generate three key sequences to encrypt the user's biometric template.The dynamic range of the optional parameters and the key space of the improved piecewise Logistic chaotic mapping system are larger than the traditional Logistic chaotic mapping system,and the security is higher.Our simulation uses finger vein features for validation,and our experimental results show that this scheme can achieve biometric authentication in the encryption domain.The improved piecewise Logistic mapping system has a good initial sensitivity,and the small change of initial parameter would lead to huge differences in the encryption template.Our scheme can revoke and replace the new template in time whenever the biometric template is attacked,and it will ensure the security of the biometric authentication system.