PEBIID: Privacy-preserving and Efficient Biometric Identification for IoV DApp

Chun Liu, Lin Yang, LinRu Ma, LiuCheng Shi, Xuexian Hu, Weipeng Cao, JingJing Zhang

Abstract: With the rapid development of Internet-of-Vehicle (IoV) and Blockchain, decentralized application (DApp) will be equipped increasingly into the IoV. To obtain individualized service of DApp with personal application data protected, the user driving a vehicle has to be authenticated. However, inputting a secure but complicated password to login in a DApp while driving may well divert the driver’s attention and even lead to a traffic accident. For the sake of traffic safety, the convenient and reliable biometric identification will better be employed as the authentication method. Due to the highly sensitivity of user’s biometric data, we propose a privacy-preserving and efficient biometric identification for IoV DApp (PEBIID) scheme based on the subtly designed invertible matrix (SDIM) to protect the biometric data. Compared with the existing matrix-transform-based privacy-preserving biometric identification scheme, the PEBIID’s privacy security is enhanced, defending against not only common attacks mentioned in the previous work, but also the novel statistical-inference attack. Further, our scheme introduces less overhead of computation and communication to protect the biometric data, meaning it is more suitable for the IoV DApp authentication scenario where the IoV topology dynamically changes greatly with a lower packet loss rate of authentication request. Moreover, by reason of less overhead introduced for biometric data protection, the limited energy and computing resources of vehicles can be saved more.