EPNS: Efficient Privacy-Preserving Intelligent Traffic Navigation From Multiparty Delegated Computation in Cloud-Assisted VANETs

Real-time navigation is an essential service in many applications, such as intelligent transportation and crowdsensing. There are, however, considerations (e.g., the potential of location privacy breach for vehicular users) in real-time navigation services. We observe that most existing state-of-the-art approaches utilize either pseudonyms or public key fully homomorphic encryption (FHE) to protect location privacy. The former (pseudonym-based approach) relies on an online certificate authority (CA) to update pseudonym certificates periodically, and FHE-based approaches incur significant computational/communication overheads. In this paper, a new cryptographic primitive, coined efficient multiparty delegated computation (MPDC), is proposed, where any one-way trapdoor permutation is required to perform in constant time (e.g., twice in our context) on each resource-constrained data provider for the encryption of (multiple) messages in a batch. We also remark that MPDC is designed to support secure evaluations (e.g., addition, multiplication, equality, and less than operations) over ciphertexts encrypted under multiple keys associated with different entities. Building on MPDC, we devise a lightweight privacy-preserving real-time intelligent traffic navigation scheme (EPNS) for cloud-assisted vehicular ad hoc networks (VANETs). The proposed approach facilitates the prediction of an optimal driving route, in terms of the shortest time from source to destination. In other words, EPNS securely evaluates the auto-regression moving average (ARMA) model with spatiotemporal correlations of high accuracy and efficiency. Therefore, neither vehicular users' private location information nor the navigation result will be disclosed, even in the presence of colluding semi-trusted cloud server (or cryptography service provider, CSP) and unauthorized users. Finally, we prove the security and demonstrate the utility of both MPDC and EPNS.

Automated summary

Real-time navigation is important in various applications, and preserving location privacy is a concern. Existing approaches use pseudonyms or fully homomorphic encryption (FHE), but they have limitations. This paper proposes an efficient multiparty delegated computation (MPDC) and a lightweight privacy-preserving real-time intelligent traffic navigation scheme (EPNS) to address these issues, providing secure evaluation and accurate prediction.