EPRIVO: An Enhanced PRIvacy-preserVing Opportunistic Routing Protocol for Vehicular Delay-Tolerant Networks

Abstract

This paper proposes an enhanced PRIvacy preserVing opportunistic routing protocol (ePRIVO) for vehicular delay-tolerant networks (VDTN). ePRIVO models a VDTN as a time-varying neighboring graph where edges correspond to neighboring relationship between pairs of vehicles. It addresses the problem of vehicles taking routing decision meanwhile keeping their information private, i.e., vehicles compute their similarity and/or compare their routing metrics in a private manner using the Paillier homomorphic encryption scheme. The effectiveness of ePRIVO is supported through extensive simulations with synthetic mobility models and a real mobility trace. Simulation results show that ePRIVO presents on average very low cryptographic costs in most scenarios. Additionally, ePRIVO presents on average gains of approximately 29% and 238% in terms of delivery ratio for the real and synthetic scenarios considered compared to other privacy-preserving routing protocols.