A privacy-preserving route planning scheme for the Internet of Vehicles

Abstract

Internet of the Things (IoT) is being integrated into applications that are continuing to reshape many elements of our daily life. One of the major application areas is the Internet of Vehicles which can enhance existing capabilities, such as efficient vehicle route planning. Such systems usually rely on real-time traffic information that includes the temporal location feed of a vehicle. Despite offering clear advantages (such as overcoming congestion, saving fuel/energy/time, and reducing CO2 emission), privacy concerns emerge due to the use of location data. Motivated by this, a privacy-preserving vehicular location (e.g. positioning) sharing scheme is developed for edge cloud-assisted vehicles. In addition, data utility bounds are theoretically analysed, and vehicle routing efficacy is empirically analysed to evaluate the impact of the proposed scheme. Rather than sharing perturbed location on two-dimensional space, we propose a graph-based differential privacy solution for sharing location. The novelty of this work relies on translating the vehicular geospatial data to the graph-structured data for its higher applicability on the road network, designing a real-time application, and empirical analysis of privacy-efficacy optimality.