A Geo-Indistinguishable Context-Based Mix Strategy for Trajectory Protection in VANETs

Abstract

Vehicular Ad-Hoc Networks (VANETs), the backbone of Intelligent Transportation Systems (ITS), can significantly improve traffic efficiency and road safety. However, preserving privacy during data transmission and storage is a significant issue in vehicular communications. To prevent malicious adversaries from tracking vehicles from the beacons they broadcast, pseudonym-changing or obfuscation strategies have been proposed in literature. However, it is challenging to protect the privacy of vehicles while simultaneously maintaining the quality of service of location-based services. In this article, we present a new context-based mix strategy that provides a higher privacy level and reduces information loss during the pseudonym-changing process. A global passive adversary model is developed to evaluate its performance and the effectiveness of the proposed mechanism is demonstrated. Finally, a uniform framework to evaluate different strategies for vehicular privacy is developed and used to compare their strengths and weaknesses in different scenarios.