CAPS: context-aware privacy scheme for VANET safety applications

Abstract

Preserving location privacy in vehicular ad hoc networks (VANET) is an important requirement for public acceptance of this emerging technology. Many privacy schemes concern changing pseudonyms periodically to avoid linking messages. However, the spatiotemporal information contained in beacons makes vehicles traceable and the driver's privacy breached. Therefore, the pseudonym change should be performed in a mix-context to discontinue the spatial and temporal correlation of subsequent beacons. Such mix-context is commonly accomplished by using a silence period or in predetermined locations (e.g., mix-zone). In this paper, we propose a location privacy scheme that lets vehicles decide when to change its pseudonym and enter a silence period and when to exit from it adaptively based on its context. In this scheme, a vehicle monitors the surrounding vehicles and enters silence when it finds one or more neighbors silent. It resumes beaconing with a new pseudonym when its actual state is likely to be mixed with the state of a silent neighbor. We evaluate this scheme against a global multi-target tracking adversary using simulated and realistic vehicle traces and compare it with the random silent period scheme. Furthermore, we evaluate the quality of service of a forward collision warning safety application to ensure its applicability in safety applications. We measure the quality of service by estimating the probability of correctly identifying the fundamental factors of that application using Monte Carlo analysis.