Paper Contents
Abstract
VANET safety applications send cooperative awareness messages (CAM) every now and then to keep vehicles updated about the nearby traffic in real time. The periodicity and spatiotemporal information from these messages enable a global adversary to monitor vehicle motion. Several privacy schemes have been suggested for VANET, but very few schemes take into account how they affect safety applications. Also, each scheme is compared on inconsistent metrics and unrealistic traces of vehicles, making it more challenging to compare the real-world performance of various schemes in the wild. In this paper, we attempt to bridge this gap and compare various privacy schemes not only on the privacy achieved but also on what impact they have on safety applications. A distortion-based measure of privacy is first introduced and compared with other well-known privacy measures demonstrating its strength in quantifying privacy. A real-world safety measure based on Monte Carlo simulation is then suggested to quantify the QoS of two safety services: forward collision warning and lane change warning. Six existing state-of-the-art VANET privacy schemes are compared and analysed using realistic traces of vehicles with respect to the suggested privacy and safety measures. Of the schemes evaluated, it is determined that the coordinated silent period scheme provides the highest level of privacy and QoS but complete synchronization of silence for all vehicles is a practical issue. The CAPS and CADS schemes offer a real-world trade-off between privacy and security as they only use the amount of silence periods necessary in order to avoid tracking and not to change pseudonyms in inconsequential circumstances. Keywords: VANETs - Vehicular Ad hoc Networks, PKC - Public-Key Cryptography, IBS - ID-based signature, IBOOS - ID-based onlineoffline signature, ITS - Intelligent Transportation Systems, CA Certification Authority, TPM - Tamper-Proof Modules.
Copyright
Copyright © 2025 GAGAN M U. This is an open access article distributed under the Creative Commons Attribution License.
