Trajectory tracking and recovery attacks in VANET systems

Abstract

Summary Investigating the mechanisms of various attacks in vehicular ad hoc networks (VANETs) provides fundamental basis to develop safeguard techniques for network security issues. As they develops, attacking strategies are also needed to be established for people to learn how to defend. Previous works failed to consider multiple victims recovery and minimum road side units (RSUs) deployment issues. In this paper, we approach the attacking strategy from an adversarial point of view and develop a trajectory tracking and recovering attack based on matrix completion (TTRA). By randomly sampling locations of vehicles, TTRA is capable of recovering locations of any user with tolerable deviation. After that, the problem of how to minimize the number of RSUs while tracking all vehicles in the whole region is also considered by converting this problem into set covering problem. A heuristic algorithm based on hierarchical clustering is proposed accordingly. To the best of our knowledge, this is the first attempt of recovering mobile users' trajectories by applying these novel techniques. Finally, simulation analysis is used to show the performance of the proposed scheme. Simulation results demonstrate the merits of the scheme in terms of tracking results, location error, and influences of sampling ratio, among others.

Attacks on Self-Driving Cars and Their Countermeasures: A Survey

“…How will the car process detailed information in the case of an attack to enable the driver to make an informed and timely decision? All these questions need to be answered for a breakthrough in the safety of driverless cars against cyberattacks [142].…”