GPS Location Spoofing Attack Detection for Enhancing the Security of Autonomous Vehicles
Mohsin Kamal, Arnab Barua, Christian Vitale, Christos Laoudias, Georgios Ellinas
Abstract
Attacks on the GPS receiver of Connected and Autonomous Vehicles (CAV) and specifically GPS location spoofing is of great concern for the automotive industry as the attacker can compromise the security of CAVs leading to serious repercussions for the drivers and pedestrians. Attack detection solutions based on specialized hardware (e.g., antenna arrays) and satellite signal processing techniques are accurate, yet bulky and expensive to mount on CAVs. Thus, lightweight and cost-effective solutions for detecting location spoofing attacks are highly desirable. This work presents an in-vehicle attack detection solution that fuses multi-source data readily available from the CAV's onboard sensors. It can be implemented in software running on cheap embedded computing platforms integrated into the CAV. The proposed solution is validated using the real-time CARLA simulator, while extensive experimental results demonstrate its effectiveness under different attack scenarios.