Modeling and Simulation of Replay Attack Detection using V2X Message in Autonomous Vehicles in WSN based IoT Environment

Abstract

An autonomous vehicle is a car that drives itself to its destination without driver intervention. Autonomous driving provides driver convenience and prevents accidents caused by driver carelessness. Autonomous vehicles recognize external environments using sensors such as cameras and riders. In addition, autonomous vehicles collect information by using vehicle-to-everything communication in places they do not recognize. During vehicle-to-everything communication, vehicle-to-infrastructure communication communicates with the infrastructure installed on the road and receives information. In other words, the autonomous vehicle receives information from the infrastructure located in an unrecognized place and grasps the road conditions. However, because infrastructure is expensive to install and maintain, technology that uses wireless sensor networks instead of infrastructure has been proposed. Since the sensor node used in the wireless sensor network is placed outside and communicates wirelessly, it is easily compromised from an attacker. Attackers can use a compromised node to attempt various attacks that affect the system, such as replay attacks. These attacks can also have a fatal effect on autonomous vehicles that use information from sensor nodes. The attacker constantly transmits false information to autonomous vehicles, causing a disruption in the driver's schedule. In addition, autonomous vehicles may cause traffic accidents due to path planning using incorrect information. The proposed scheme in this paper uses an autonomous vehicle to defend against replay attacks and detects compromised nodes. The sensor node sends a message to the base station and the autonomous vehicle to notify them when an event occurs. Thereafter, the message is transmitted to the traffic management center and the base station to be mutually verified. This paper shows that by modeling and simulating EF-ITS, it is possible to defend against replay attacks with a probability of 90% and detect compromised nodes.