2018/04/18 by Rens W. van der Heijden, van der Heijden, Rens W., Thomas Lukaseder +3 · 1 citation
Computer Science · Engineering · #Adversarial Robustness in Machine Learning #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #User Authentication and Security Systems #Vehicular Ad Hoc Networks (VANETs)
paper · pdf · doi:10.48550/arxiv.1804.06701
openalex publication_date 2018/04/18 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28
Vehicular networks are networks of communicating vehicles, a major enabling\ntechnology for future cooperative and autonomous driving technologies. The most\nimportant messages in these networks are broadcast-authenticated periodic\none-hop beacons, used for safety and traffic efficiency applications such as\ncollision avoidance and traffic jam detection. However, broadcast authenticity\nis not sufficient to guarantee message correctness. The goal of misbehavior\ndetection is to analyze application data and knowledge about physical processes\nin these cyber-physical systems to detect incorrect messages, enabling local\nrevocation of vehicles transmitting malicious messages. Comparative studies\nbetween detection mechanisms are rare due to the lack of a reference dataset.\nWe take the first steps to address this challenge by introducing the Vehicular\nReference Misbehavior Dataset (VeReMi) and a discussion of valid metrics for\nsuch an assessment. VeReMi is the first public extensible dataset, allowing\nanyone to reproduce the generation process, as well as contribute attacks and\nuse the data to compare new detection mechanisms against existing ones. The\nresult of our analysis shows that the acceptance range threshold and the simple\nspeed check are complementary mechanisms that detect different attacks. This\nsupports the intuitive notion that fusion can lead to better results with data,\nand we suggest that future work should focus on effective fusion with VeReMi as\nan evaluation baseline.\n