Vehicle-to-Vehicle Flooding Datasets using MK5 On-board Unit Devices

The availability of information is a key requirement for the proper functioning of any network. When the availability problem is brought to vehicular networks, it may hinder novel vehicular services and applications and potentially put human lives at risk, as malicious users can send a massive number of spurious packets to disrupt them. Although flooding attacks in vehicular contexts have been the focus of attention of the research community, most proposed datasets are generated using simulated data and only contain the modeled network's behavior. In this work, we generated datasets of such attacks using three realistic vehicular devices, i.e., MK5 On-board Unit (OBU). We applied a machine learning algorithm to get the first insights into the complexity of the proposed datasets, reporting the achieved Accuracy, F1-Score, Precision, and Recall.