Image: ghcr.io/vs-uulm/veremi-nextgen:latest

VeReMi NextGen

A comprehensive dataset and dataset generator for evaluating Misbehavior Detection in Vehicle-to-Everything (V2X) communication.

Paper Reference

If you are using our dataset, please use the following citation:

Hermann, A., Remmers, J. N., Eissermann, D., Erb, B and Kargl, F. 2026. VeReMi NextGen: A Dataset for Evaluating Misbehavior Detection Systems in VANETs. Proceedings of the 2026 IEEE Vehicular Networking Conference Conference (Montreal, Canada, 2026).

@inproceedings{Hermann2026vereminextgen,
	author = {Hermann, Artur and Remmers, Jan Niklas and Eisermann, Dennis and Erb, Benjamin and Kargl, Frank},
	booktitle = {2026 {IEEE} {Vehicular} {Networking} {Conference} ({VNC})},
	date = {2026-06},
	location = {Montreal, Canada},
	title = {VeReMi {NextGen}: A {Dataset} for {Evaluating} {Misbehavior} {Detection} {Systems} in {VANETs}},
}

Overview

VeReMi NextGen provides:

• Dataset with 15 attack types on V2X Cooperative Awareness Messages
• Train/Val/Test Sets for machine learning experiments
• Plug'n Play Solution for recreating the VeReMi Baseline with the option to change parameters
• Attack Generator to create custom attack scenarios
• CaTCH-MBD System implementing CaTCH adopted to VeReMi NextGen
• Parameter Optimization for systematically obtaining the best thresholds

Repository Structure

VeReMi-NextGen/
├── Dataset/          # Download-Links for each set of VeReMi NextGen
├── Generator/        # Everything to recreate VeReMi NextGen or an own new Dataset
└── Documentation/    # Detailed documentation of VeReMi NextGen 

Documentation

Full Documentation

• Home – Overview, message format, quick start
• Getting Started - Guide to run the simulation on you own machine
• Architecture – System design & Overview
• Processes – Processes that occure during simulation
• Evaluation – Tools used for evluating VeReMi NextGen
• Post-Processing – Post-Processing pipeline for integrating MB and more
• Eclipse MOSAIC - Eclipse MOSAIC Simulation Setup