TY - GEN
T1 - Simutack - An Attack Simulation Framework for Connected and Autonomous Vehicles
AU - Finkenzeller, Andreas
AU - Mathur, Anshu
AU - Lauinger, Jan
AU - Hamad, Mohammad
AU - Steinhorst, Sebastian
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - With the ongoing efforts toward autonomous driving, modern vehicles become increasingly digital and smart. Hence, the vehicle architecture including smart sensors, ECUs, and in-vehicle communication also faces new challenges to satisfy the ever-changing safety and security requirements. The complexity of the system naturally exposes many attack surfaces that demand for sound security solutions to protect the vehicle from potential intrusions. State-of-the-art approaches such as intrusion detection and intrusion response systems require lots of training and testing against various attack scenarios. However, implementing such attacks in real environments is difficult, expensive, and involves many legal and safety considerations. With Simutack, we present an open-source attack simulation framework that is capable of generating realistic attack scenarios for comprehensive security testing in the automotive development process. The framework integrates several classes of attacks, for instance, smart sensor attacks, V2X attacks, and attacks targeting the in-vehicle networks, which are all among the most commonly exploited attack vectors. We evaluate three common attack scenarios that showcase the applicability and capabilities of our work. In each scenario, the generated attack data is processed and returned to the simulation to visualize the attack's effect on the vehicle and its environment. Furthermore, a custom autopilot application demonstrates the attack's impact on autonomous driving systems.
AB - With the ongoing efforts toward autonomous driving, modern vehicles become increasingly digital and smart. Hence, the vehicle architecture including smart sensors, ECUs, and in-vehicle communication also faces new challenges to satisfy the ever-changing safety and security requirements. The complexity of the system naturally exposes many attack surfaces that demand for sound security solutions to protect the vehicle from potential intrusions. State-of-the-art approaches such as intrusion detection and intrusion response systems require lots of training and testing against various attack scenarios. However, implementing such attacks in real environments is difficult, expensive, and involves many legal and safety considerations. With Simutack, we present an open-source attack simulation framework that is capable of generating realistic attack scenarios for comprehensive security testing in the automotive development process. The framework integrates several classes of attacks, for instance, smart sensor attacks, V2X attacks, and attacks targeting the in-vehicle networks, which are all among the most commonly exploited attack vectors. We evaluate three common attack scenarios that showcase the applicability and capabilities of our work. In each scenario, the generated attack data is processed and returned to the simulation to visualize the attack's effect on the vehicle and its environment. Furthermore, a custom autopilot application demonstrates the attack's impact on autonomous driving systems.
KW - Attack Generation
KW - Connected and Autonomous Vehicle Security
KW - Security Framework
KW - Simulation
KW - V2X
UR - http://www.scopus.com/inward/record.url?scp=85169785292&partnerID=8YFLogxK
U2 - 10.1109/VTC2023-Spring57618.2023.10200555
DO - 10.1109/VTC2023-Spring57618.2023.10200555
M3 - Conference contribution
AN - SCOPUS:85169785292
T3 - IEEE Vehicular Technology Conference
BT - 2023 IEEE 97th Vehicular Technology Conference, VTC 2023-Spring - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 97th IEEE Vehicular Technology Conference, VTC 2023-Spring
Y2 - 20 June 2023 through 23 June 2023
ER -