An official website of the United States government
The smart city landscape is rife with opportunities for mobility and economic optimization, but also presents many security concerns spanning the range of components and systems in the smart ecosystem. One key enabler for this ecosystem is smart transportation and transit, which is foundationally built upon connected vehicles. Ensuring vehicular security, while necessary to guarantee passenger and pedestrian safety, is itself challenging due to the broad attack surfaces of modern automotive systems. A single car contains dozens to hundreds of small embedded computing devices known as electronic control units (ECUs) executing 100s of millions of lines of code; the inherent complexity of this tightly-integrated cyber-physical system (CPS) is one of the key problems that frustrates effective security. We describe an approach to help reduce the complexity of security analyses by leveraging unsupervised machine learning to learn clusters of messages passed between ECUs that correlate with changes in the CPS state of a vehicle as it moves through the world. Our approach can help to improve the security of vehicles in a smart city, and can leverage smart city infrastructure to further enrich and refine the quality of the machine learning output.
more »
« less
Towards Reverse Engineering Controller Area Network Messages Using Machine Learning
The automotive Controller Area Network (CAN) allows Electronic Control Units (ECUs) to communicate with each other and control various vehicular functions such as engine and braking control. Consequently CAN and ECUs are high priority targets for hackers. As CAN implementation details are held as proprietary information by vehicle manufacturers, it can be challenging to decode and correlate CAN messages to specific vehicle operations. To understand the precise meanings of CAN messages, reverse engineering techniques that are time-consuming, manually intensive, and require a physical vehicle are typically used. This work aims to address the process of reverse engineering CAN messages for their functionality by creating a machine learning classifier that analyzes messages and determines their relationship to other messages and vehicular functions. Our work examines CAN traffic of different vehicles and standards to show that it can be applied to a wide arrangement of vehicles. The results show that the function of CAN messages can be determined without the need to manually reverse engineer a physical vehicle.
more »
« less
- Award ID(s):
- 1645987
- PAR ID:
- 10198349
- Date Published:
- Journal Name:
- Proceedings of the IEEE World Forum on Internet of Things (WF-IoT)
- Page Range / eLocation ID:
- 1 to 6
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Cyber Physical Systems (CPS) consist of integration of cyber and physical spaces through computing, communication, and control operations. In vehicular CPS, modern vehicles with multiple Electronic Control Units (ECUs) and networking with other vehicles help autonomous driving. Vehicular CPS is vulner-able to multitude of cyber attacks, including false data injection attacks. This paper presents an Asynchronous Federated Learning (AFL) with a Gated Recurrent Unit (GRU) model for identifying False Data Injection (FDI) attacks in a VCPS. The AFL model continuously monitors the network and constructs a digital twin using the data obtained from a VCPS for intrusion detection. The proposed model is evaluated using different evaluation metrics. Numerical results show that the AFL model outperforms other existing models.more » « less
-
Commercial vehicles are a vital component of modern logistics and transportation, forming part of the critical infrastructure and representing safety-critical cyber-physical systems. Contemporary automotive operations are dominated by embedded computing systems that engage through standardized protocols, which constitute the infrastructure of vehicular communication networks. Within the commercial vehicle sector, these systems utilize high-level protocols that operate over the Controller Area Network (CAN) protocol for internal exchanges in medium and heavy-duty vehicles. The Unified Diagnostic Services (UDS) protocol, as described in International Standards Organization (ISO) 14229 (UDS) and ISO 15765 (Diagnostic Communication over CAN), plays a pivotal role by providing vital diagnostic capabilities. This research introduces four specific scenarios that expose deficiencies in the diagnostic protocol standards and how these can be manipulated to initiate attacks on in-vehicle computers within commercial vehicles, circumventing existing security frameworks. In the first three scenarios, we demonstrate three flaws within the ISO 14229 protocol standards. Following this, the fourth and final scenario elucidates a flaw unique to the ISO 15765 protocol standards. For the purpose of demonstration, test setups incorporating actual Electronic Control Units (ECUs) linked to a CAN bus were employed. Further experiments were performed using a fully equipped cab assembly from a 2018 Freightliner Cascadia truck, set up as a testing environment. The experimental outcomes demonstrate how attacks targeting these specific protocols can undermine the integrity of individual ECUs, leading to denial of service. Additionally, within the Freightliner Cascadia configuration, a network architecture typical of contemporary vehicles was observed, featuring a gateway unit that isolates internal ECUs from diagnostic interfaces. Although this gateway is engineered to prevent conventional message injection and spoofing attacks, it permits all diagnostic communications. This selective permeability inadvertently introduces a susceptibility to diagnostic protocol flaws, highlighting an essential area for security improvements within commercial vehicle networks. These insights are vital for engineers and developers tasked with integrating the diagnostic protocols into their network subsystems, underscoring the urgency for improved security provisions.more » « less
-
The Controller Area Network (CAN) protocol used in vehicles today was designed to be fast, reliable, and robust. However, it is inherently insecure due to its lack of any kind of message authentication. Despite this, CAN is still used extensively in the automotive industry for various electronic control units (ECUs) and sensors which perform critical functions such as engine control. This paper presents a novel methodology for in-vehicle security through fingerprinting of ECUs. The proposed research uses the fingerprints injected in the signal due to material imperfections and semiconductor impurities. By extracting features from the physical CAN signal and using them as inputs for a machine learning algorithm, it is possible to determine the sender ECU of a packet. A high classification accuracy of up to 100.0% is possible when every node on the bus has a sufficiently different channel length.more » « less
-
The Controller Area Network (CAN) protocol used in vehicles today was designed to be fast, reliable, and robust. However, it is inherently insecure due to its lack of any kind of message authentication. Despite this, CAN is still used extensively in the automotive industry for various electronic control units (ECUs) and sensors which perform critical functions such as engine control. This paper presents a novel methodology for in-vehicle security through fingerprinting of ECUs. The proposed research uses the fingerprints injected in the signal due to material imperfections and semiconductor impurities. By extracting features from the physical CAN signal and using them as inputs for a machine learning algorithm, it is possible to determine the sender ECU of a packet. A high classification accuracy of up to 100.0% is possible when every node on the bus has a sufficiently different channel length.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Conference Paper:
- https://doi.org/10.1109/WF-IoT48130.2020.9221383
