As the adoption of electric vehicles (EVs) accelerates, ensuring the security of the EV charging infrastructure has become a critical challenge in the field of cyber-physical system (CPS) security. The EV charging system represents a complex CPS environment where physical and digital components are intricately coupled with intertwined layered communication protocols. These characteristics necessitate a comprehensive analysis of threat flows rather than merely identifying fragmented vulnerabilities. However, conventional threat modeling techniques often rely on static system analyses, providing threat checklists or security requirement specifications, limiting their ability to capture realistic attack paths and stepwise adversarial behaviors. More dynamic threat modeling approaches that incorporate attacker behavior flows are needed to address this limitation. This paper proposes a threat modeling method that analyzes system operation flows and asset interactions to identify critical attack points. This work constructs an attack tree based on this analysis and applies behavior-based threat classification at the leaf-node level. The resulting threat scenarios are linked to the MITRE adversarial tactics, techniques, and common knowledge framework with flows to generate actionable and executable attack scenarios. The proposed method is applied to a core component of the EV charging infrastructure (i.e., EV supply equipment) to derive scenarios, including remote intrusions, physical access attacks, and supply chain compromises. The scenarios are designed to be directly executable in breach and attack simulation environments, enabling automated and repeatable validation of security controls and facilitating proactive defense planning.
Connected and automated vehicles rely on data from various entities to support safety-critical applications such as Cooperative Adaptive Cruise Control (CACC). However, unauthorized data manipulation through, for example, data injection attacks can compromise vehicle safety and lead to incidents. Existing vehicular security mechanisms, such as Misbehavior Detection System (MBD), have limitations in detecting and mitigating all types of threats on their own. To address these limitations, our prior work has proposed the concept of a Trust Assessment Framework (TAF), which assesses data trustworthiness by combining evidence from multiple security systems operating as trust sources. However, TAF as a concept has not been extensively evaluated in safety-critical Cooperative Driving (CD) applications. In this work, we refine the architecture of the TAF and implement a software prototype based on it. We integrate the TAF prototype with a CACC simulation environment and implement three types of data injection attacks. We demonstrate that by incorporating multiple security mechanisms as trust sources, the TAF significantly improves attack detection performance and reduces the number of crashes by 86% compared to using a single security mechanism, such as MBD.
Modern vehicles increasingly depend on wireless technologies such as Remote Keyless Entry (RKE) systems for user convenience. However, these systems are susceptible to various security vulnerabilities. The RollJam attack, for example, demonstrated that rolling code mechanisms, intended to prevent unauthorized access, can be circumvented through wireless interception and replay techniques. In addition, attackers have exploited access to the Controller Area Network (CAN) bus to carry out a range of in-vehicle attacks. In this paper, we introduce RollCAN, a novel attack that adapts the principles of RollJam from the wireless domain to the CAN bus. RollCAN enables adversaries with physical access to intercept, block, and replay RKE messages over the CAN bus, thereby allowing unauthorized locking and unlocking of vehicle doors. We demonstrate the practicality of the RollCAN attack through two distinct implementations: one using professional automotive-grade equipment and another using a low-cost Raspberry Pi-based setup with a CAN interface. Furthermore, we discuss potential mitigation strategies to counteract the RollCAN attack.
Driven by technological progress and, above all, software, our world is becoming ever faster, more flexible and more complex. Vehicles in particular are complex systems due to the number of components, communication, and millions of lines of code. Recent trends such as autonomous driving and artificial intelligence further accelerate these developments. As a result, the attack surface of vehicles is constantly increasing. To proactively address future security risks, we present the Automotive Cybersecurity Trend Radar (ACTR), an approach that applies the Innovation Radar methodology to identify and temporally classify emerging technologies relevant for the automotive cybersecurity sector. By systematically analyzing technology trends, academic research, and industry reports, this radar provides a structured view of upcoming challenges and opportunities. Our research highlights key topics for the future, including the impact of post-quantum cryptography, AI-driven security mechanisms, and confidential computing technologies. By incorporating these insights into early-stage strategic planning, manufacturers and suppliers can improve their preparedness for emerging cybersecurity-relevant technologies as well as benefit from technology developments, allowing strategic investments. The ACTR thus serves as a tool for industry stakeholders to anticipate, prioritize, and address automotive cybersecurity challenges before they become critical.
Modern car insurance companies can dynamically adjust premiums with mobile sensor units that constantly record, analyze, and grade driving behavior. However, attackers can tamper with these systems to cheat the insurance company and reduce their premiums. We developed an attacker model for this threat in conjunction with a risk assessment using the ISO/SAE 21434 Risk Rating Methodology, a widely accepted standard for evaluating and managing risks in automotive systems. To support and showcase our model, we examined the static IT security of the mobile sensor unit DriveWell and its companion Android application HUK Mein Auto. In particular, we utilized digital forensic tools to analyze the stored data in the sensor’s internal storage and the app’s smartphone storage, with a focus on encryption mechanisms for locally stored data and internet communication. During this analysis, we spotted a critical vulnerability that enables attackers to manipulate their insurance scores and potentially reduce premiums by up to 30%. Our model and real-world analysis provide a solid foundation for reducing the risk of such manipulation of insurance premiums.
