Scientists Develop AI Tool to Safeguard Vehicles from Cyber Threats

Researchers have developed an artificial intelligence tool to safeguard vehicles from cyber threats, addressing a major challenge in the Internet of Vehicles (IoV) ecosystem. The IoV refers to a network where vehicles communicate with each other and intelligent devices in parking lots, pedestrians, and road infrastructure. However, the security of these vehicles is vulnerable to cyberattacks that can cause catastrophic events.

Scientists from the University of Sharjah, the University of Maryland, and Abdul Wali Khan University Mardan have proposed an ML-based authentication scheme that trains and classifies vehicles at edge servers in a distributed manner, preserving privacy and minimizing bandwidth consumption and delay.

The scheme uses machine learning algorithms to analyze and verify real-time communication patterns, strengthening security against common cyber-attacks. The researchers’ approach extends the decision power of vehicles and edge servers to identify adversaries, ensuring faster and more efficient communication.

Safeguarding Vehicles from Cyber Threats: A Novel AI Tool

The increasing reliance on the Internet of Vehicles (IoV) has brought many benefits, including real-time communication and enhanced operational capabilities. However, this interconnected network of vehicles, roadside units, and infrastructure also poses significant security risks. The limited resources of onboard units and embedded sensors in vehicles make them vulnerable to cyberattacks, which can have catastrophic consequences. To address these challenges, researchers from the University of Sharjah, the University of Maryland, and Abdul Wali Khan University Mardan have developed an artificial intelligence (AI) tool that consolidates the privacy of vehicles and their drivers.

The Need for Lightweight Authentication Schemes

The IoV ecosystem is characterized by vehicle geographical mobility and insufficient resources. The onboard units (OBUs) and roadside units (RSUs) are resource-constrained, making it difficult to support computationally complex security and privacy preservation schemes. Moreover, the communication system installed in vehicles encounters challenges related to bandwidth scarcity and delays in responses from cloud-located services within a stipulated time. These limitations make it essential to design lightweight but reliable authentication schemes to combat various attacks.

The Proposed AI-Based Authentication Mechanism

The researchers propose an ML-based authentication scheme that trains and classifies vehicles at the edge servers in a distributed manner, preserving the privacy of communicating entities and minimizing bandwidth consumption and delay experienced by vehicles. This novel approach extends the decision power of cars and edge servers to identify adversaries, ensuring faster and more efficient communication. The proposed scheme requires each car to participate in an offline phase, where a trusted authority shares a list of masked identities (MaskIDs) and secret keys of legitimate vehicles and edge servers.

Enhancing Security with Machine Learning

The machine learning algorithm embedded in the proposed scheme analyzes and verifies communication patterns in real-time, strengthening security against common cyber-attacks including man-in-the-middle or impersonation attacks. The AI tool is designed to prune against well-known adversarial attacks by embedding a timespan in the payload of each encrypted message. This approach ensures that vehicles can authenticate each other without relying on cloud servers, reducing the computational load on the vehicle.

Simulation Results and Performance Evaluation

The research team conducted experiments in a simulated environment using comparative analysis of the proposed scheme with existing state-of-the-art schemes. The simulation results concluded that the proposed scheme is not only pruned against well-known intruder attacks but also lightweight and effective concerning various performance evaluation metrics such as computation, communication, and storage overheads. The exceptional performance of the proposed scheme in terms of computational overhead, communication overhead, and storage overhead makes it a promising solution for safeguarding vehicles from cyber threats.

Conclusion

The increasing reliance on the IoV ecosystem necessitates the development of robust security measures that can combat emerging cyber threats. The researchers’ novel AI-based authentication mechanism addresses the challenges posed by limited resources and bandwidth scarcity in vehicles, ensuring faster and more efficient communication. This approach can potentially revolutionize how we secure our vehicles, making them safer and more reliable for drivers and passengers alike.