Scientists have developed a new artificial intelligence (AI) tool to improve the security and privacy of vehicles connected through the Internet of Vehicles (IoV).
This innovation aims to tackle challenges like cyberattacks and resource limitations, which threaten the safety of these high-tech vehicles.
What is the Internet of Vehicles?
The IoV is a network where vehicles communicate with each other, road infrastructure, and devices like parking systems and pedestrian tools.
This technology provides real-time updates, making cities smarter and driving safer. However, this system also opens doors to cyber threats, as hackers can intercept or alter vehicle communications.
Modern vehicles in the IoV are equipped with onboard units (OBUs) and sensors that collect data and send it to roadside units (RSUs) or cloud servers.
Machine learning (ML) and deep learning (DL) algorithms process this data in real-time, enhancing the vehicles’ capabilities.
Despite these advancements, the communication systems still face issues such as delays, limited bandwidth, and unreliable cloud servers. These problems can lead to dangerous situations on the road.
Key challenges in vehicle security
According to scientists, existing onboard systems and sensors lack the resources to handle complex security measures. Cloud servers, even when aided by advanced AI algorithms, cannot always provide the quick responses vehicles need. Additionally, these systems face high computational demands, creating vulnerabilities that hackers could exploit.
To solve these issues, researchers from the University of Sharjah (UAE), the University of Maryland (USA), and Abdul Wali Khan University (Pakistan) have designed an AI-based authentication scheme. This system aims to protect vehicle communication, improve privacy, and reduce delays without overloading the onboard systems.
How does the AI tool work?
The proposed system uses machine learning at edge servers—devices located near vehicles—to identify legitimate vehicles and detect potential threats. Here’s how it works:
- Offline Phase: Each vehicle goes through an initial setup where a trusted authority provides a unique set of masked identities (MaskIDs) and secret keys. These identifiers allow vehicles and servers to recognize each other securely without depending on cloud servers.
- Real-Time Authentication: When a vehicle sends a message, the nearest edge server uses its MaskIDs and secret keys to verify its identity. This reduces the computational burden on the vehicle and speeds up communication.
- Advanced Security Features: The system embeds a time-based identifier in every encrypted message to prevent cyberattacks like impersonation or man-in-the-middle attacks. It also uses machine learning to analyze communication patterns, ensuring that only legitimate vehicles can connect to the network.
The researchers tested their system in a simulated environment and found it outperforms existing methods. The key advantages include:
- Lightweight and Efficient: The system is designed to work with limited resources, making it suitable for vehicles with minimal computational power.
- Fast and Reliable: It reduces delays and minimizes bandwidth usage, ensuring smooth communication even in resource-constrained environments.
- Secure Against Cyberattacks: The use of ML algorithms and time-based encryption significantly strengthens the protection against common cyber threats.
This new system could revolutionize the way vehicles interact within the IoV. It ensures that communication is both secure and efficient, paving the way for safer autonomous driving, smart transportation systems, and connected cities.
The scientists believe their innovation addresses critical privacy and security issues, helping vehicles operate more reliably in real-world environments. By reducing dependency on cloud servers, their solution could make IoV networks faster and safer, marking a significant step forward for the future of transportation technology.
