Improving security in 5G vehicular networks using ASCON and machine learning based NIDS

Improving security in 5G-enabled vehicular networks is critical to address the growing cyber threats targeting embedded systems within the Internet of Vehicles (IoV). The IoV significantly enhances vehicle safety, traffic flow management, collision avoidance, and urban transportation efficiency—forming the backbone of smart city initiatives. However, expanding IoV adoption also amplifies exposure to security vulnerabilities, requiring comprehensive protection mechanisms to defend vehicular communications from unauthorized access and malicious tampering.This research proposes a robust, two-tier security framework tailored for 5G vehicular environments. The first tier employs ASCON, a lightweight cryptographic algorithm, to safeguard data exchanges between vehicles and infrastructure. ASCON effectively mitigates diverse attacks such as password brute-force, authentication compromises, and sophisticated cryptographic exploits. Implemented on embedded platforms like Raspberry Pi and validated with real-world industrial vehicle datasets, it ensures strong data integrity through optimized encryption parameters.The second tier integrates a machine learning-based Network Intrusion Detection System (NIDS) focused on detecting and countering Denial of Service (DoS) attacks. Utilizing a voting classifier ensemble, the NIDS delivers outstanding detection accuracy, precision, and recall rates, strengthening the system’s overall resilience. By combining lightweight cryptography with advanced intrusion detection, this framework offers a comprehensive, scalable security solution that addresses the unique challenges of IoT-enabled automotive systems, advancing safer, more reliable, and eco-friendly intelligent transportation networks.