Enhanced Machine Learning Framework for DDoS Attack Endurance in Vehicles Security for 5G Networks

Over the last few years, security has emerged as an issue of great concern in vehicular networks (VANETs). Attacks classified as Distributed Denial of Service (DDoS) and Denial of Service (DoS) are some major security threats. These attacks have the potential to destroy services by bombarding systems with excessive traffic. This makes them a big threat to cyber security as they are able to cause mass destruction. These attacks can degrade the operation of the network by flooding it with too many packets. Such weaknesses threaten the quality and safety of vehicle-to-vehicle commutation systems. This study focuses on improving the security of vehicular networks by addressing threats arising from DoS and DDoS attacks. Introduces a machine learning framework that takes advantage of data from the application layer to enhance threat detection capabilities. This framework effectively extracts and analyzes important network traffic aspects by combining Random Projection (RP) and Randomized Matrix Factorization (RMF). It makes it easier to spot important patterns in complicated data. Through exhaustive evaluation, the framework exhibits outstanding performance using models such as the Extra Tree Classifier (ETC), Logistic Regression (LR), and Random Forest (RF). Together, these models achieve an impressive detection accuracy of 0.99, outperforming conventional methods. By effectively identifying and mitigating malicious activities, the proposed approach significantly fortifies the security and reliability of vehicular communications. This research provides a robust solution to enhance vehicular network protection, ensuring more secure communication in connected vehicles, and establishes a foundation for future progress in safeguarding vehicular systems against cyber threats.