Intelligent Hybrid Machine Learning-Based Cyber Threat Detection and Network Performance Optimization Framework for Vehicular Ad Hoc Networks

Abstract

The advent of Vehicular Ad Hoc Networks (VANETs) has enabled real-time communication between vehicles and the infrastructure on the road and has made it a crucial part of the intelligent transportation system. The special topology, mobility and diverse communication landscape, however, makes VANETs vulnerable to advanced cyber attacks which have the potential to severely impact network reliability, security and communication efficiency. For secure and efficient vehicular communication, this study presents an Intelligent Hybrid Machine Learning-Based Cyber Threat Detection and Network Performance Optimization Framework, which combines Deep feature Extraction, Ensemble learning, Anomaly Detection, and Adaptive network optimization. The framework utilizes extensive data pre-processing, intelligent intrusion detection and multi-class cyber threat classification to accurately detect malicious activity and dynamically optimize routing and communication performance. The proposed framework is an intelligent and adaptive cybersecurity solution that is scalable and usable for improving network performance and the detection capability for threats, making it an ideal solution for next generation intelligent transportation systems.