Volume 18, No. 5, 2021
A Hybrid Intrusion Detection Model For VANET Using SDN And Growing Hierarchical Self-Organizing Maps
K. Savitha , Dr. C.Chandrasekar
Abstract
VANET is the most promising, fast-growing and modern technology. VANETs enable Vehicleto-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I). The VANET's security is one of the most crucial concerns from a prediction and prevention standpoint. Machine Learning based Intrusion Detection System (IDS) can be used to sense the abnormalities, but it suffers from low detection rates, high false alarm rates, and, further, it requires massive amounts of data for prediction accuracy. This paper recommended a framework for the 5G-based VANET system that combines Software-Defined Networking (SDN) with Growing Hierarchical Self-organizing Maps (GHSOM). The suggested system will be a new mix of SDN and a Growing Hierarchical Selforganizing Maps (GHSOM)-based network solution to improve security in both dimensions, detecting and blocking assaults. The suggested method investigates the effect of GHSOM settings on the detection ratio of various types of attacks on a VANET. Experiment findings demonstrate that our technology can sense malicious network traffic efficiently, prevent and mitigate DDoS assaults, and improve VANET’s security and recovery speed from assaulting traffic. Furthermore, the proposed system has a high accuracy rate. The experimental results illustrate the proposed model's usefulness and efficiency in terms of detection accuracy and other metrics investigated.
Pages: 158-182
Keywords: Software-Defined Networking, VANET, Vehicle-to-Vehicle, DDoS Attacks, Accuracy