Efficient fuzzy ranking with ensemble machine learning network for attack detection and classification in VANET

Vehicular Ad Hoc Networks (VANETs) are an essential element of the recent. Intelligent Transportation Systems (ITS), enabling as the interface among vehicles, roadside transportation, and various services. These networks offer enhanced safety, traffic management, and infotainment , but their openness and reliance on wireless communication make them vulnerable to various security attacks and threats. As VANETs become more widespread, ensuring their security is paramount to their success and adoption. One of the critical challenges in securing VANETs is the detection and classification of attacks that can disrupt the network’s functionality, compromise safety, or degrade performance. To overcome the abovementioned issues, a new attack detection model is developed for VANET. Through the support of online sources, the network data are obtained. The traffic data undergoes the data cleaning process and then, the cleaned data is considered to perform the normalization for organizing the data. Then, Modernized Random Parameter-based Green Anaconda Optimization (MRP-GAO) algorithm is developed for performing the Optimal Weighted Feature Selection (OWFS). To detect VANET attacks, the Ensemble Machine Learning Model (EMLM) is established, where the Bayesian network , AdaBoost , Support Vector Machine (SVM), and Multi-Layer Perception (MLP) are incorporated. Followed by this, the fuzzy ranking method is considered for determining the final attack detection outcome in VANET model. The efficiency analysis on the proposed model is done to verify the detection performance when compared to other comparative techniques.