DETECTING FOG-ASSISTED IOVS NETWORK ANOMALIES USING DEEP LEARNING

Abstract

Our review involves Deep Learning to detect anomalies in Fog-Assisted IoVs to further develop security. We recognized IoVs network attacks with 97% accuracy utilizing SVM, Random Forest, Decision Tree, Naive Bayes, DNN, and DNN Autoencoder. We added ensemble draws near, for example, the Voting Classifier, which accomplished 100 percent accuracy. This overhaul safeguards against validation breaks, information trustworthiness concerns, DDoS attacks, and malware dangers by further developing correspondence. Our work fortifies IoVs network security and addresses neighborhood haze hub weaknesses with Fog-Assisted layers to make trustworthy and safe wise transportation frameworks. Our endeavors lead to get information move, upgraded street wellbeing, and reliable astute
transportation administrations for clients, IoV proprietors, and society. Our methodology underlines the significance of refined innovation in making transportation more secure and more effective by diminishing auto collisions and further developing
correspondence.