A Hybrid Approach for Detecting of Intrusion in Vanet Using Machine Learning with Optimization Approach
Year: 2024
Type: article
Abstract: In recent times, there has been a growing focus among researchers on VANET (Vehicular Ad-hoc Network) and its diverse applications, including the improvement of traffic safety through the collection and distribution of traffic event information. Malfunctions in vehicles significantly affect both human safety and road safety, underscoring the importance of addressing vehicle network security as a crucial challenge. The significance of carefully analyzing the Machine Learning (ML) methods used to improve the security aspects of intrusion detection systems (IDSs) is highlighted by this delicate research focus in VANET. This entails dealing with issues like the computational complexity of machine learning difficulties brought on by the increase in vehicle data. In order to better address the issues raised by rapid development, this research presents a hybrid machine learning approach intended to enhance the efficacy of intrusion detection systems (IDSs). This network's main goals are to improve general privacy and thwart vulnerable attacks. Support Vector Machine with Fish Swarm Optimization (SVM-FSO), a cutting-edge machine learning approach, is used in our suggested system to identify DDoS attacks and provide vehicle information while maintaining anonymity. The CICIDS 2017 IDS dataset is used for the evaluation, and MATLAB is used to implement the unique machine learning technique. When performance evaluation takes into account parameters like latency, network lifetime, throughput, delivery ratio, and drop, the results are better than with other approaches like SVM, ANN, KNN, and DNN.
Author: Ganga T G
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7218241679
