Intrusion Detection for Internet of Vehicles using Machine Learning

The Internet of Vehicles (IoV) has replaced vehicular networks as the preferred paradigm as a result of the enormous expansion in computer and network capabilities. Because of the dynamic IoV’s diverse nature necessitates effective resource management, which calls for cutting-edge technologies like Software Defined Networking (SDN), Machine Learning (ML), and others. In Software Defined-IoV (SD-IoV) networks, Road Side Units (RSUs) are in charge of network effectiveness and provide a number of safety features. However, it is not practical to deploy enough RSUs, and the current RSU placement does not provide complete coverage of an area. Furthermore, any lapse in network security or performance has a negative influence on driving. Thus, the objective of this study is to increase security in an IoV network by using different types of Machine learning Algorithm to increase network efficiency. As a result, it is suggested to use XG-BOOST Learning Algorithm placement method to decrease communication time while expanding coverage among IoV devices. Along with this method, this paper is works on the IoV network by using CAN-OITDS Dataset. The comparative study of conventional ML algorithms shows that the IDS detects the malicious attack on IoV with the help of XGBOOST with high accuracy of 96.04%.