An Online Adaptive Random Forest (OARF) for SDN-Managed Real-Time IoT Intrusion Detection-Resource Usage Analysis

Abstract

The Internet of Things and Software Defined Networking are being used together more and more. This makes it easier to manage networks. However it also makes it easier for hackers to get in. This paper is about the Online Adaptive Random Forest framework. The Online Adaptive Random Forest framework is used for detecting intrusions in time in networks that use Software Defined Networking to manage the Internet of Things. The Online Adaptive Random Forest model uses a kind of learning to update its decision trees all the time based on what is happening on the network. This helps the Online Adaptive Random Forest model stay safe from attacks. The Online Adaptive Random Forest framework is part of the Software Defined Networking control plane. This means the Online Adaptive Random Forest framework can quickly change the flow rules to stop threats. Tests show that the Online Adaptive Random Forest framework is better than methods, at detecting intrusions. The Online Adaptive Random Forest framework is also more accurate. Uses less of the computers resources. The results show that the Online Adaptive Random Forest framework is a way to keep the Internet of Things and Software Defined Networking safe. The Online Adaptive Random Forest framework is an option because it is easy to use and can handle a lot of work.