Venkateshwara Gera Rao
Computer Science
March 2023
In order to identify and stop potential attacks, Software Defined Networking (SDN) requires an intrusion detection system (IDS). In this research, we suggest an intrusion detection system (IDS) for SDN that leverages machine learning approaches to increase intrusion detection accuracy. The suggested system combines the Decision Tree (DT) and Random Forest machine learning techniques (RF).The DT and RF algorithms are used by the IDS to monitor and analyse network traffic in real-time. The system constructs a model to identify any aberrant behaviour using the characteristics of typical traffic patterns. The system generates an alert when an intrusion is found to warn the network administrator of the potential danger.The testing findings show how well the suggested IDS works at spotting several kinds of attacks, including DoS, DDoS, and port scanning. With a low false-positive rate of 1.1%, the suggested IDS attain a high detection rate of 98.4%. In addition, the suggested IDS has a low overhead and good accuracy, which makes it a workable solution for protecting SDN settings.To sum up, the suggested IDS for SDN employing machine learning techniques offers a trustworthy and effective method of identifying potential network intrusions. Combining DT and RF algorithms has proven to be useful in raising intrusion detection accuracy. Network administrators can strengthen the security of their SDN environments and fend off potential threats with the aid of the suggested technology.
264- 292
© Airo Journals 2021
