Software Defined Network (SDN) Based Intrusion Detection Technique to Enhance Network Performance Using Machine Learning

Abstract

The internet has brought about a significant transformation in the world. It serves as a tool for individuals to maintain their social connections and reach out to others within their networks for support. However, the act of sharing personal and professional information online exposes individuals and organizations to various risks. Given the integral role of the internet in our daily lives, the security of our data is constantly under threat. Consequently, the role of Intrusion Detection Systems (IDS) in safeguarding internet users against malicious network attacks is paramount. An Intrusion Detection System (IDS) is a security mechanism that continuously monitors network traffic for any suspicious activities and promptly alerts users upon detecting such anomalies. This paper aims to explore three distinct classifications, focusing on machine learning algorithms such as Artificial Neural Network (ANN), Support Vector Machine (SVM), and K-Nearest Neighbors (KNN). These algorithms will be employed to determine the most accurate approach using the SDN Intrusion Dataset in the initial phase. Subsequently, based on the outcomes of the first stage, the most effective algorithm will be applied to process our database. The goal of this work is to detect suspicious attacks on the SDN network by using supervised machine learning methods. A flood attack is an attack that has a huge impact on the performance of the network because it can prevent the victim from reacting. Response, preventing him from providing services and wasting network resources. To maintain the performance of the network, these attacks must be detected. So we proposed a supervising machine learning model. The model SVM, KNN, and ANN received performance evaluations of 93.78, 96.56, and 93.63, respectively. Based on the result KNN model yielded the best accuracy results, KNN classifier strategies achieved the desired results.