An Intelligent Traffic Pattern Analysis and Optimization Using Machine Learning in SDN Enable Network Infrastructure

Nowadays a large number of traffic passes through the network. However, the network system is unreliable and has safety issues. Different attacking activities may arise on the network traffic. Again the performance, privacy, latency, and control overhead requirements of real-world networks are the fundamental issues in terms of Network performance analysis. Hence analyzing traffic patterns is a very efficient fact that will help to detect anomalies, monitor network availability, and maximize the performance of the network. Recent innovations in networking, such as the revolutionary Software-Defined Networking (SDN), have disrupted the status of network administration. The SDN has the power to decouple its control plan from the data plan which results in more effective observation in terms of traffic analysis. Therefore, the pattern analysis of network traffic has been accomplished by a method based on machine learning. In this work, Our mission is to analyze the network traffic using machine learning algorithms for SDN to enable network flows to observe the traffic pattern. We collect SDN-based data and prepossessed it with advanced preprocessing tools and techniques. Afterward, Machine Learning models i.e. Random Forest (RF) and KNN are applied to find out the traffic pattern. Our experiments conducted in this study demonstrate that machine learning model-based intelligent pattern analysis for network traffic achieves superior outcomes when compared to traditional methods.