Machine Learning based Security Enhancement Framework for Software Defined Networks

The significance of adopting stringent security measures to safeguard network infrastructures as Software Defined Networks (SDNs) become prevalent cannot be underestimated. This paper outlines a comprehensive framework for enhancing SDN’s security through machine learning methods. This suggested approach entails capturing Packet Capture (pcap) files, by Wireshark, from an SDN controller and then using IDS rules to analyze them with Snort Intrusion Detection System (IDS). The output is in the form of a log file that helps identify possible risks nodes within the SDN. This Snort IDS log file can later be transformed into structured CSV format via a systematic pipeline integrated into our framework. Furthermore, Python based machine learning model analyzes, classifies and evaluates this CSV file. To improve how malicious activities in SDNs are identified and classified, the machine learning technique aims at suggesting workable strategy for detecting and resolving security issues. An experiment suite validating this suggestion shows accuracy and efficiency of the proposed framework while separating benign from malevolent network activity… And it is provided by means of such metrics as an accuracy level report and a detailed categorization analysis.