Predictive malware detection in SDN-enabled wireless networks
K. Muthamil Sudar, P. Nagaraj, V. Vaissnave · 2024
Software-defined networking (SDN) has become an innovative technology in today’s digitally connected world, offering dynamic network management and adaptability. However, because of their inherent flexibility, SDN environments are also susceptible to new types of security risks, such as malware. This chapter investigates how malware cases can be predicted and mitigated within SDN infrastructures using supervised machine learning approaches, particularly random forest and support vector machines (SVMs). To analyze network traffic data gathered from SDN controllers, we make use of random forest and SVM. To extract pertinent data, including flow statistics, packet headers, and communication patterns, feature engineering approaches are used. The experimental findings show that random forest and SVM are effective at correctly identifying malware occurrences in SDN systems. A comparative study exposes the advantages and disadvantages of each model, highlighting their functional properties. We also demonstrate how these supervised machine learning techniques can be tailored to handle the particular issues associated with SDN security.