Design optimization-based software-defined networking scheme for detecting and preventing attacks

In this paper, we design a Spider Monkey-based Elman Spike Neural Network (SM-ESNN) to identify intrusion threats in Software Defined Networks (SDN). Utilizing analysis of multidimensional Internet Protocol (IP) flows to find intrusion and flooding assaults against central controllers. Moreover, information is first gathered from the ISCXIDS2012 dataset and updated to the SDN's secure defensive system. The developed software defense system has two sub-modules: a detection module and a mitigation module. The developed technique's key benefit is improving SDN security by quickly and accurately identifying and stopping assaults. First, the proposed SM-ESNN method is implemented in Python. The assessment measures in this scenario include accuracy, specificity, sensitivity, precision, and false alarm rate (FAR). Furthermore, the suggested SM-ESNN approach obtained improved average performances of 98.24% accuracy, 97.34% specificity, 98.68% sensitivity, and 98.33% precision, which highlights its efficiency in detecting the attacks.