Network Situation Assessment Method Based on Improved BP Neural Network

Although a software defined network (SDN) realizes the flexible configuration and centralized control of resources, there are potential security risks challenges. Network situation awareness (NSSA) technology associates integrates multi-source heterogeneous information to analyze impact on whole network, assessment can grasp in real time. However, existing methods have low accuracy, most studies focus traditional networks, while few SDN environment. In this paper, by summarizing important index parameters SDN, model based improved back propagation (BP) neural (based cuckoo search algorithm) is proposed, step factor algorithm (CS) was improve accuracy. The maps elements layers optimizes weights thresholds BP through obtain global optimal solution; it finally purpose comprehensive rating evaluation verified set up Mininet. experimental results show that curve closer value, accuracy rate 97.61%, with good results.