Research on Algorithm Based on Secure Computer Network Defense

摘要

In order to solve the problems that the network security defense measures are independent, passive and lagging, and anomaly detection needs an effective training set, a scalable dynamic compound virtual network framework and strategy for active defense is designed and realized in this paper, a classification method based on real network data is also proposed. While PSO-FCM clustering algorithm is used to analyze the data in real network, immune evolutionary algorithm is used to dynamically adjust the number of clusters. Experimental results show that this algorithm has correct cluster and standard of network data, and can extract relatively pure training data from the network data.