Research On The Model Of Network Security Situation Prediction Based On Time Series Analysis

At present, the network has become an indispensable part of information society. Various organizations are dependent on the network to work and do daily management. At the same time, various network security events occur frequently, causing a serious threat. Although most of the organizations were deployed a large number of network security devices to protect the security of the network, because of their different emphasis, these devices just pay attention on their own field. Therefore, the security situation analysis technology that can comprehensively analyze network information has become a hot research topic in the field of network security.This paper researches the current methods of security situation analysis and prediction. According to the specific network environment, the paper designs the situation analysis prediction model. And the function of network security situation analysis is added to the existing security management system according to the model. The situation value that calculated by the system is used to do the prediction analysis. The SAS software is used to assist the prediction analysis so as to improve the efficiency and accuracy.First of all, the paper introduced the background and the current situation of the research in network security situation analysis technology field. It also introduces the related technology and research of situation analysis, including data pretreatment, event correlation, situation assessment and situation prediction etc.Secondly, the paper designs a network security situation prediction model according to the actual network environment, discusses the entire process of generating and forcasting network security situation value from data collection and normalization, correlation analysis, risk calculation situation prediction etc. The situation value is divided into four parts including threat index, vulnerability index, index of the host control ability and index of the equipment running ability. The classification calculation is in order to ensure the comprehensiveness of the model. Considering the distinction to the hierarchical deployment of the platform, the value is divided into local risk and regional risk. Then, According to the steps of stationary test, model identification, parameter calculation, residual test, this paper describes the process of network security situation prediction based on time series analysis.Finally, the situation analysis module implementation and page functions are described. The paper introduces the design of the database table and backstage function, introduces the related page configuration method and the various charts. By generating the simulated events it tests and verifies the security situation prediction model.