基于混沌理论网络攻击频率预测研究  

Modeling and simulation of Network Attack Frequency Prediction

作　　者：陈萍[1] 张莉华[1] 

机构地区：[1]黄淮学院计算机科学系,河南驻马店463000

出　　处：《计算机仿真》2011年第10期125-128,353,共5页Computer Simulation

基　　金：河南省重点科技攻关项目(112102210335);河南省高等学校青年骨干教师资助计划(070016)

摘　　要：研究入侵网络攻击准确预测问题。网络攻击频率是进行网络入侵检测的一个重要特征,网络攻击频率具有随机性、混沌性和不连续性等特性,传统预测方法不能对随机的攻击频率数据进行有效地分析测试,导致预测精度低。为了提高网络攻击预测精度,提出一种根据混沌理论的网络攻击频率预测模型。首先对网络攻击频率进行相空间重构,然后将重构后的数据输入神经网络中学习并预测,获得网络攻击频率的预测结果。结果表明,解决了传统预测方法不能很好测试网络攻击频率数据特征的难题,提高网络攻击频率预测精度,对降低网络系统的安全风险具有重要参考价值。Study the problem of network security.Because network attack frequency has the properties of randomness,chaotic and uncontinuity,the traditional forecasting methods cannot identify chaos,and prediction accuracy is low.In order to improve the accuracy of network attack frequency,the paper puts forward a network attack frequency prediction method based on chaotic time series forecasting model.Firstly,the maximum Lyapunov index method is used to decide the network attack frequency of chaotic,and then the phase space reconstruction is used to reconstruct the original data.Lastly,the reconstruction data is inputted into the neural network for learning.The prediction results of the attack frequency are obtained.Simulation results show that chaos time prediction method reveals the network attack frequency characteristics,solves the problem that traditional forecasting method can not depict network attack frequency data features,improves the network attack rate prediction accuracy,and has important reference value for reducing network system security risk.

关 键 词：网络攻击频率 相空间重构 混沌时间序列 预测 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]