基于混沌理论和LSSVM的网络攻击预测算法  

Network Attack Prediction Based On Chaotic T heory and Least Square Support Vector Machine

作　　者：韩国彬[1] 

机构地区：[1]河北青年管理干部学院,石家庄450012

出　　处：《科技通报》2012年第8期140-141,144,共3页Bulletin of Science and Technology

摘　　要：针对网络攻击具有多样性、时变性,传统预测方法预测精度较差的问题,提出一种混沌理论和LSSVM相融合的网络攻击预测算法。利用网络攻击频率时间序列预测模型参数之间的联系,采用粒子群优化算法对模型参数进行组合优化。采用最优参数的预测模型对具体网络攻击频率数据进行仿真测试,并与其它预测算法进行对比。实验结果表明,该方法对网络攻击频率预测精度要高于对比算法,是一种泛化能力好、预测结果可靠的网络攻击预测算法。The traditional method＇s prediction accuracy is poor because network attack has diversity,variability.This paper proposes a network attack prediction algorithm based on chaos theory and LSSVM.Network attack frequency time series are predicted using the relation between model parameters,and the parameters of the model are optimized by particle swarm optimization algorithm.The simulation test is carried out with the optimal parameters on network attack frequency data.The experimental results show that compared with the other prediction methods,the proposed model has improved the prediction accuracy,it has good generalization ability and prediction results are reliable.

关 键 词：网络攻击 混沌理论 预测 最小二乘支持向量机 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]