虚拟机间性能互扰下的web网络入侵检测  

Web Network Intrusion Detection under Performance Mutual Interference among Virtual Machines

作　　者：汪刚[1] 

机构地区：[1]江苏大学计算机科学与通信工程学院,江苏镇江212013

出　　处：《计算机仿真》2015年第8期323-326,342,共5页Computer Simulation

摘　　要：在虚拟机间性能互扰的状态下,web网络的开放性和共享性导致虚拟机间信号互扰,如采用传统算法进行web网络入侵检测,其衡量标准无法确定,存在检测时间较长,稳定性差、精确度低、实用性小等问题。提出改进支持向量机算法的web网络入侵检测方法。上述方法将web网络的入侵检测精度作为衡量标准,融合粒子群算法(PSO),通过提取web网络入侵特征子集来获到SVM参数,并将结果作为约束标准来组建虚拟机间性能互扰下的web网络入侵检测数学模型,借助粒子群的强大搜索能力对建立的web网络入侵检测数学模型进行优化,得到全局最优解。仿真结果证明,改进支持向量机算法在虚拟机间性能互扰的状态下,web网络入侵检测方法的精确度高,实用性强。Under the state of performance mutual interference among virtual machines, the openness and sharing of web network lead to signal mutual interference among virtual machines. In the paper, a web network intrusion de- tection method was proposed based on improved support vector machine （SVM） algorithm. In this method, the preci- sion of web network intrusion detection is taken as a measurement criterion with convergence of particle swarm optimi- zation algorithm （PSO）. By extracting the character subset of web network intrusion, the parameters of SVM were obtained, and the results were taken as a standard constraints to build the mathematical model of web network intru- sion detection under performance mutual interference among virtual machines. In virtue of the powerful search ability of particle swarm, the established web network intrusion detection model was optimized to obtain the global optimal solution. Simulation experiment proves that under the state of performance mutual interference among virtual ma- chines, the web network intrusion detection method based on improved support vector machine （SVM） algorithm has high precision and strong practicability.

关 键 词：网络 入侵检测 粒子群算法 

分 类 号：TP127[自动化与计算机技术—控制理论与控制工程]