基于禁忌算法和RBF神经网络的网络安全态势预测  

Prediction of Network Security Situation Based on Compound Tabu Algorithm and RBF Neural Network

作　　者：邵伯乐[1] SHAO Bo-le(Bozhou Vocational and Technical College,Bozhou 236800,Chin)

机构地区：[1]亳州职业技术学院实验实训中心,安徽亳州236800

出　　处：《兰州工业学院学报》2018年第3期54-57,共4页Journal of Lanzhou Institute of Technology

基　　金：安徽省教育厅高职教育创新发展行动计划项目(13343XM-6)

摘　　要：针对网络安全态势预测的现有方法预测精度不高和实时性不强,难以从整体上关注网络安全状况的问题,提出了一种基于禁忌算法和RBF神经网络的在线网络安全态势预测方法.首先,建立了基于RBF神经网络的网络安全态势模型;其次,对网络的结构、适应度函数和算法的个体等进行了定义,并描述了采用禁忌算法对这些参数进行优化的具体方法;最后,定义了基于禁忌算法和RBF神经网络的网络安全态势的具体算法.为了验证所提方法的优越性,以某校园网网络管理部门获取的部分黑客攻击数据作为仿真数据,并将所提模型与其他方法进行比较,仿真结果表明文中方法能有效地进行预测,并具有较高的预测准确率.The traditional network security situation method has the problems such as the low accuracy of the prediction and unpunctuality. Therefore,aiming at the overall security situation of the network,a network security situation method based on Tabu algorithm and RBF neural network is proposed. Firstly,the network security situation model based on RBF neural network is established. Then the network structure,fitness function and the solution are defined. The specific algorithm based on Tabu algorithm is used to optimize the parameters. Finally,the algorithm based on Tabu algorithm and RBF neural network for predicting the network security situation is defined. To verify the priority of the proposed model,the hike data fetched from the management department are regarded as the simulation data,and this model is compared with the other methods. The simulation shows this method can predict effectively with the high prediction accuracy.

关 键 词：径向基函数神经网络 网络安全态势 预测 训练误差 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]