摘要
在医院网络非法入侵检测中,支持向量机的检测泛化性能和参数设定存在较高关联性。为了提升医院网络非法入侵检测率,设计一种基于蚁群优化算法和支持向量机相结合的医院网络非法入侵检测模型,把支持向量机参数设成蚂蚁的方位向量,使用非静止随机提取方法判断目标个体指引蚁群实施全局检索,并在最佳蚂蚁邻域里实施小步长局部检索,获取支持向量机最佳参数,使用最佳参数实现医院网络非法入侵检测。实验结果表明,所设计模型对医院网络非法入侵的误检率最大值仅有1.55%,检测耗时低,且应用效果评价较高。
In the detection of illegal intrusion into hospital network,there is a high correlation between the generalization performance and parameter setting of SVM(support vector machine).A hospital network illegal intrusion detection model based on the combination of ACO(ant colony optimization)algorithm and SVM is designed to improve the detection rate of illegal intrusion into hospital network.The parameters of SVM are set as the position vector of ants,the non static random extraction method is used to judge the target individual and guide the ant colony to implement global search,and the small step local retrieval is implemented in the best ant neighborhood to obtain the best parameters of SVM.The best parameters is used to realize the detection of illegal intrusion into the hospital network.The experimental results show that the maximum false detection rate of the designed model for the detection of illegal intrusion into hospital network is only 1.55%,its detection timeconsuming is low,and its application effect evaluation is good.
作者
吴永芬
徐为
WU Yongfen;XU Wei(College of Command&Control Engineering,Army Engineering University of PLA,Nanjing 210007,China;Department of General Education,Army Engineering University of PLA,Nanjing 210007,China)
出处
《现代电子技术》
北大核心
2020年第22期78-81,共4页
Modern Electronics Technique
基金
国家自然科学基金项目(11401581)。
关键词
医院网络
非法入侵检测
蚁群优化算法
支持向量机
入侵检测模型
全局搜索
hospital network
illegal intrusion detection
ant colony optimization algorithm
support vector machine
intrusion detection model
global search
