开放实验室内部网络非法入侵准确检测仿真  

Accurate Detection and Simulation of Illegal Intrusion in Open Laboratory Internal Network

作　　者：吴延慧[1] 杨凌凤[1] WU Yan-hui;YANG Ling-feng(Informatization Center,Nantong University,Jiangsu Nantong 226019,China)

机构地区：[1]南通大学信息化中心,江苏南通226019

出　　处：《计算机仿真》2023年第9期411-415,共5页Computer Simulation

摘　　要：为了确保开放实验室网络的安全运行,降低入侵风险检测误检率,提出一种开放实验室内部网络非法入侵检测方法。引入RBF神经网络构建开放实验室内部网络非法入侵检测模型,提取隐含层权值参数,对隐含层权值优化处理;采用自适应蛙跳算法对非法入侵检测模型求解,在经典蛙跳算法的基础上引入变异操作,并通过层次分析方法调整不同影响参数权重,避免陷入局部最优,实现入侵检测。仿真结果表明,所提方法可以有效提升网络非法入侵检测率,降低误检率,获取更加精准的检测结果。In order to ensure the safe operation of the open laboratory network and reduce the false detection rate,a method of detecting illegal intrusion in the internal network of the open laboratory was proposed.Firstly,RBF neural network was introduced to build a model for detecting illegal intrusion in the internal network of the open laboratory,thus extracting and optimizing the hidden layer weights.Secondly,the adaptive leapfrog algorithm was adopted to solve the model.Based on the classical leapfrog algorithm,the mutation operation was introduced.Meanwhile,the weights of different influence parameters were adjusted by the analytic hierarchy process,thus avoiding falling into the local optimum.Finally,intrusion detection was achieved.Simulation results show that the proposed method can effectively improve the detection rate of illegal intrusion,reduce the false detection rate,and obtain more accurate detection results.

关 键 词：开放实验室 内部网络 非法入侵检测 神经网络 自适应蛙跳算法 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]