移动物联网中重复博弈论入侵检测模型仿真  

Simulation of Repeated Game Theory Intrusion Detection Model in Mobile Internet Of Things

作　　者：张淋江 刘志龙 周红雷 刘建朋 ZHANG Lin-jiang;LIU Zhi-long;ZHOU Hong-lei;LIU Jian-peng(Henan University of Animal Husbandry and Economy,Network Management CenterNetwork Management Center,Zhengzhou Henan,450011 China)

机构地区：[1]河南牧业经济学院网络管理中心

出　　处：《计算机仿真》2019年第5期337-340,共4页Computer Simulation

基　　金：河南省科技攻关计划项目课题(142102210607);河南省省级教改一般项目课题(2017SJGLX500);河南省高等学校重点科研项目课题(15A520002);河南省社科联、经团联调研课题课题(SKL-2018-2645)

摘　　要：为了提高移动物联网在运行过程中的安全性和准确性,需要对移动物联网入侵检测模型进行研究。采用当前入侵检测模型对移动物联网进行入侵检测时,存在检测结果准确率低和节点背离率高的问题。提出一种移动物联网中重复博弈论入侵检测模型,通过四元组构建重复博弈模型,在重复博弈模型的基础上得到入侵检测系统和攻击者的收益函数,采用期望效用函数计算重复博弈模型的收益矩阵,根据计算结果得到模型的纳什均衡。采用均衡算法将贴现因子引入效益函数中对重复博弈论入侵检测模型进行优化,并在重复博弈论入侵检测模型中引入惩戒机制,完成移动物联网的入侵检测。仿真结果表明,所提模型的检测结果准确率高、节点背离率低。Research on intrusion detection model can improve accuracy and security of Internet of Things during operational process.Accuracy of detection result of current detection model is low.Meanwhile,deviation rate of node is high.Therefore,a new intrusion detection model based on repeated game theory was proposed in this article.Firstly,repeated game model was built via tetrad.Based on the model,revenue function of intrusion detection system and intruder was obtained.Payoff matrix of the model was calculated via expected utility function and s,Nash equilibrium of the model was obtained according to the calculation result.Then,discount factor was introduced into benefit function to optimize the intrusion detection model using load-balancing algorithm and punishment mechanism was introduced in the intrusion detection model.Thus,the intrusion detection of mobile Internet of Things was completed.Simulation results show that accuracy of detection result of the method is higher and deviation rate of node is lower.

关 键 词：移动物联网 重复博弈论 入侵检测 

分 类 号：TP391.44[自动化与计算机技术—计算机应用技术]