引用本文:李建瑞,胡松林,岳东,陈小莉,马勇.DoS攻击下随机系统的攻击参数依赖型观测器与控制器协同设计[J].控制理论与应用,2023,40(6):986~994.[点击复制]
LI Jian-rui,HU Song-lin,YUE Dong,CHEN Xiao-li,MA Yong.Attack parameter dependent observer and controller co-design for stochastic systems under DoS attacks[J].Control Theory and Technology,2023,40(6):986~994.[点击复制]
DoS攻击下随机系统的攻击参数依赖型观测器与控制器协同设计
Attack parameter dependent observer and controller co-design for stochastic systems under DoS attacks
摘要点击 2185  全文点击 565  投稿时间:2022-01-04  修订日期:2023-05-25
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2022.20004
  2023,40(6):986-994
中文关键词  DoS攻击  随机系统  均方指数稳定  随机时变李雅普诺夫函数  L2增益
英文关键词  denial-of-service attacks  stochastic system  mean square exponential stability  time-varying stochastic lyapunov function  L2-gain
基金项目  国家重点研发计划项目(2018YFA0702200), 国家自然科学基金项目(62173187, 52022073), 江苏省自然科学基金项目(BK20201377), 江苏省六大 人才高峰项目(RLD201810), 河北省自然科学基金重点项目(E2020203139)
作者单位E-mail
李建瑞 南京邮电大学 571967597@qq.com 
胡松林* 南京邮电大学 songlin621@126.com 
岳东 南京邮电大学 碳中和先进技术研究院  
陈小莉 南京财经大学 信息工程学院  
马勇 武汉理工大学 航运学院  
中文摘要
      本文研究了拒绝服务攻击下网络随机控制系统的攻击参数依赖型观测器与控制器协同设计问题. 首先, 将 拒绝服务(DoS)攻击建模为周期脉宽调制干扰信号, 并构造了Luenberger观测器来估计不可测的系统状态. 其次, 设 计了基于观测器的控制器, 提出一种新的切换随机系统模型. 然后, 引入DoS攻击参数依赖型随机时变李雅普诺夫 函数分析切换随机系统. 给出了切换随机系统均方指数稳定的判据, 并使闭环系统具有L2增益性能水平. 同时通过 应用矩阵不等式技术, 给出了攻击参数依赖型观测器与控制器的协同设计方案. 最后, 以一种飞行器系统为例, 验证 了该方案的有效性.
英文摘要
      The co-design problem of attack parameter-dependent observer and controller is investigated for a class of networked stochastic control systems under periodic denial-of-service (DoS) attacks. First, the DoS attack is modeled as a periodic pulse width modulation (PWM) interference signal, and a Luenberger observer is constructed to estimate the unmeasured system state. Second, an observer-based controller is designed, and a new switched stochastic system model is introduced. Third, the DoS attack parameter-dependent stochastic time-varying Lyapunov function is introduced to analyze the switching stochastic system. The criterion of the mean square exponential stability of the switched stochastic system is given, and the closed-loop system has L2-gain performance. At the same time, by applying the matrix inequality technology, the co-design scheme of attack parameter-dependent observer and controller is presented. Finally, an air vehicle system is taken as an example to show the effectiveness of the obtained theoretical results.