节点文献
拒绝服务攻击下信息物理系统安全控制问题研究
Research on Security Control of Cyber Physical System under Denial of Service Attack
【作者】 黄鹏;
【作者基本信息】 兰州理工大学 , 控制工程(专业学位), 2022, 硕士
【摘要】 随着信息物理系统的飞速发展,智能化、自动化已经融入生产生活的方方面面,信息物理系统已成为目前自动化领域的研究热点。依赖网络互联来完成信息交互的信息物理系统也将面临许多威胁与挑战。受到网络带宽的限制,各数据节点为了获得网络访问权也将不可避免地造成冲突问题。同时,数据在传输过程中也将受到来自外部的恶意网络攻击,也会对系统造成严重影响。本文研究了存在通信约束和拒绝服务攻击的信息物理系统状态估计与控制策略的设计问题,主要内容如下:首先,针对信息物理系统的安全控制问题,提出拒绝服务攻击下基于事件触发的保性能控制方法。考虑到拒绝服务攻击行为的随机特性,将拒绝服务攻击描述为具有固定丢包数的伯努利随机分布过程;为了降低网络通信负担,采用离散事件触发通信机制对传感器的量测信息进行传输;根据李雅普诺夫稳定性理论和线性矩阵不等式方法给出拒绝服务攻击下系统稳定的充分条件,进而在给定的保性能水平下设计了基于事件触发机制的保性能控制器。其次,研究了拒绝服务攻击下具有随机调度协议的信息物理系统H_∞状态估计问题。考虑到传感器节点数据碰撞问题,利用随机调度协议对传感器节点获得网络访问权的顺序进行调度,将通信网络中由于拒绝服务攻击和信道干扰造成的数据包丢失建模为一个具有能量限制的随机模型;然后借助李雅普诺夫稳定性理论和线性矩阵不等式方法给出拒绝服务攻击下估计误差系统渐近稳定的充分条件,进而设计了具有给定H_∞性能的信息物理系统状态估计器。再次,研究了测量通道和控制通道受到拒绝服务攻击的系统安全控制问题。考虑到通信网络中存在的带宽受限、时延、拒绝服务攻击和信道干扰造成的数据包丢失问题,采用WTOD动态调度策略对传感器节点和控制器节点通信进行调度,将网络中存在的随机时延描述为系统的不确定性,将拒绝服务攻击和信道干扰造成的数据包丢失建立为一个统一随机干扰模型;应用李雅普诺夫稳定理论以及线性矩阵不等式技术得到了增广误差系统渐近稳定的充分条件,进而设计了基于观测器的鲁棒H_∞控制器。最后,对全文主要研究工作做了总结,给出了本文的一些研究成果,并对未来的研究工作进行了梳理与展望。
【Abstract】 With the rapid development of Cyber-Physical Systems(CPS),intelligence and automation have been integrated into all aspects of production and life.CPS has also become a research hotspot in the field of automation.CPS has the advantages of implementation perception,remote dynamic control and high integration.CPS that relies on network interconnection to complete information interaction will also face many threats and challenges.Due to the limitation of network bandwidth,each data node will inevitably cause conflict in order to obtain network access.At the same time,the data will also be attacked by malicious networks attacks from outside in the transmission process,which will also have a serious impact on the system.This thesis studies the design of state estimation and control strategy of CPS with communication constraints and denial of service(DoS)attacks.The main contents are as follows:Firstly,the event-triggered guaranteed cost control strategy is investigated for security control problem of CPS under DoS attacks.Firstly,the effect of DoS attacks is described as a Bernoulli random distribution process with the fixed number of packet losses by considering the characteristic of DOS attacks.Then,in order to reduce the network communication burden,a discrete event-triggered communication mechanism is employed to transmit the measured information of sensors.In what follows,the sufficient conditions of system stability are conducted by utilizing Lyapunov stability theory and linear matrix instability technology,and then a guaranteed cost controller based on event triggering mechanism is designed under a given guaranteed cost level.Secondly,the H_∞state estimation problem of CPS with random scheduling protocol under DoS attacks is studied.Considering the data collision problem of sensor nodes,the order in which sensor nodes obtain network access rights is scheduled by using Stochastic Communication Protocol.The packet loss caused by DoS attack and channel interference in communication network is modeled as a stochastic model with energy limitation;Then,with the help of Lyapunov stability theory and linear matrix inequality method,the sufficient conditions for the asymptotic stability of estimation error system under DoS attack are given,and then the state estimator of CPS with given H_∞performance is designed.Thirdly,the security control problem of measurement channel and control channel subjected to DoS attacks is studied.Considering the problems of packet loss caused by bandwidth limitation,delay,DoS attack and channel interference in communication network,WTOD dynamic scheduling strategy is adopted to schedule sensor nodes and controller nodes.The uncertainty delay in the network is described as the uncertainty of the system and the packet loss caused by DoS attack and channel interference is established as a unified random interference model.A sufficient condition for asymptotic stability of the augmented error system is obtained by using Lyapunov stability theory and linear matrix inequality(LMI)technique,and a robust H_∞controller based on observer is designed.Finally,it summarizes the main research work of the full text,gives some research results of this paper,and combs and prospects the future research work.