Investigations For Networked Systems With Limited Resources And Stochastic Cyber-attacks

This paper is concerned with the problem of stability of networked control systems(NCSs),filter design and distributed state estimators design subject to the limited networked resources and stochastic cyber-attacks.Due to the important meaning of theory and application in NCSs,the study of constraint resource and networked security have attracted an increasing interests among experts and scholars.To make fully utilize of the network,a hybrid-triggered scheme,which consists of time-triggered scheme and event-triggered scheme,is introduced.The switching rule between the two schemes is described by a Bernoulli variable.By considering the effect of time-delay,packet loss,external interference and stochastic cyber-attacks,the corresponding mathematical model is constructed.By using Lyapunov stability theory and linear matrix inequality(LMI)techniques,sufficient conditions which can guarantee the stability of desired system can be obtained and the designed algorithm can be presented in detail.The main structure is list as follows:The research background and significance of this paper are described in Section 1.The main contents,lemmas and notations of this paper are also given which are important in the main results.Section 2 concentrates on the quantized stabilization for T-S fuzzy systems with hybrid-triggered scheme and stochastic cyber-attacks.A hybrid-triggered scheme is adopted to alleviate the burden of network,which is described by Bernoulli variable.By considering the influence of network-induced delay,packet loss and cyber-attacks,a mathematical model of NCSs with quantization is constructed by using T-S fuzzy approach.Sufficient conditions which can ensure the stability of closed-loop system are derived and controller gains are presented explicitly with Lyapunov stability theory and LMI techniques.Finally,the effectiveness of the designed algorithm is demonstrated by examples.Section 3 is concerned with the problem of H-infinity filtering for NCSs with hybrid-triggered scheme and stochastic cyber-attacks.In order to mitigate the burden of network,a hybrid-triggered scheme is introduced.Meanwhile,stochastic cyberattacks are considered which is governed by the other independent Bernoulli variable to construct the model of filtering error system.By applying Lyapunov stability theory,one can get the sufficient conditions to guarantee the stability of system.The parameters of designed filter can be obtained with LMI techniques.The feasibility of the proposed algorithm can be illustrated in the simulation section.Section 4 addresses the issue of distributed event-triggered H-infinity state estimators subject to deception attacks for networked sensor systems.Whether current sampling data of each sensor is delivered or not is determined by the corresponding event-triggered scheme,which can reduce the transmission data effectively.A novel model is established by considering the influence of time-delay,packet dropouts and deception attacks.Sufficient conditions can be achieved to stabilize the estimating error system by Lyapunov stability theory.Decoupling gains and coupling gains are expressed by using LMI techniques.Finally,simulations are presented to illustrate the usefulness of the desired algorithm.Section 5 makes a conclusion briefly and points out the disadvantages of this research.Meanwhile,some investigating directions are expected in the future study.