The State Estimation And Synchronization Of Discrete-time Complex Dynamical Networks With Random Packet Losses

In the transmission process of real network, data losses often happen because of a variety of factors, such as the smaller width and malicious attacks. Data losses contain two kinds of cases, including packet losses and missing measurements. Packet losses refer to data losses that happen between the internal nodes of networks during the data transmission, yet missing measurements refer to data losses that happen between two or more networks during the data transmission. So, in order to understand the behavioral characteristics of a complex dynamical network better, we study the state estimation and synchronization of discrete-time complex dynamical networks with random packet losses and missing measurements. The main contents and contributions of the dissertation are as follows:Firstly, the state estimation problem is addressed for a class of discrete-time complex networks with single packet losses and single missing measurements, where the packet losses probability and missing measurements probability are known. The process of packet losses of states of complex dynamical networks and missing measurements are described with the independent identically distributed Bernoulli random variable, respectively. By applying the Lyapunov stability theory combined with stochastic analysis method, a sufficient condition for state estimation is obtained in the form of linear matrix inequalities(LMIs). It is demonstrated by simulations that the proposed state estimation can estimate the networks states effectively.Secondly, synchronization design is studied for a class of discrete-time complex networks with single packet losses, where the packet losses probability is known. The process of packet losses of states of complex dynamical networks and missing measurements are described with the independent identically distributed Bernoulli random variable, respectively. By applying the Lyapunov stability theory combined with stochastic analysis method, a sufficient condition of synchronization is obtained in the form of linear matrix inequalities(LMIs). Finally,it is demonstrated by simulations that the proposed synchronization method is effective.Thirdly, synchronization design is addressed for a class of discrete-time complex networks with single packet losses and single missing measurements. First, we obtain all the state variables of complex dynamical networks by the method of state estimator, then, the synchronization design is studied by applying the state variables got from the state estimator. The process of packet losses of states of complex dynamical networks and missing measurements are described with the independent identically distributed Bernoulli random variable, respectively. By applying the Lyapunov stability theory combined with stochastic analysis method, a sufficient condition of synchronization is obtained in the form of linear matrix inequalities(LMIs). Finally,it is demonstrated by simulations that the proposed synchronization method with the observed states can realize the network synchronization effectively.