The State Estimation Of Discrete-time Complex Network

Network monitoring and fault diagnosis often require the state information of all nodes in the network, however, the measurement of that is unrealistic. Therefore, in order to get all the state variables and understand network’s behaviors better, we need to utilize the measured nodes’ output to estimate the states of network. The main research work and achievements are as follows:Firstly, various uncertain factors in real networks certainly will exert a negative impact on state estimation of the complex network. Under the circumstances that there are missing measurements and noises in the data transmission, this paper is concerned with the idea of robust guaranteed cost state estimation of discrete-time complex networks, which focuses on reducing the influence of missing measurements and noises on the state estimation. By using Lyapunov stability theory combined with stochastic analysis, an existence condition of the gain matrix and the relevant design criteria, which apply to the robust guaranteed cost state estimation, are proposed in the form of linear matrix inequalities(LMIs). Numerical simulation examples are provided to demonstrate the effectiveness and applicability of the proposed design approach.Secondly, we use complex network with the node output coupling to be the model which conforms to the real engineering networks. Under the circumstances that there are missing measurements and noises in the data transmission, this paper is concerned with the problem of robust guaranteed cost state estimation for discrete-time complex networks with coupling output, which focuses on reducing the influence of missing measurements and noises on the state estimation. By using Lyapunov stability theory combined with stochastic analysis, an existence condition of the gain matrix and the relevant design criteria are proposed in the form of linear matrix inequalities(LMIs), which apply to the robust guaranteed cost state estimation. The estimation error dynamics are exponentially mean-square stable and the robust guaranteed cost performance requirements are satisfied. Numerical simulation examples are provided to demonstrate the effectiveness and applicability of the proposed design approach.Finally, we use the method of pinning control in the state estimation for discrete-time complex network. As is more practical in engineering applications, we only control a small part of the nodes to estimate the states of the whole complex network. By using Lyapunov stability theory, pinning state estimator design criteria are proposed. Numerical simulation examples are provided to demonstrate the effectiveness and applicability of the proposed design approach.