State Estimation Of Complex Networks Based On Not Fully Measured

Network monitoring and troubleshooting often require real-time control of all node state information in the network.The direct measurement of the network node status information is unrealistic due to the complexity of the network size and the more nodes.In order to obtain the status information of each node in the network and then to understand the behavior characteristics of the network,we need to estimate the whole network state information through the output information of the measured node.In addition,the reality of the network will produce the delay and packet loss inevitable during the procedure of the transmission due to the various factors.So the study of incomplete measurement of complex network state estimation is more realistic guidance.The main research work and achievements of this paper are as follows:(1)The reality of the network in the transmission process will produce the packet loss,delay and other uncertain factors of interference,the emergence of these disturbances will certainly influence the state estimation of the complex network.In this paper,based on the existing theory,a robust guaranteed state estimator is designed to reduce the influence about the state estimation due to the data packet loss and delay.Based on Lyapunov stability theory and stochastic analysis method,the design criterion of robust guaranteed state estimator is given in the form of linear matrix inequality.Finally,the validity of the design scheme is verified by numerical simulation of complex networks of different sizes.(2)The reality of the network may appear the phenomenon about the part of nodes output unpredictable,while the network delay in the output at the moment k more dependent on the output of the moment (k-1).In this paper,the method of pinning control is applied to the state estimation of discrete complex networks with one step induction delay.By applying feedback control to some nodes to achieve the state estimation of the whole complex network,the cost of state estimation is greatly reduced,which is more in line with the practical engineering application.Based on Lyapunov stability theory,the design criterion of the control state estimator is introduced.Finally,numerical simulation is used to verify the feasibility of the results.(3)According to the structural control theory,the driving node is selected to pinning control.Based on Lyapunov stability theory,the design criterion of gain feedback controller is obtained.Finally,theeffectiveness and feasibility of the designed feedback controller are verified by simulation.