Recursive fusion estimation for stochastic discrete time-varying complex networks under stochastic communication protocol: The state-saturated case

Abstract

In this paper, we investigate the recursive fusion estimation problem for time-varying state-saturated complex networks under stochastic communication protocol (SCP). To cater for physical limitations of network components, the phenomenon of state saturations is taken into account in the complex network model. The underlying communication mechanism is to ensure that just one sensor node is permitted to send its collected measurement at each time, and the SCP determines the permission to use the network channel for each sensor at each transmission time. A key issue of the addressed problem is to construct a time-varying state estimator such that an upper bound is guaranteed on the filtering error covariance subjected to both the state saturations and the SCP. By applying two sets of matrix difference equations, we first derive an upper bound according to the error covariance of the state estimation and then minimize such an upper bound by precisely calculating the estimator parameters. Then, the performance analysis of the obtained state estimator is given in terms of the boundedness. Finally, we provide a simulation example to illustrate the validity of the designed state estimator.