Least square fault estimation for a class of sensor networks

In this paper, the problem of fault estimation is considered for a class of time-varying systems through a sensor network. The measurements collected via the sensor network are subject to probabilistic data missing. A set of least square estimators are designed for the addressed systems such that the estimation error variance is minimized at each tim step. By solving a set of Riccati-like matrix equations, the parameters of the desired estimators are calculated recursively. Numerical simulations are exploited to illustrate the effectiveness of the proposed algorithm.