A variance-constrained approach to event-triggered distributed extended Kalman filtering with multiple fading measurements

In this paper, a distributed extended Kalman filtering problem is studied for discrete-time nonlinear systems with multiple fading measurements. To alleviate the network communication burden, the event-triggered communication scheme is employed in both sensor-to-estimator channel and estimator-to-estimator channel. As such, the data transmission is executed only when the predefined event occurs. In addition, a set of independent random variables with known statistical properties is defined to represent the phenomenon of multiple fading measurements. The variance-constrained approach is adopted to derive an upper bound for the estimation error covariance in consideration of the event-triggered mechanism and truncated error by linearization. The filter gain for each node is then designed to minimize such an upper bound by recursively solving two Raccati-like difference equations. By virtue of the stochastic stability theory, a sufficient condition is provided to guarantee the boundedness of the estimation error. Finally, a simulation example is presented to illustrate the feasibility and effectiveness of the proposed filtering algorithm.