Optimal predictor and smoother design for networked systems with communication constraint

This paper investigates the optimal linear predictor and smoother design problem for networked systems with communication constraint. The communication constraint is such a problem that only one network node (which may contain more than one sensors) is allowed to gain access to a shared communication channel. The various network nodes are scheduled to transmit packets according to a specified media accessing control protocol, and the remote estimator perform the estimation task with only partially available measurements. The channel accessing processes of the various spatially distributed network nodes are modeled by Bernoulli processes, and optimal linear predictors and smoothers are designed by using the orthogonal projection principle and the innovation analysis approach. An illustrative example is given to show the effectiveness of the proposed results.