Distributed simultaneous state-input estimation over sensor networks under quantized communication

This paper is concerned with the distributed state estimation problem over sensor networks with a careful eye towards unknown inputs and quantized communication. Based on singular value decomposition, a unified estimator is developed to simultaneously estimate system states and unknown inputs, in which the estimator gain is determined by minimizing an upper bound on the updated error covariance. Then, a novel distributed state estimator is constructed by enforcing that each node uniformly quantizes the local estimates and the upper bounds on local error covariances before transmission. Furthermore, it is proved that the fused estimation error in each node is uniformly bounded in mean square. Finally, an illustrative example is provided to show the practical effectiveness of the proposed techniques.