Sequential fusion for state estimation in multirate wireless sensor networks

摘要

This paper is concerned with the optimal state fusion estimation for a linear discrete-time stochastic dynamic process with packet losses in the measurement. It is assumed that the dynamical process is a multi-rate scheme and the measurement transmission period equals to the estimates updating period, and the sampling period is shorter than the estimates updating period. The first algorithm is designed by the projection formula combing with the sequential fusion method, and the filter gains are given by a set of stochastic Riccati equations which involve the packet losses variables. While the second algorithm is developed by a different innovation analysis method combining with the sequential fusion formula, and the filter gains are given in terms of a set of generalized deterministic Riccati equations which are just involve the probability of the packet losses. The results of this note can be applied to the dynamic target tracking and locating problem, while the simulation result is shown in the finally illustrative example.