Centralized Fusion Quadratic Estimators in Multi-Sensor Systems with Correlated Missing Measurements

Abstract

In this paper, the centralized fusion quadratic estimation problem in linear discrete-time stochastic systems with missing measurements coming from multiple sensors is addressed when the Bernoulli variables describing the phenomenon of missing measurements are correlated at instants that differ m sampling times. For this purpose, an appropriate augmented system is defined and the required quadratic estimators of the original state are obtained from the linear estimators of the augmented state. By using an innovation approach, recursive algorithms for the least-squares linear filtering and fixed-point smoothing problems of the augmented system are derived. The performance of the proposed estimators is illustrated by a simulation example where centralized fusion linear and quadratic estimators are compared in terms of their error variances for different missing probabilities and values of m.