Distributed bounded-error state estimation

This paper presents a distributed bounded-error state estimation algorithm adapted, e.g., to wireless sensor networks. Contrary to centralized estimation, where all data are collected to a central processing unit, here, each datum is processed locally by the sensors, the results are broadcasted to the network and taken into account by the other sensors. A first analysis of the conditions under which distributed and centralized estimators provide the same results in a bounded-error context is presented. An application to the tracking of a moving source using a network of sensors measuring the strength of the signal emitted by the source is considered.