On the Consistency and Confidence of Distributed Dynamic State Estimation in Wireless Sensor Networks

TL;DR

This paper investigates distributed dynamic state estimation in wireless sensor networks, emphasizing the importance of consistency and confidence, and proposes a distributed hybrid information fusion algorithm that guarantees these properties while improving estimation accuracy.

Contribution

It introduces a novel distributed hybrid information fusion algorithm that ensures estimate consistency and enhances confidence, with proven convergence and efficiency improvements.

Findings

The proposed algorithm guarantees estimate consistency.

It improves confidence compared to existing methods.

The algorithm converges under certain conditions.

Abstract

The problem of distributed dynamic state estimation in wireless sensor networks is studied. Two important properties of local estimates, namely, the consistency and confidence, are emphasized. On one hand, the consistency, which means that the approximated error covariance is lower bounded by the true unknown one, has to be guaranteed so that the estimate is not over-confident. On the other hand, since the confidence indicates the accuracy of the estimate, the estimate should be as confident as possible. We first analyze two different information fusion strategies used in the case of information sources with, respectively, uncorrelated errors and unknown but correlated errors. Then a distributed hybrid information fusion algorithm is proposed, where each agent uses the information obtained not only by itself, but also from its neighbors through communication. The proposed algorithm not only guarantees the consistency of the estimates, but also utilizes the available information sources in a more efficient manner and hence improves the confidence. Besides, the proposed algorithm is fully distributed and guarantees convergence with the sufficient condition formulated. The comparisons with existing algorithms are shown.