New distributed fusion filtering algorithm based on covariances over sensor networks with random packet dropouts

ABSTRACT

This paper studies the distributed fusion estimation problem from multisensor measured outputs perturbed by correlated noises and uncertainties modelled by random parameter matrices. Each sensor transmits its outputs to a local processor over a packet-erasure channel and, consequently, random losses may occur during transmission. Different white sequences of Bernoulli variables are introduced to model the transmission losses. For the estimation, each lost output is replaced by its estimator based on the information received previously, and only the covariances of the processes involved are used, without requiring the signal evolution model. First, a recursive algorithm for the local least-squares filters is derived by using an innovation approach. Then, the cross-correlation matrices between any two local filters is obtained. Finally, the distributed fusion filter weighted by matrices is obtained from the local filters by applying the least-squares criterion. The performance of the estimators and the influence of both sensor uncertainties and transmission losses on the estimation accuracy are analysed in a numerical example.

KEYWORDS:

• Distributed fusion estimation
• covariance information
• random parameter matrices
• correlated noises
• packet dropouts

Acknowledgements

This research is supported by Ministerio de Economía y Competitividad and Fondo Europeo de Desarrollo Regional FEDER (grant no. MTM2014-52291-P).

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

Ministerio de Economía y Competitividad and Fondo Europeo de Desarrollo Regional FEDER [grant number MTM2014-52291-P].