Distributed Information Filter for Linear Systems with Colored Measurement Noise

  • Yanbo Yang
  • , Yuemei Qin
  • , Quan Pan
  • , Yanting Yang
  • , Zhi Li

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Scopus citations

Abstract

This paper considers the distributed filtering problem for discrete-time linear systems with colored measurement noise obeying an autoregressive process in sensor networks. For the considered system in the centralized fusion framework, a novel information-type filter is proposed based on the measurement difference approach. Here, the dimension of the estimate error covariance (i.e., the information matrix) in the proposed information filter is the same as that of the original system state, with the help of the block matrix inverse operation. Then, the average consensus-based distributed implementation is designed, to ensure that the final state estimate in each sensor node is asymptotically consistent with the centralized filtering result as closely as possible. An example about target tracking with colored measurement noise in sensor networks validates the proposed method.

Original languageEnglish
Title of host publicationFUSION 2019 - 22nd International Conference on Information Fusion
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9780996452786
StatePublished - Jul 2019
Event22nd International Conference on Information Fusion, FUSION 2019 - Ottawa, Canada
Duration: 2 Jul 20195 Jul 2019

Publication series

NameFUSION 2019 - 22nd International Conference on Information Fusion

Conference

Conference22nd International Conference on Information Fusion, FUSION 2019
Country/TerritoryCanada
CityOttawa
Period2/07/195/07/19

Keywords

  • average consensus
  • block matrix inverse
  • colored measurement noise
  • distributed filtering
  • information filtering

Fingerprint

Dive into the research topics of 'Distributed Information Filter for Linear Systems with Colored Measurement Noise'. Together they form a unique fingerprint.

Cite this