Multi-level diffusion Kalman filter algorithm for adaptive networks

Xiaorui Yang, Wenxuan Han, Lijuan Jia*, Shunshoku Kanae

*Corresponding author for this work

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

Abstract

We study the problem of distributed state estimation over adaptive networks, where agents collaborate to estimate a common state parameter vector. If the sensing target area is too large or we want to improve the convergence speed of a large adaptive network, single-level diffusion algorithms do not have better performance, so we study the multi-level diffusion Kalman filter algorithm where a network running a diffusion strategy is enhanced by defining some special nodes called supernodes. In order to improve the estimation accuracy, we also study the weight-normalized diffusion Kalman filter algorithm. Simulation results show that multi-level diffusion Kalman filter algorithm has better accuracy and convergence performance than single-level diffusion Kalman filter algorithm. Furthermore, in order to further improve the algorithm's performance, we studied weight-normalized methods which are better than average weight methods.

Original languageEnglish
Title of host publicationProceedings of the 36th Chinese Control Conference, CCC 2017
EditorsTao Liu, Qianchuan Zhao
PublisherIEEE Computer Society
Pages8895-8900
Number of pages6
ISBN (Electronic)9789881563934
DOIs
Publication statusPublished - 7 Sept 2017
Event36th Chinese Control Conference, CCC 2017 - Dalian, China
Duration: 26 Jul 201728 Jul 2017

Publication series

NameChinese Control Conference, CCC
ISSN (Print)1934-1768
ISSN (Electronic)2161-2927

Conference

Conference36th Chinese Control Conference, CCC 2017
Country/TerritoryChina
CityDalian
Period26/07/1728/07/17

Keywords

  • Kalman filter
  • adaptive networks
  • diffusion strategy
  • distributed state estimation

Fingerprint

Dive into the research topics of 'Multi-level diffusion Kalman filter algorithm for adaptive networks'. Together they form a unique fingerprint.

Cite this