Recursive LMMSE centralized fusion with recombination of multi-radar measurements

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

6 Scopus citations

Abstract

For target tracking with radar measurements, recursive LMMSE (Linear Minimum Mean Squared Error) filtering outperforms the popular measurement conversion based Kalman filters, which have some serious drawbacks in terms of both estimation accuracy and credibility. The existing recursive LMMSE with measurements from a single radar is first extended to the multi-radar case. It is then shown that recombination plays an important role in performance improvement for recursive LMMSE centralized fusion using multiple radars. Here, "recombination" means shuffling all scalar measurements from the multiple radars, dimension by dimension. This differs from the case of centralized fusion with linear measurements from multiple sensors. Numerical simulation examples are provided to illustrate the use of recombination in recursive LMMSE centralized fusion for the nonlinear radar measurements.

Original languageEnglish
Title of host publicationFusion 2011 - 14th International Conference on Information Fusion
StatePublished - 2011
Event14th International Conference on Information Fusion, Fusion 2011 - Chicago, IL, United States
Duration: 5 Jul 20118 Jul 2011

Publication series

NameFusion 2011 - 14th International Conference on Information Fusion

Conference

Conference14th International Conference on Information Fusion, Fusion 2011
Country/TerritoryUnited States
CityChicago, IL
Period5/07/118/07/11

Keywords

  • Centralized fusion
  • Nonlinear filtering
  • Radar measurements
  • Recombination
  • Recursive LMMSE filtering
  • Target tracking

Fingerprint

Dive into the research topics of 'Recursive LMMSE centralized fusion with recombination of multi-radar measurements'. Together they form a unique fingerprint.

Cite this