Robust kriged Kalman filtering

Brian Baingana, Emiliano Dall'Anese, Gonzalo Mateos, Georgios B. Giannakis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Scopus citations

Abstract

Although the kriged Kalman filter (KKF) has well-documented merits for prediction of spatial-temporal processes, its performance degrades in the presence of outliers due to anomalous events, or measurement equipment failures. This paper proposes a robust KKF model that explicitly accounts for presence of measurement outliers. Exploiting outlier sparsity, a novel ℓ1-regularized estimator that jointly predicts the spatial-temporal process at unmonitored locations, while identifying measurement outliers is put forth. Numerical tests are conducted on a synthetic Internet protocol (IP) network, and real transformer load data. Test results corroborate the effectiveness of the novel estimator in joint spatial prediction and outlier identification.

Original languageEnglish (US)
Title of host publicationConference Record of the 49th Asilomar Conference on Signals, Systems and Computers, ACSSC 2015
EditorsMichael B. Matthews
PublisherIEEE Computer Society
Pages1525-1529
Number of pages5
ISBN (Electronic)9781467385763
DOIs
StatePublished - Feb 26 2016
Event49th Asilomar Conference on Signals, Systems and Computers, ACSSC 2015 - Pacific Grove, United States
Duration: Nov 8 2015Nov 11 2015

Publication series

NameConference Record - Asilomar Conference on Signals, Systems and Computers
Volume2016-February
ISSN (Print)1058-6393

Other

Other49th Asilomar Conference on Signals, Systems and Computers, ACSSC 2015
CountryUnited States
CityPacific Grove
Period11/8/1511/11/15

Bibliographical note

Funding Information:
The work of B. Baingana and G. B. Giannakis was supported by NSF grants 1343248, 1423316, 1442686, 1508993, 1509040, and grant ARO W911NF- 15-1-0492

Publisher Copyright:
© 2015 IEEE.

Keywords

  • IP path delay monitoring
  • Kalman filter
  • Robust estimation
  • kriging
  • sparsity

Fingerprint Dive into the research topics of 'Robust kriged Kalman filtering'. Together they form a unique fingerprint.

Cite this