Skip to main navigation Skip to search Skip to main content

Robust state estimation based on quadratic programming for network packet dropout

  • Zonggen Yi
  • , Zhongjie Wang

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

3 Scopus citations

Abstract

With the development of network, the wide use of network makes the state sensing more complex, and there inevitably exists the effect of network's uncertainty. In this paper, a new model which considers the state estimation with the network packet dropout is proposed and this is a robust state estimation problem. In this paper a robust state estimation based on quadratic programming(RSE-QP) method is proposed. In this algorithm, with the use of l1 term to describe and handle the packet dropout, there needs some transformations to express it as a standard l1 regularized least squares problem(LSP) for convenient and fast solving. Through solving a l 1 regularized LSP and associating with the Kalman filter updating procedure, the good estimated state is gained effectively even when there exists network packet dropout.

Original languageEnglish
Title of host publication2011 International Conference on Electronics, Communications and Control, ICECC 2011 - Proceedings
Pages1647-1652
Number of pages6
DOIs
StatePublished - 2011
Event2011 International Conference on Electronics, Communications and Control, ICECC 2011 - Ningbo, China
Duration: Sep 9 2011Sep 11 2011

Publication series

Name2011 International Conference on Electronics, Communications and Control, ICECC 2011 - Proceedings

Conference

Conference2011 International Conference on Electronics, Communications and Control, ICECC 2011
Country/TerritoryChina
CityNingbo
Period09/9/1109/11/11

Keywords

  • l-regularized least squares
  • network packet dropout
  • quadratic programming
  • robust state estimation

Fingerprint

Dive into the research topics of 'Robust state estimation based on quadratic programming for network packet dropout'. Together they form a unique fingerprint.

Cite this