Evidence filtering

Duminda A. Dewasurendra, Peter H. Bauer, Kamal Premaratne

Research output: Contribution to journalArticlepeer-review

33 Scopus citations

Abstract

A novel framework named evidence filtering for processing information from multiple sensor modalities is presented. This approach is based on conditional belief notions in Dempster-Shafer (DS) evidence theory and enables one to directly process temporally and spatially distributed sensor data and infer on the "frequency" characteristics of various events of interest. The method can accommodate partial and incomplete information from multiple sensor modalities during the process. Certain restrictions on the coefficients impose several challenges in the design of evidence filters suggesting that arbitrary frequency shaping is not possible. A design procedure and the analysis of nonrecursive evidence filters is presented. A threat assessment scenario is simulated and the results are presented to illustrate the applications of evidence filtering.

Original languageEnglish (US)
Pages (from-to)5796-5805
Number of pages10
JournalIEEE Transactions on Signal Processing
Volume55
Issue number12
DOIs
StatePublished - Dec 1 2007

Keywords

  • Dempster-Shafer (DS) belief theory
  • Evidence filters
  • Filter design
  • Inference mechanisms
  • Partial and incomplete information
  • Positive systems
  • Sensor modalities
  • Signature detection
  • Spatiotemporal sensor data
  • Uncertainty

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing

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

Cite this