A Discussion of Dempster-Shafer Theory and its Application to Identification Fusion

Abstract

This paper outlines some of the basics of Dempster-Shafer Theory, which is a mathematical theory for combining evidence from different sources to obtain a degree of belief in a proposition. In particular, different combination rules available within the context of Identification Fusion and the assumptions and implications of each of those rules are outlined and investigated. However, the belief function arising from combining evidence under Dempster-Shafer Theory is often insufficient to support decision-making, and a transformation from the belief function to a probability distribution is required. Several different transformations and illustrative examples implementing Dempster-Shafer Theory for Identification Fusion are provided. The results and some possible directions for future work are discussed.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 2015
Accession Number
ADA621365

Entities

People

  • E. El-mahassni
  • K. White

Organizations

  • Defence Science and Technology Group

Tags

Communities of Interest

  • C4I
  • Materials and Manufacturing Processes
  • Sensors

DTIC Thesaurus Topics

  • Automatic Identification Systems
  • Classification
  • Computational Complexity
  • Detection
  • Detectors
  • Identification
  • Identification Systems
  • Models
  • Multitarget Tracking
  • National Security
  • Probability
  • Probability Distributions
  • Radio Frequency
  • Random Variables
  • Security
  • Sensor Fusion
  • Target Detection

Readers

  • Approximation Theory.
  • Theoretical Analysis.