A Theory of Informational Exchanges - Random Set Formalism

Abstract

This paper describes a theory of data fusion in random-set formalism. Data fusion problems are defined as problems for estimating random sets of targets, i.e., an unknown number of objects whose states are to be estimated, based on information given in terms of random sets, i.e., a collection of sets of unknown numbers of observables with unknown origins. In this theory, information, i.e., a state of knowledge, is described, both a priori and a posteriori, in terms of random-set probability density functions, sometimes known as Janossy densities. Using this formalism, this paper considers an abstract distributed information processing system consisting of multiple information processing agents that, in addition to processing local information obtained through local information gathering sources, exchanges information with each other to achieve a globally optimal informational state collectively.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1998
Accession Number
ADA399141

Entities

People

  • Shozo Mori

Organizations

  • RTX

Tags

Communities of Interest

  • Sensors

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Data Fusion
  • Data Processing
  • Data Sets
  • Detectors
  • Distributed Data Processing
  • Information Exchange
  • Information Processing
  • Information Science
  • Mathematics
  • Multitarget Tracking
  • Probability
  • Probability Density Functions
  • Sensor Networks
  • Statistics
  • Target Tracking

Fields of Study

  • Computer science

Readers

  • Database Systems and Applications
  • Statistical inference.