DocumentCode :
2002914
Title :
Application of Dempster-Shafer theory of evidence to the correlation problem
Author :
Morelli, Michael ; DeSimone, Anthony J., Jr.
Author_Institution :
Naval Electron. & Surveillance Syst. Surface Syst., Lockheed Martin, Moorestown, NJ, USA
Volume :
2
fYear :
2002
fDate :
8-11 July 2002
Firstpage :
759
Abstract :
We apply the Dempster Shafer theory of evidence to the spatial correlation problem. Usually the correlation problem is solved using a Bayesian approach by evaluating the likelihood function for each possible assignment and choosing the maximum likelihood function as the correct assignment. A simulation comparison is then made between the decisions arrived at using the Dempster Shafer theory and those found using the (traditional) Bayesian approach. The results show (for the cases examined) that the decisions made by both theories are identical. Since the Dempster Shafer theory is much more computationally intensive than the Bayesian approach and there is no gain (or loss) in the outcome, one should be much more appreciative of the more traditional Bayesian solution to the problem.
Keywords :
Bayes methods; inference mechanisms; maximum likelihood estimation; sensor fusion; uncertainty handling; Bayesian approach; Dempster Shafer theory of evidence; correct assignment; decisions; maximum likelihood function; simulation; spatial correlation problem; Bayesian methods; Computational modeling; Contracts; Covariance matrix; Government; Measurement errors; Model driven engineering; Surveillance;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information Fusion, 2002. Proceedings of the Fifth International Conference on
Conference_Location :
Annapolis, MD, USA
Print_ISBN :
0-9721844-1-4
Type :
conf
DOI :
10.1109/ICIF.2002.1020882
Filename :
1020882
Link To Document :
بازگشت