Research on the Combination Rule Properties of Belief Function Theory
Belief function theory is an effective information fusion method for uncertain data. Belief function theory mainly includes Dempster-Shafer theory and Dezert-Smarandanche theory. Combination rules play an important role in belief function fusion. The fused result depends on the performance of combination rules. In this paper, properties of Dempster-Shafer theory and Dezert-Smarandanche theory combination rule are presented. The properties of combination rules are analyzed in three aspects as combination property, time series property and engineering property. The simulation results show that the properties presented are useful for analysis, improvement and application of combination rules in belief function theory.
KeywordsBelief function theory Dempster-shafer theory Dezert-smarandanche theory Combination rule Property analysis
Supported by National Science Foundation of China (Grant No. 61102168) Innovation Foundation in Air Force Radar Academy (Grant No. X11QN106).
- 3.Dezert J (2002) Foundations for a new theory of plausible and paradoxical reasoning. Inform Secur J 12(1):26–30Google Scholar
- 4.Dezert J, Smarandache F (2003) On the generation of hyper-power sets for the DSmT. In: Proceedings of the 6th international conference on information fusion. Piscataway, IEEE vol 6, pp 1118–1125Google Scholar
- 5.Zadeh LA (1984) Review of Shafer’s “a mathematical theory of evidence”. AI Mag 5(3):81–83Google Scholar
- 6.Hou Jun (2006) The research on combination rule, performance indexes and application of evidence reasoning. Northwestern Polytechnical UniversityGoogle Scholar
- 7.Pan Quan, Zhang Shan Ying, Cheng Yong Mei, Zhang Hong Cai (2001) Some research on robustness of evidence theory. Acta Automat Sin 27(6):798–805Google Scholar
- 8.Jin Hong bin, Lan Jiang qiao (2011) Proportional conflict redistribution rules based on preponderant belief. Modern Rad 33(9):58–61Google Scholar