A novel combination rule based on the evidence conflicting ntensity
Based on the framework of evidence theory, data fusion aims at obtaining a single Basic Probability Assignment(BPA) fuction by combining several belief functions from distinct information sources. Dempsterrule of combination is the most popular rule, but it is a poor solution for the management of the conflict between various information sources at the normalization step. Even when it faces high conflict evidences, the classical Dempster-Shafers(D-S) evidence theory can involve counter-intuitive results. This paper analyzes the origins of the evidence conflict with the Dempsters rule of combination, and presents the evidence conflicting intensity(ECI) to measure the extent of evidence conflict. Based on ECI, a new evidence combination rule is proposed, which can compose both consistent and conflicting evidences. The results of simulation show the validity of this rule.
Information processing Data Fusion Evidence Theory Conflict Combination Rule
Liu Xinglai
North University of China, Shanxi Taiyuan, 030051, P.R.China
国际会议
北京
英文
2007-08-05（万方平台首次上网日期，不代表论文的发表时间）