Weighted Combination of Conflicting Evidence Based on Evidence Variance
Yang Yi · Dianzi xuebao · 2011
Dempster-Shafer evidence theory is widely used in the fields of decision-level information fusion.To suppress the counter-intuitive results encountered when using Dempster's rule of combination,a modified sequential weighted evidence combination approach based on variance of evidence is proposed.According to Jousselme's distance,variance of bodies of evidence is defined.In each combination step,the weights are generated based on the variances of the sequences of available evidence combination results before and after adding new arriving body of evidence.Then the weights generated are used to modify the bodies of evidence including the previous combination result and the new arriving body of evidence at current step.Finally,according the Dempster's rule of combination,the weighted average combination result can be obtained.Some numerical examples provided show the efficiency and rationality of the proposed approach.