Weighted Data Fusion Based on Evidence Confidence Distance
Jiayi Li, Xin Wang · 2019
Aimed at the problem that D-S evidence theory cannot deal with high conflict, a new weighted combination algorithm based on confidence distance is proposed in this paper. This algorithm introduces the concept of confidence distance measure, in which the variance factor is computed by two evidential variance determination methods. The data source is processed by confidence distance and transformed into distance matrix. A new set of evidence is obtained by weighting the similarity matrix. Finally, the evidence is fused by D-S evidence theory. An example is given to show the effectiveness of the proposed algorithms.