Abstract: In order to improve the accuracy of evidence conflict measure and evidence fusion, a local conflict information redistribution algorithm based on evidence ranking fusion is proposed. The evidence conflict is firstly measured based on the evidence distance and the conflict coefficient in the new algorithm, and on this basis, the order of evidence fusion is optimized and the conflict measure algorithm of different focal elements in different pieces of evidence is improved. Further, during the sequential fusion of evidence, the new conflict measure results of the evidence and the focal element are applied to the redistribution of local conflict information. The performances of the new algorithm and the related algorithms are comparatively analyzed in theory and application, and the results show that the evidence fusion effect of the new algorithm is more stable and reliable.

Key words: evidence combination rule, target recognition, conflict measure, fusion