A novel information fusion method based on Dempster-Shafer evidence theory for conflict resolution

Evidence conflict that may cause the counter-intuitive results is one of the most concerns for information fusion by Dempster-Shafer's (D-S) evidence theory. To deal with the issue and manage evidence conflict greatly for the improvement of belief convergence, evidence conflict and belief convergence are investigated based on the analysis of the coherence degree between two sources of evidence. Moreover, the stochastic interpretation for basic probability assignment (BPAs) is illustrated. In addition, a few methods in dealing with evidence conflict are analyzed and compared. Then, a new paradox combination algorithm based on an absolute difference factor of two pieces of evidence and a relative difference factor of two pieces of evidence for a specific hypothesis are proposed with the consideration of local attributions to local conflict. The newly proposed algorithm is verified by the numerical example. The analysis shows the efficiency of the proposed method to improve the performance of belief convergence, in which the comparison studies indicate the advantages of the proposed method as well.