A belief logarithmic similarity measure based on Dempster-Shafer theory and its application in multi-source data fusion

Dempster-Shafer theory (DST) has attracted widespread attention in many domains owing to its powerful advantages in managing uncertain and imprecise information. Nevertheless, counterintuitive results may be generated once Dempster's rule faces highly conflicting pieces of evidence. In order to handle this flaw, a new belief logarithmic similarity measure (epsilon BLSM) based on DST is proposed in this paper. Moreover, we further present an enhanced belief logarithmic similarity measure (epsilon BLSM) to consider the internal discrepancy of subsets. In parallel, we prove that EBLSMsatisfies several desirable properties, like bounded, symmetry and non-degeneracy. Finally, a new multi-source data fusion method based on EBLSMis well devised. Through its best performance in two application cases, specifically those pertaining to fault diagnosis and target recognition respectively, the rationality and effectiveness of the proposed method is sufficiently displayed.

Automated summary

DST theory is popular due to its advantages in managing uncertain information, but it may produce counterintuitive results when faced with conflicting evidence. To address this flaw, a new belief logarithmic similarity measure (εBLSM) based on DST is proposed. An enhanced belief logarithmic similarity measure (εBLSM) is also presented to consider subset discrepancies. Moreover, a new multi-source data fusion method based on εBLSM is devised, which demonstrates rationality and effectiveness in fault diagnosis and target recognition cases.