The TBM global distance measure for the association of uncertain combat ID declarations

Abstract

The problem is sequential association of combat ID declarations in the multi-target environment. Being random and non-specific, the combat ID declarations are represented by belief functions and manipulated using the tools of the belief function theory as interpreted by the transferable belief model (TBM). The solution is provided in the framework of ''object to ID declaration'' association based on assignment techniques. For that purpose, the paper derives the global cost of assignment (i.e. a dissimilarity measure) based on the plausibility of the global assignment. This measure is directly related to the conflict as described in the TBM. The performance of the proposed method was evaluated by Monte Carlo simulations and a comparison with various alternative dissimilarity measures is carried out in the framework of multi-object classification.