A Novel Approach for Multisensors Target Recognition with Conflicting Information Based on Entropy and Distance

Combining uncertain, ambiguous, and conflicting evidences has always been a sensitive and critical technique in evidence combination theory. Firstly, the Lance distance is employed to characterize the reliability between evidence piece, and Deng entropy is utilized to measure the information volume of uncertain information, a coefficient is constructed to determine the weight coefficients for each evidence. Secondly, a novel approach is proposed to deal with multi-sensors information combination with conflicts, the Murphy's method is employed to obtain BPA of each target. Finally, the effectiveness and reliability of the proposed approach are verified through the numerical experimental. Comparative analysis with existing methods from other literature demonstrates the superiority of our approach in handling highly conflicting evidence.