Improving D-S Evidence Theory for Data Fusion System

Dempster-Shafer evidence theory is an efficient method to process uncertain, incomplete and vague information in data fusion. The combination of conflicting bodies of evidence has been one of the most important issues in Dempster-Shafer evidence theory since the example of counter-intuitive results produced by Dempster’s rule was found. Based on the framework of the Dempster-Shafer evidence theory, this paper provides a novel approach to handling this problem. The proposed method reapportions the mass in the null set among the other subsets that is originally assigned to the null set. Our simulations show that the proposed conflicting evidence combination method can effectively solve thorny problems such as military target identification of multisensor.