基于综合冲突度量的高冲突证据融合算法

为提高冲突证据的合成效果，针对已有基于改进Jousselme证据距离函数的冲突度量算法进行研究，提出基于非包含度和改进Jousselme证据距离函数共同度量证据冲突的综合冲突度量算法，并依此计算证据权重；在此基础上基于焦元被其他证据中同一焦元的共同支持度对焦元可信度计算式进行改进，并将上述改进结果应用于优化两两证据合成顺序的局部冲突信息再分配算法中。将新算法和已有相关算法进行算例与仿真实验比较分析，结果表明，在证据之间存在高冲突情况下，所提新算法的正确目标识别概率高于已有其他证据融合算法，且对于一般证据，新算法也具有较满意的证据融合效果，新算法是一种具有较高灵活性和较强抗干扰性的证据融合算法。

A fusion algorithm for high conflicting evidence based on comprehensive conflict measure

In order to enhance the fusion effect of conflicting evidence, the existing conflict measure algorithm which improves Jousselme evidence distance function is studied, and a comprehensive conflict measure algorithm based on both non-inclusive degree and improved Jousselme evidence distance function is proposed. And then the weight of each piece of evidence is calculated. On this basis, the calculation formula of the credibility of the focal element is improved by using the total support degree of the same focal element from different pieces of evidence, and the calculation result is applied to the local conflict information redistribution algorithm which optimizes the two-two evidence fusion sequence. The new algorithm is compared with the related algorithms by using examples and simulation experiments. The results show that the correct target recognition probability of the new algorithm proposed in this paper is higher than that of other related evidence fusion algorithms in the case of high conflict evidence, and the new algorithm also has satisfactory evidence fusion effect for the general evidence, and it is a kind of evidential fusion algorithm with high flexibility and strong anti-interference ability.