一种基于众信度函数的相关证据合成方法

Combination of non-distinct bodies of evidence based on commonality function

实际应用中,Dempster规则要求的证据独立性可能难以满足.在相关源证据已知的假设条件下,基于证据的众信度函数,提出一种相关证据合成方法.该方法无需辨识独立证据的过程,可直接得到解析的合成结果,并且对相关源证据的形式没有要求.最后通过算例验证了所提出方法的有效性.

The distinctness assumption of the Dempster＇s rule of combination is often not fulfilled in practice. A combination rule for non-distinct bodies of the evidence is proposed, by using the commonality function under the assumption that the dependent source is available. The identification of distinct bodies of the evidence is not necessary in the proposed approach, and an analytic combination result is generated without additional assumption that the dependent source is non-dogmatic. A numerical example is provided to illustrate the effectiveness of the proposed combination rule.