一种新的基本概率函数构建方法及应用

A New Basic Probability Function Conformation Method and Application

针对Dempster-Shafer(D-S)证据推理中基本概率赋值函数的构造问题,基于模糊聚类分析给出了一种新的构建方法。将它应用到雷达目标识别的仿真实验中,并与灰关联法相比较,结果表明该方法切实可行,不仅提高了基本概率赋值的准确性和稳定性,而且利用了数据的结构信息,有效缓解了证据的冲突。

Aiming at the problem of basic probability assignment function conformation to Dempster-Shafer evidence reasoning, this paper presents a new conformation method based on fuzzy cluster analysis, then applies it to simulation experiments of radar target recognition, and compares it to the gray correlation method. The example results show that this method is practical and feasible, not only improve the basic probability assignment accuracy and stabilization, but also makes use of the structure information, alleviate the evidence conflict effectively.