基于条件证据更新的机动目标分类

Maneuvering Target Classification Based on Conditional Evidence Updating

以Fagin-Halpern Dempster-Shafer条件证据理论为依据,提出了一种基于条件证据更新的机动目标分类方法。该方法将空中目标的机种类型与机动状态相互作为条件证据,以获得交叉验证的条件信任度分配函数,并通过识别框架的拓展,实现基本信任度分配的更新。实验结果验证了所提方法能够有效提高目标分类的准确性和可信度。

According to Fagin-Halpern Dempster-Shafer conditional evidence theory, this paper proposes a classification method of maneuvering target based on conditional evidence updating. In the proposed method, target type and maneuvering status are taken as conditional evidences for each other in order to obtain the assignment function of conditional confidence. Moreover, by expanding the frame of identification, the updating on assignment of basic probability is made. A numerical example is given to show the accuracy and reliability of the proposed approach.