基于广义证据理论的识别融合解模糊方法  

Method of Recognition Fusion Ambiguity-resolving Based on Generalized Evidence Theory

作　　者：梁雨竹 张玥 LIANG Yuzhu;ZHANG Yue(The 29th Research Institute of China Electronic Technology Group Corporation,Chengdu 610036,China)

机构地区：[1]中国电子科技集团公司第二十九研究所,成都610036

出　　处：《移动信息》2024年第11期218-222,共5页Mobile Information

摘　　要：在复杂的电磁环境下,不同型号雷达辐射源存在电磁参数混叠的情况,导致传统基于三大参数的识别方法出现了型号错误、节拍间型号跳变的情况。为解决识别结果反复跳变的问题,解识别模糊成为辐射源识别领域的重点研究方向。文中提出了一种能解识别模糊的雷达辐射源的识别方法,采用强约束的广义基本概率指派生成方法计算出多属性的广义基本概率指派值,再利用辨识框架可扩展的D-S证据理论进行证据组合,然后在多个时间周期内进行时域递归的证据组合,并根据组合证据进行结果判决,最后通过仿真实验验证了所提方法的有效性。相较于传统方法,文中提出的识别融合方法可有效解决识别融合中的模糊问题,提高识别的准确率和稳定性。In the complex electromagnetic environment,different types of radar radiation sources have electromagnetic parameter aliasing,which leads to model errors and model jumps between beats in traditional identification methods based on three parameters.In order to solve the problem of repeated jumps in recognition results,solving the identification ambiguity has become the key research direction in the field of radiation source identification.In this paper,a method for identifying fuzzy radar radiation sources is proposed.The generalized basic probability assignment generation method with strong constraints is used to calculate the generalized basic probability assignment values of multiple attributes.Then,the D-S evidence theory with scalable identification framework is used to combine the evidence.Then the time-domain recursive evidence combination is carried out in multiple time periods,and the result judgment is made according to the combined evidence.Finally,the effectiveness of the proposed method is verified by simulation experiments.Compared with traditional methods,the recognition fusion method proposed in this paper can effectively solve the ambiguity problem in recognition fusion and improve the accuracy and stability of recognition.

关 键 词：识别融合 解模糊 D-S证据理论 时域递归 

分 类 号：TP212.9[自动化与计算机技术—检测技术与自动化装置]