基于随机历史集的有反馈MSIF目标识别方法  

RandomHistory Set-based MSIF Target Identification Method with Feedback

作　　者：李山 权文 苏力德 LI Shan;QUAN Wen;SU Lide(Air Traffic Control and Navigation School,Air Force Engineering University,Xi’an 710051,China)

机构地区：[1]空军工程大学空管领航学院,西安710051

出　　处：《火力与指挥控制》2024年第4期130-135,共6页Fire Control & Command Control

基　　金：陕西省自然科学基金资助项目(2021JM-226)。

摘　　要：为提高复杂战场环境下的目标识别准确度,提出了一种分布式MSIF模型,并通过加入反馈信息流,对时域内的序贯信息进行处理。以经典D-S证据理论为基础,针对其不能处理高冲突证据的弊端,采用费雪耶兹洗牌方法,置乱泛传感器矩阵中的信息排序,生成随机历史集,并随着运算次数增加,算法反复迭代。通过改进算法,修正了由于高冲突证据过早出现所带来的识别结果偏差。数值算例表明,该方法既能提高信息利用率,同时可克服信息先入性陷阱的影响,准确度较高,且计算量较小,能有效减轻系统负担,适合MSIF目标识别中的大数据量处理。To improve the accuracy of target identification in complex battlefield environments,a distributed MSIF model is proposed and continuous information in the time domain is processed by incorporating feedback information flow.Based on the classical D-S evidence theory,the Fisher-Yates shuffling method is used to disorder the information ordering in the ubiquitous sensor matrix to generate a random history set for traditional algorithm’s drawback of not being able to process high conflict evidence,the algorithm is iterated with the increase of number of operations.By improving the algorithm,the bias of recognition results due to the premature appearance of high conflict evidence was corrected.The numerical examples show that the method can both improve information utilization and overcome the influence of information precedence trap with high accuracy and low computational complexity,which can effectively reduce the system burden and is suitable for large data volume processing in MSIF target recognition.

关 键 词：目标识别 证据融合 MSIF 随机历史集 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]