鲁棒自适应的机载外辐射源雷达多目标跟踪算法  

Robust adaptive multi-target tracking algorithm for airborne passive bistatic radar

作　　者：单靖原 卢雨 凌寒羽 SHAN Jingyuan;LU Yu;LING Hanyu(International School,Beijing University of Posts and Telecommunications,Beijing 100876,China;Naval Aviation University,Yantai 264001,China;Unit 91917 of the PLA,Beijing 102401,China)

机构地区：[1]北京邮电大学国际学院,北京100876 [2]海军航空大学,山东烟台264001 [3]中国人民解放军91917部队,北京102401

出　　处：《系统工程与电子技术》2024年第9期2902-2915,共14页Systems Engineering and Electronics

基　　金：国家自然科学基金青年科学基金(62001503)资助课题。

摘　　要：针对未知杂波环境下机载外辐射源雷达的多目标跟踪问题,提出一种鲁棒自适应的标签多伯努利滤波器。首先基于标签多伯努利滤波器算法框架对多目标跟踪问题进行建模,然后针对目标新生参数、杂波参数以及目标检测概率未知的问题,提出采用量测驱动的目标新生模型和基于势均衡多目标多伯努利估计器的在线参数估计方法,最后考虑到机载外辐射源雷达量测的非线性,采用序贯蒙特卡罗方法对所提算法进行实现。实验结果表明,所提滤波器能够利用外辐射源量测准确估计多目标航迹,且在未知杂波环境下的性能可以逼近杂波参数已知的广义标签多伯努利滤波器,鲁棒性更好。In order to address the multi-target tracking problem of airborne passive bisstatic radar(APBR)in an unknown clutter environment,a robust adaptive labelled multi-Bernoulli(RA-LMB)filter is proposed.Firstly,a model for multi-target tracking problem is established on the basis of the LMB filter algorithm framework.Then,for the problems of unknown target newborn parameters,clutter parameters and target detection probability,the measurement-driven target newborn model and the online parameter estimation method based on the cardinality-balanced multi-target multi-Bernoulli estimator are proposed.Finally,considering the non-linearity of APBR measurements,the sequential Monte Carlo method is used to implement the proposed algorithm.The experimental results show that the proposed filter is able to estimate the multi-target trajectory using the APBR measurements,and the performance in the unknown clutter environment can be approximated to the generalized LMB filter with known clutter parameters,with better robustness.

关 键 词：外辐射源雷达 多目标跟踪 鲁棒跟踪 标签多伯努利滤波器 随机有限集 

分 类 号：TN957.51[电子电信—信号与信息处理]