基于动态模态分解的弹道目标平动补偿与微动特征提取方法  

Translational compensation and micro-motion feature extraction method of ballistic targets based on dynamic mode decomposition

作　　者：李开明[1,2] 代肖楠 张袁鹏[4] 姚佳文 罗迎[1,2] LI Kaiming;DAI Xiaonan;ZHANG Yuanpeng;YAO Jiawen;LUO Ying(College of Information and Navigation,Air Force Engineering University,Xi’an 710077,China;Collaborative Innovation Center of Information Sensing and Understanding,Air Force Engineering University,Xi’an 710077,China;Unit 95806 of the PLA,Beijing 100076,China;The First Department,Air Force Early Warning Academy,Wuhan 430019,China;Beijing Research Institute of Telemetry,Beijing 100094,China)

机构地区：[1]空军工程大学信息与导航学院,陕西西安710077 [2]空军工程大学信息感知技术协同创新中心,陕西西安710077 [3]中国人民解放军95806部队,北京100076 [4]空军预警学院一系,湖北武汉430019 [5]北京遥测技术研究所,北京100094

出　　处：《系统工程与电子技术》2025年第2期451-462,共12页Systems Engineering and Electronics

基　　金：国家自然科学基金面上项目(62371468,62131020,62301599,62271500)资助课题。

摘　　要：针对弹道目标平动导致微动特征难以准确提取的问题,提出一种基于动态模态分解(dynamic mode decomposition, DMD)的弹道目标平动补偿与微动特征提取方法。首先,在弹道目标微动回波建模的基础上,对目标的慢时间-距离像序列进行微多普勒(micro-Doppler, m-D)特征曲线分离;其次,将分离后的数据向量移位堆叠构建为增广数据矩阵,并对其进行DMD;然后,利用分解后的模态幅值对各模态进行排序,结合损失函数等信息选取主要模态;同时,利用主要模态中的零频率模态完成弹道目标的平动补偿,从其他主要模态中提取出自旋频率和锥旋频率等微动特征信息;最后,对基于DMD的弹道目标平动补偿与微动特征提取方法进行性能分析与对比实验,验证了所提方法的可行性和稳健性。To address the problem of difficulty in accurately extracting micro-motion features caused by ballistic target translational motion,a ballistic target translational motion compensation and micro-motion feature extraction method based on dynamic mode decomposition(DMD)is proposed.Firstly,based on the modeling of micro-motion echoes of ballistic targets,the slow-time range profile sequence of the target is subjected to micro Doppler(m-D)feature curve separation.Secondly,the separated data vectors are shifted and stacked to construct an augmented data matrix,and DMD is performed.Then,the decomposed mode amplitudes are used to rank each mode,and the main mode is selected combined with information such as the loss function.The zero frequency mode in the main modes is utilized to achieve translational compensation of ballistic targets,and micro-motion feature information such as spin frequency and cone rotation frequency from other main modes are extracted.Finally,performance analysis and comparative experiments are conducted on the DMD-based ballistic target translational compensation and micro-motion feature extraction methods,verifying the feasibility and robustness of the proposed methods.

关 键 词：动态模态分解 弹道目标 微多普勒 平动补偿 特征提取 

分 类 号：TN95[电子电信—信号与信息处理]