基于高分辨谱估计的雷达目标分类方法

doi:10.12305/j.issn.1001-506X.2025.07.08

摘要/Abstract

摘要：

针对常规窄带雷达体制下利用高分辨谱估计的方法对人、车、无人机等地面及低空运动目标的分类问题，首先通过高分辨谱估计的方法提取回波信号频谱的能量分布特征。然后，对回波信号时域和频域特性进行分析, 提取目标的速度、雷达截面积(radar cross section, RCS)、时域波形熵、幅度相对值、频域方差等特征, 并根据目标实际运动情况划分速度区间, 设计相应的反向传播(back-propagation, BP)神经网络模型。最后，基于实测雷达回波数据验证了所提高分辨谱估计的目标分类算法具有较好的分类效果。

关键词: 高分辨谱估计, 能量分布特征, 速度区间, 反向传播神经网络

Abstract:

To address the classification problem of ground and low-altitude moving targets such as people, vehicles, and unmanned aerial vehicles using high-resolution spectrum estimation in conventional narrowband radar system, the energy distribution characteristics of the echo signal spectrum are extracted through high-resolution spectral estimation firstly. Then, the time domain and frequency domain characteristics of the echo signal are analyzed, and the target speed, radar cross section (RCS), time domain waveform entropy, amplitude relative value, and frequency domain variance are extracted. According to the actual motion of the target, the velocity interval is divided, and a corresponding back-propagation (BP) neural network model is designed. Finally, the experiment based on the measured radar echo data proves that the target classification algorithm with high-resolution spectrum estimation has a significant classification effect.

Key words: high-resolution spectrum estimation, energy distribution characteristic, velocity interval, back-propagation (BP) neural network

中图分类号:

TN959.1

张梦泽, 田黎育. 基于高分辨谱估计的雷达目标分类方法[J]. 系统工程与电子技术, 2025, 47(7): 2146-2153.

Mengze ZHANG, Liyu TIAN. Radar target classification method based on high resolution spectral estimation[J]. Systems Engineering and Electronics, 2025, 47(7): 2146-2153.

图/表 12

图1

图2

图3

图4

图5

图6

表1

表2

表3

表4

表5

表6

参考文献 32

1	LIUZ,CAIY F,WANGH,et al.Robust target recognition and tracking of self-driving cars with radar and camera information fusion under severe weather conditions[J].IEEE Trans.on Intelligent Transportation Systems,2021,23(7):6640-6653.
2	PERSICOA R,CLEMENTEC,GAGLIONED,et al.On model, algorithms, and experiment for microDoppler-based recognition of ballistic targets[J].IEEE Trans. on Aerospace and Electronic Systems,2017,53(3):1088-1108.
3	ZHANGQ,LUOY,CHENY A.Micro Doppler characteristics of radar targets[M].Cambridge:Academic Press,2017.
4	张群,胡健,罗迎.微动目标雷达特征提取、成像与识别研究进展[J].雷达学报,2018,7(5):531-547.
	ZHANGQ,HUJ,LUOY.Research progresses in radar feature extraction, imaging, and recognition of target with micro-motions[J].Journal of Radars,2018,7(5):531-547.
5	CHEN V C. Analysis of radar micro-Doppler signature with time-frequency transform[C]//Proc. of the IEEE Workshop on Statistica Signaland Array Processing, 2000: 463-466.
6	CHENV C,LIF,HOS S,et al.Micro-Doppler effect in radar: phenomenon, model, and simulation study[J].IEEE Trans.on Aerospace and Electronic Systems,2006,42(1):2-21.
7	DUL,LIL S,WANGB S.Micro-Doppler feature extraction based on time-frequency spectrogram for ground moving targets classification with low resolution radar[J].IEEE Sensors Journal,2016,16(10):3756-3763.
8	ZHU L Z, ZHAO H C, XU H L. Classification of ground vehicles based on micro-Doppler effect and singular value decomposition[C]//Proc. of the IEEE Radar Conference, 2019.
9	LI Y B, DU L, LIU H W. Moving vehicle classification based on micro-Doppler signature[C]//Proc. of the IEEE International Conference on Signal Processing, Communications and Computing, 2011.
10	吴嗣亮.近代信号处理[M].北京:北京理工大学出版社,2007.
	WUS L.Modern signal processing[M].Beijing:Beijing Institute of Technology Press,2007.
11	HU H, WU H B, CHEN X B. Spectrum estimation of AR model based on Burg method[C]//Proc. of the International Conference on Optoelectronic Materials and Devices, 2023.
12	SHAO S J, ZHU W G, LI Y G. Radar detection of low-slow-small UAVs in complex environments[C]//Proc. of the IEEE 10th Joint International Information Technology and Artificial Intelligence Conference, 2022: 1153-1157.
13	ZHUL Z,ZHANGS N,MAQ,et al.Classification of UAV-to-ground targets based on enhanced micro-Doppler features extracted via PCA and compressed sensing[J].IEEE Sensors Journal,2020,20(23):14360-14368.
14	DAUBECHIESI.The wavelet transform, time frequency localization and signal analysis[J].IEEE Trans.on Information Theory,1990,36(5):961-1005.
15	张龙,王兴理,张琪.基于小波变换和距离引导值的雷达杂波抑制分析[J].现代雷达,2020,42(7):23-28.
	ZHANGL,WANGX L,ZHANGQ.Analysis of radar clutter suppression based on wavelet transform and distance guidance value[J].Modern Radar,2020,42(7):23-28.
16	RIOULO,DUHAMELP.Fast algorithms for discrete and continuous wavelet transforms[J].IEEE Trans.on Information Theory,1992,38(2):569-586.
17	GUO D F, ZHU W H, GAO Z M, et al. A study of wavelet thresholding denoising[C]//Proc. of the 5th International Conference on Signal Processing Proceedings, 2000: 329-332.
18	ALLABAKASHS,YASODHAP,REDDYS V.Wavelet transform-based methods for removal of ground clutter and denoising the radar wind profiler data[J].IET Signal Processing,2015,9(5):440-448.
19	刘金星. 现代谱估计技术的研究应用与FPGA实现[D]. 成都: 电子科技大学, 2016.
	LIU J X. Research and application of modern spectrum estimation technology and realization of FPGA[D]. Chengdu: University of Electronic Science and Technology of China, 2016.
20	FANY F,LUOF,LIM.Fractal properties of autoregressive spectrum and its application on weak target detection in sea clutter background[J].IET Radar, Sonar and Navigation,2015,9(8):1070-1077.
21	POOJA G, KAR S P. MUSIC and improved MUSIC algorithm to estimate direction of arrival[C]//Proc. of the International Conference on Communications and Signal Processing, 2015: 757-761.
22	XIY P,LIUJ,ZHAOW H.SATCOM earth station arrays anti-jamming based on MVDR algorithm[J].Applied Sciences,2023,13(14):8337.
23	孙伟峰,李雯,范陈清,等.基于AR模型的多普勒散射计回波功率谱估计[J].遥感信息,2023,38(2):40-47.
	SUNW F,LIW,FANC Q,et al.An echo power spectral density estimation method based on AR model for Doppler scat terometter[J].Remote Sensing Information,2023,38(2):40-47.
24	KAYS M.Modern spectral estimation: theory and application[M].Englewood Cliffs, NJ:Prentice Hall,1994.
25	王潇怡,罗运华,喻忠军,等.基于EfficientNet的多通道雷达目标微动特征分类方法[J].系统工程与电子技术,2024,46(9):2988-2998. doi: 10.12305/j.issn.1001-506X.2024.09.11
	WANGX Y,LUOY H,YUZ J,et al.A multi-channel radar target micro-motion feature classification method based on EfficientNet[J].Systems Engineering and Electronics,2024,46(9):2988-2998. doi: 10.12305/j.issn.1001-506X.2024.09.11
26	HANIFA,MUAZM,HASANA.Micro-Doppler based target recognition with radars: a review[J].IEEE Sensors Journal,2022,22(4):2948-2961.
27	李开明,张群,罗迎.基于微多普勒效应和AR模型的车辆目标分类方法[J].电子学报,2018,46(4):805-813.
	LIK M,ZHANGQ,LUOY.A classification method of vehicle targets based on micro-Doppler effect and auto regression model[J].Acta Electronica Sinica,2018,46(4):805-813.
28	JIANG W, LIU Y, WEI Q C, et al. A high-resolution radar automatic target recognition method for small UAVs based on multi-feature fusion[C]//Proc. of the IEEE 25th International Conference on Computer Supported Cooperative Work in Design, 2022: 775-779.
29	COSKUNA,BILICZS.Target classification using radar cross-section statistics of millimeter-wave scattering[J].The International Journal for Computation and Mathematics in Electrical and Electronic Engineering,2023,42(5):1199-1211.
30	丁鹭飞.雷达原理[M].3版西安:西安电子科技大学出版社,2014.
	DINGL F.Radar principles[M].3rd edXi'an:Xidian University Press,2014.
31	LIX Y,HUANGY H,YINK Y.Classification of vehicles with high-speed airborne radar based on micro-Doppler signatures[J].IETE Technical Review,2017,35(2):180-189.
32	LI T, TANG Y H. Frequency estimation based on modulation FFT and MUSIC algorithm[C]//Proc. of the International Conference on Pervasive Computing, Signal Processing and Applications, 2010: 525-528.

目标类别	速度分布范围/(m/s)	距离分布范围/m
人	1~5	200~20 000
车	1~22	200~20 000
无人机	1~15	200~20 000

目标类别	低速区	高速区	样本总数
人	18 848	—	18 848
车	12 118	6 844	18 962
无人机	16 172	16 668	32 840

区域	测试集识别率
区域	特征谱法	FFT法	MUSIC谱估计	AR模型谱估计
低速区	92.73	94.55	92.81	95.21
高速区	95.79	97.21	96.25	97.92

区域	测试集识别率
区域	17维能量分布特征	5维时域和频域特征	22维组合特征
低速区	88.64	91.35	95.21
高速区	91.17	96.08	97.92

目标类型	人	车	无人机	测试集识别率/%	总识别率/%
人	3 590	122	60	95.17	95.21
车	95	2 175	66	93.11	95.21
无人机	59	45	3 132	96.78	95.21