基于MGMD空间的窄带雷达空中目标识别方法

doi:10.12305/j.issn.1001-506X.2025.04.10

系统工程与电子技术 ›› 2025, Vol. 47 ›› Issue (4): 1136-1145.doi: 10.12305/j.issn.1001-506X.2025.04.10

基于MGMD空间的窄带雷达空中目标识别方法

郭泽坤¹, 杨洪飞², 刘峥¹^,*, 谢荣¹, 冉磊¹, 李嘉楠¹

1. 西安电子科技大学雷达信号处理全国重点实验室, 陕西西安 710071
2. 江南机电设计研究所, 贵州贵阳 550025

收稿日期:2023-11-21 出版日期:2025-04-25 发布日期:2025-05-28
通讯作者: 刘峥
作者简介:郭泽坤 (1994—), 男, 博士研究生, 主要研究方向为雷达目标识别、深度学习、小样本学习
杨洪飞 (1988—), 男, 工程师, 博士, 主要研究方向为探测系统信息处理、计算机视觉
刘峥 (1964—), 男, 教授, 博士, 主要研究方向为雷达精确制导、多传感器信息融合、雷达信号处理的理论与系统设计
谢荣 (1981—), 男, 副教授, 博士, 主要研究方向为雷达精确制导、雷达智能抗干扰
冉磊 (1989—), 男, 副教授, 博士, 主要研究方向为高速高机动平台雷达成像、多通道阵列雷达成像
李嘉楠 (2000—), 女, 硕士研究生, 主要研究方向为多传感器信息融合目标识别、雷达目标回波仿真
基金资助:
国家自然科学基金(62001346);雷达信号处理全国重点实验室支持计划(KGJ202205)

Narrow-band radar airborne target recognition method based on MGMD space

Zekun GUO¹, Hongfei YANG², Zheng LIU¹^,*, Rong XIE¹, Lei RAN¹, Jia'nan LI¹

1. National Key Laboratory of Radar Signal Processing, Xidian University, Xi'an 710071, China
2. Jiangnan Electromechanical Design Institute, Guiyang 550025, China

Received:2023-11-21 Online:2025-04-25 Published:2025-05-28
Contact: Zheng LIU

摘要/Abstract

摘要：

针对窄带雷达对短时观测回波(echoes from short-term observation, ESTO)序列特征识别性能较低且易受诱饵干扰的问题, 基于最大边缘高斯混合分布(maximum-margin Gaussian mixture distribution, MGMD)空间提出一种空中目标识别方法。首先, 构建MGMD空间, 对库内类别预设类中心, 将每个类别中心之间的距离最大化, 并且类中心对应MGMD的均值。其次, 构建包含特征注意力机制的深度网络, 将ESTO序列映射至MGMD空间。然后, 通过训练使每类深度特征服从于MGMD, 从而使其边缘最大化, 以此提升模型的分类泛化性能和对诱饵的鉴别能力。实验结果表明, 所提方法能够有效提升窄带雷达对目标ESTO序列的分类性能和对诱饵目标的鉴别能力。

关键词: 窄带雷达, 空中目标识别, 短时观测回波, 最大边缘高斯混合分布

Abstract:

To solve the problem that narrow-band radar has low performance in recognizing echoes from short-term observations (ESTO) sequence features and is susceptible to decoy interference, an aerial target recognition method is proposed based on the maximum-margin Gaussian mixture distribution (MGMD) space. Firstly, the MGMD space is constructed and the class center for the categories in the library is preset. The distance between each category center is maximized and the class center corresponds to the mean value of MGMD. Secondly, a deep network containing the feature attention mechanism is built to map the ESTO sequence to the MGMD space. Then, through training, each type of deep feature is subject to MGMD, which maximizes its edge and improves the model's classification generalization performance and its ability to recognize the baits. Experimental results show that the proposed method can effectively improve the narrow-band radar classification performance of target ESTO sequences and its ability to recognize the decoy targets.

Key words: narrow-band radar, airborne target recognition, echoes from short-term observation (ESTO), maximum-margin Gaussian mixture distribution (MGMD)

中图分类号:

TN959.1

郭泽坤, 杨洪飞, 刘峥, 谢荣, 冉磊, 李嘉楠. 基于MGMD空间的窄带雷达空中目标识别方法[J]. 系统工程与电子技术, 2025, 47(4): 1136-1145.

Zekun GUO, Hongfei YANG, Zheng LIU, Rong XIE, Lei RAN, Jia'nan LI. Narrow-band radar airborne target recognition method based on MGMD space[J]. Systems Engineering and Electronics, 2025, 47(4): 1136-1145.

图/表 18

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

表1

表2

图12

图13

表3

图14

表4

参考文献 30

1	FENG X J, HUO C Y, ZHU C X, et al. Multi-dimensional feature fusion based on narrow-band RCS for UAV target recognition[C]//Proc. of the Cross Strait Radio Science & Wireless Technology Conference, 2022.
2	CHEN V C , LI F Y , SHEN S H , 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.
3	陈一畅, 熊鑫, 王万田. 基于稀疏FrFT的窄带雷达目标架次识别方法[J]. 系统工程与电子技术, 2021, 43 (8): 2129- 2136. doi: 10.12305/j.issn.1001-506X.2021.08.14
	CHEN Y C , XIONG X , WANG W T . Target sortie identification method of narrow-band radar based on sparse fractional Fourier transform[J]. Systems Engineering and Electronics, 2021, 43 (8): 2129- 2136. doi: 10.12305/j.issn.1001-506X.2021.08.14
4	WANG W T , TANG Z Y , CHEN Y C , et al. Aircraft target classification for conventional narrow-band radar with multi-wave gates sparse echo data[J]. Remote Sensing, 2019, 11 (22): 2700- 2712. doi: 10.3390/rs11222700
5	XIA S Q , ZHANG C W , CAI W Y , et al. Aircraft target classification method for conventional narrow-band radar based on micro-Doppler effect[J]. Mathematical Problems in Engineering, 2022, 3154854.
6	林青松, 胡卫东, 虞华, 等. 低分辨雷达回波序列轮廓像目标分类方法研究[J]. 现代雷达, 2005, 27 (3): 24- 28.
	LIN Q S , HU W D , YU H , et al. A study of target classification method based on low-resolution radar return sequences image profile[J]. Modern Radar, 2005, 27 (3): 24- 28.
7	梁复台, 李宏权, 张晨浩. 基于深度迁移学习的窄带雷达群目标识别方法[J]. 兵器装备工程学报, 2020, 41 (4): 143- 147.
	LIANG F T , LI H Q , ZHANG C H . Narrow-band radar unresolved targets recognition method based on deep transfer learning[J]. Journal of Ordnance Engineering, 2020, 41 (4): 143- 147.
8	GAO Y , ZHOU Y , WANG Y , et al. Narrow-band radar automatic target recognition based on a hierarchical fusing network with multidomain features[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 18 (6): 1039- 1043.
9	REN K , DU L , WANG B S , et al. Statistical compressive sensing and feature extraction of time-frequency spectrum from narrow-band radar[J]. IEEE Trans. on Aerospace and Electronic Systems, 2019, 56 (1): 326- 342.
10	WENGROWSKI E , PURRI M , DANA K , et al. Deep CNNs as a method to classify rotating objects based on monostatic RCS[J]. IET Radar, Sonar & Navigation, 2019, 13 (7): 1092- 1100.
11	TIAN X D , BAI X R , ZHOU F . Recognition of micro-motion space targets based on attention-augmented cross-modal feature fusion recognition network[J]. IEEE Trans. on Geoscience and Remote Sensing, 2023, 61 (1): 5104909.
12	GAO Y , ZHOU Y , WANG Y , et al. Narrow-band radar automatic target recognition based on a hierarchical fusing network with multidomain features[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 18 (6): 1039- 1043.
13	MCLACHLAN G J . Mahalanobis distance[J]. Resonance, 1999, 4 (6): 20- 26.
14	TIAN L , CHEN B , GUO Z K , et al. Open set HRRP recognition with few samples based on multi-modality prototypical networks[J]. Signal Processing: the Official Publication of the European Association for Signal Processing, 2022, 193, 108391.
15	PANG T Y, DU C, ZHU J. Max-mahalanobis linear discriminant analysis networks[C]//Proc. of the International Conference on Machine Learning, 2018: 4016-4025.
16	WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Proc. of the European Conference on Computer Vision, 2018: 3-19.
17	胡明春, 王建明, 孙俊, 等. 雷达目标识别原理与实验技术[M]. 北京: 国防工业出版社, 2017: 12- 13.
	HU M C , WANG J M , SUN J , et al. Principle and experiments of radar target recognition technology[M]. Beijing: National Defense Industry Press, 2017: 12- 13.
18	GOODFELLOW I , POUGET-ABADIE J , MIRZA M , et al. Generative adversarial nets[J]. Communications of the ACM, 2014, 11 (64): 2672- 2680.
19	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
20	CHEN J , XU S Y , CHEN Z P . Convolutional neural network for classifying space target of the same shape by using RCS time series[J]. IET Radar, Sonar & Navigation, 2018, 12 (11): 1268- 1275.
21	CHEN J, XU S Y, HU P J, et al. Precession period extraction of axisymmetric space target from RCS sequence via convolutional neural network[C]//Proc. of the Progress in Electromagnetics Research Symposium, 2018.
22	AL-SHALABI L , SHAABAN Z , KASASBEH B . Data mining: a preprocessing engine[J]. Journal of Computer Science, 2006, 2 (9): 735- 739.
23	TAX D M J , DUIN R P W . Support vector data description[J]. Machine Learning, 2004, 54 (1): 45- 66.
24	PARVIN H , ALIZADEH H , MINAEIBIDGOLI B . MKNN: modified k-nearest neighbor[J]. Lecture Notes in Engineering & Computer Science, 2008, 2173 (1): 2054- 2062.
25	RUFF L, VANDERMEULEN R A, NICO G, et al. Deep one-class classification[C]//Proc. of the International Conference on Machine Learning, 2018: 4393-4402.
26	WAN J W , CHEN B , XU B , et al. Convolutional neural networks for radar HRRP target recognition and rejection[J]. EURASIP Journal on Advances in Signal Processing, 2019, 2019, 5.
27	LEE K, MAJI S, RAVICHANDRAN A, et al. Meta-learning with differentiable convex optimization[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 10657-10665.
28	AKITA T, MITA S. Object tracking and classification using millimeter-wave radar based on LSTM[C]//Proc. of the IEEE Intelligent Transportation Systems Conference, 2019: 1110-1115.
29	LAURENS V D M , HINTON G . Visualizing data using t-SNE[J]. Journal of Machine Learning Research, 2008, 9 (11): 2579- 2605.
30	KOBAK D , LINDERMAN G C . Initialization is critical for preserving global data structure in both t-SNE and UMAP[J]. Nature Biotechnology, 2021, 39 (2): 156- 157.

算法	鉴别率/%	时间/s
SVDD^[23]	78.03	0.052
w-KNN^[24]	76.81	0.046
Deep-SVDD^[25]	91.55	0.108
SCNN^[26]	82.76	0.079
MCTR-Net	95.93	0.073

算法	分类准确率/%	时间/s
SVM	35.62	3.52
CNN-SVM^[27]	86.78	2.86
CNN-KNN	86.59	2.89
LSTM^[28]	90.16	3.23
SCNN^[26]	88.24	3.18
MCTR-Net	96.61	2.68

算法	分类准确率/%	时间/s
CNN-Softmax	87.35	3.262
CNN-FAM-Softmax	91.42	2.630
CNN-MMC	93.95	2.598
MCTR-Net (CNN-FAM-MMC)	96.61	2.677

标准差	0.1	0.5	1	2	3	4
RMSE	0.068	0.334	0.668	1.336	2.004	2.673
识别率/%	95.76	92.52	90.11	87.88	83.97	84.63

标准差	5	6	7	8	9	10
RMSE	3.341	4.009	4.670	5.345	6.013	6.681
识别率/%	83.63	82.42	83.64	83.23	84.04	82.83

标准差	12	14	16	18	20	-
RMSE	8.018	9.354	10.690	12.026	13.363	-
识别率/%	80.20	78.99	77.17	76.77	75.98	-

[1]	尹建国, 盛文, 蒋伟. 基于深度残差收缩网络的雷达空中目标识别[J]. 系统工程与电子技术, 2024, 46(9): 3012-3018.
[2]	刘帅康, 曹伟, 管志强, 杨学岭, 许金鑫. 基于RMDLPP的雷达空中目标分类[J]. 系统工程与电子技术, 2024, 46(4): 1220-1228.
[3]	陈一畅, 熊鑫, 王万田. 基于稀疏FrFT的窄带雷达目标架次识别方法[J]. 系统工程与电子技术, 2021, 43(8): 2129-2136.
[4]	李靖卿, 冯存前, 龙戈农, 王义哲. 基于多站微多普勒融合的弹道目标特征提取[J]. 系统工程与电子技术, 2016, 38(7): 1549-1554.
[5]	胡晓伟, 童宁宁, 何兴宇, 雷腾, 王宇晨. 基于窄带雷达组网的高速自旋目标成像[J]. 系统工程与电子技术, 2016, 38(4): 792-798.
[6]	杨秋, 张群, 王敏, 孙莉. 基于机载窄带雷达的舰船目标多普勒特性分析[J]. 系统工程与电子技术, 2015, 37(12): 2733-2738.
[7]	黄璟，肖志河，任红梅. 窄带相参雷达调制谱超分辨处理方法[J]. Journal of Systems Engineering and Electronics, 2010, 32(9): 1894-1897.
[8]	苏军海, 吕孝雷, 邢孟道, 张守宏. 窄带雷达高速多目标检测及其运动参数估计[J]. Journal of Systems Engineering and Electronics, 2009, 31(7): 1539-1543.

基于MGMD空间的窄带雷达空中目标识别方法

Narrow-band radar airborne target recognition method based on MGMD space

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 18

参考文献 30

相关文章 8

编辑推荐

Metrics

本文评价