基于生成对抗网络的谐波SAR自聚焦方法

doi:10.12305/j.issn.1001-506X.2026.11.10

系统工程与电子技术 ›› 2026, Vol. 48 ›› Issue (11): 3779-3791.doi: 10.12305/j.issn.1001-506X.2026.11.10

基于生成对抗网络的谐波SAR自聚焦方法

史智明(), 易建新(), 万显荣(), 龚子平()

武汉大学电子信息学院，湖北武汉 430072

收稿日期:2025-09-01 修回日期:2025-09-29 接受日期:2025-09-30 出版日期:2026-10-25 发布日期:2026-01-19
通讯作者: 万显荣 E-mail:shichiminh@whu.edu.cn;jxyi@whu.edu.cn;xrwan@whu.edu.cn;zpgong@whu.edu.cn
作者简介:史智明（1993—），男，博士研究生，主要研究方向为合成孔径雷达成像处理、雷达图像检测
易建新（1989—），男，副教授，博士，主要研究方向为外辐射源雷达信号处理、非线性信号处理
龚子平（1977—），男，讲师，博士，主要研究方向为高频无线电波传播、天线理论分析与设计
第一联系人：
共同作者
基金资助:
国家自然科学基金（61931015, 62071335, 61831009）；湖北省自然科学基金（2021CFA002）资助课题

Autofocus method based on generative adversarial networks for harmonic SAR

Zhiming Shi(), Jianxin Yi(), Xianrong Wan(), Ziping Gong()

School of Electronics and Information，Wuhan University，Wuhan 430072，China

Received:2025-09-01 Revised:2025-09-29 Accepted:2025-09-30 Online:2026-10-25 Published:2026-01-19
Contact: Xianrong Wan E-mail:shichiminh@whu.edu.cn;jxyi@whu.edu.cn;xrwan@whu.edu.cn;zpgong@whu.edu.cn

摘要/Abstract

摘要：

针对谐波合成孔径雷达（harmonic synthetic aperture radar, HSAR)非平稳平台偏离理想轨迹产生运动误差，且谐波信号对该误差敏感而加剧成像散焦的问题，提出一种基于生成对抗网络的自聚焦方法。首先，通过多尺度分割与拼接模块捕捉多尺度运动误差特征，强化对微小运动误差的感知能力。然后，利用压缩−激励模块，自适应分配不同通道的权重，突出关键图像特征并抑制运动误差特征。仿真与实测结果表明，所提方法可显著改善非平稳平台运动误差引起的成像散焦问题，有效缓解HSAR对运动误差的敏感性，提升HSAR成像聚焦质量。

关键词: 谐波合成孔径雷达, 运动误差, 自聚焦, 生成对抗网络

Abstract:

To address the problem that for harmonic synthetic aperture radar (HSAR), non-stationary platforms deviate from the ideal trajectory, generating motion errors, and harmonic signals are sensitive to such errors, thereby exacerbating imaging defocus, a generative adversarial networks-based autofocus method is proposed. Firstly, the multiscale split and concat module is utilized to capture multi-scale motion error features, enhancing the ability to perceive tiny motion errors. Secondly, the squeeze and excitation module is employed to adaptively allocate weights across different channels, highlighting key image features while suppressing motion error features. The simulated and measured results demonstrate that this method significantly mitigates the imaging defocus caused by motion errors of non-stationary platforms, effectively reduces the sensitivity of HSAR to motion errors, and improves the focusing quality of HSAR imaging.

Key words: harmonic synthetic aperture radar (HSAR), motion error, autofocus, generative adversarial network (GAN)

中图分类号:

TP 183

史智明, 易建新, 万显荣, 龚子平. 基于生成对抗网络的谐波SAR自聚焦方法[J]. 系统工程与电子技术, 2026, 48(11): 3779-3791.

Zhiming Shi, Jianxin Yi, Xianrong Wan, Ziping Gong. Autofocus method based on generative adversarial networks for harmonic SAR[J]. Systems Engineering and Electronics, 2026, 48(11): 3779-3791.

图/表 24

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

表1

图11

图12

图13

图14

图15

表2

图16

图17

图18

图19

图20

图21

表3

参考文献 30

1	Braun T T, Hansen S, Pohl N. Emerging trends in radar: harmonic radar systems with mm waves[J]. IEEE Aerospace and Electronic Systems Magazine, 2025, 40 (6): 24. doi: 10.1109/MAES.2025.3540038
2	Zeng L Y, Li C Z. Recent advances in intermodulation and harmonic radar sensing and tracking[C]//IEEE Asia Pacific Microwave Conference, 2023: 846.
3	Wu B C, Zheng X Y, Li C M, et al. A locating harmonic radar-based sensor for vital-sign health monitoring with ultra-low power harmonic conversion loss compensation[J]. IEEE Trans. on Microwave Theory and Techniques, 2025, 73 (8): 4402. doi: 10.1109/TMTT.2025.3564475
4	Hansen S, Nowok S, Shoykhetbrod A, et al. Distributed sensor network for 3D tag localization using harmonic radar at 61/122 GHz ISM band [C]//IEEE 20th European Radar Conference, 2023: 343.
5	Ala R, Rouse C D, Colpitts B G. An extra low-mass harmonic radar transponder for insect tracking applications[J]. IEEE Trans. on Radar Systems, 2023, 1, 146. doi: 10.1109/TRS.2023.3286270
6	Braun T T, Schopfel Chöpfel J, Bredendiek C, et al. Introducing inharmonic radar: tag detection in the automotive bands of present and future at 76–81/134–141 GHz via fractional multiplication[J]. IEEE Journal of Microwaves, 2024, 4 (3): 473. doi: 10.1109/JMW.2024.3412415
7	Chen X, Dong Z, Zhang Z H, et al. Very high resolution synthetic aperture radar systems and imaging: a review[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2024, 17, 7104. doi: 10.1109/JSTARS.2024.3374429
8	Manzoni M, Tagliaferri D, Rizzi M, et al. Motion estimation and compensation in automotive MIMO SAR[J]. IEEE Trans. on Intelligent Transportation Systems, 2022, 24 (2): 1756.
9	Navarro Castillo J P, Scheiber R, Jäger M, et al. Accurate relative motion compensation for digital beamforming in airborne SAR[C]//European Conference on Synthetic Aperture Radar, 2024: 795.
10	Fan L, Wang H Q, Yang Q, et al. High-quality airborne terahertz video SAR imaging based on echo-driven robust motion compensation[J]. IEEE Trans. on Geoscience and Remote Sensing, 2024, 62, 2001817.
11	Li Y K, Liu C, Wang Y F, et al. A robust motion error estimation method based on raw data[J]. IEEE Trans. on Geoscience and Remote Sensing, 2012, 50 (7): 2780. doi: 10.1109/TGRS.2011.2175737
12	Ran L, Liu Z, Li T, et al. Extension of map-drift algorithm for highly squinted SAR autofocus[J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2017, 10 (9): 4032.
13	Wang G Y, Zhang M, Huang Y, et al. Robust two-dimensional spatial-variant map-drift algorithm for UAV SAR autofocusing[J]. Remote Sensing, 2019, 11 (3): 340. doi: 10.3390/rs11030340
14	Huang Y, Liu F Y, Chen Z Y, et al. An improved map-drift algorithm for unmanned aerial vehicle SAR imaging[J]. IEEE Geoscience and Remote Sensing Letters, 2021, 18 (11): 1. doi: 10.1109/LGRS.2020.3011973
15	Evers A, Jackson J A. Generalized phase gradient autofocus algorithm[J]. IEEE Trans. on Computational Imaging, 2019, 5 (4): 606. doi: 10.1109/TCI.2019.2899453
16	Zhang T, Li Y, Zhang T, et al. Expediting phase gradient autofocus algorithm for SAR imaging[C]//IEEE International Geoscience and Remote Sensing Symposium, 2020: 1145.
17	Miao Y X, Wu J J, Yang J Y. Azimuth migration-corrected phase gradient autofocus for bistatic SAR polar format imaging[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 18 (4): 697.
18	Wang J, Liu X. SAR minimum-entropy autofocus using an adaptive-order polynomial model[J]. IEEE Geoscience and Remote Sensing Letters, 2006, 3 (4): 512. doi: 10.1109/LGRS.2006.878446
19	Xiong T, Xing M, Wang Y, et al. Minimum-entropy-based autofocus algorithm for SAR data using Chebyshev approximation and method of series reversion, and its implementation in a data processor[J]. IEEE Trans. on Geoscience and Remote Sensing, 2014, 52 (3): 1719. doi: 10.1109/TGRS.2013.2253781
20	Meng Z C, Zhang L, Ma Y, et al. Accelerating minimum entropy autofocus with stochastic gradient for UAV SAR imagery[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19, 4017805.
21	Liu Z, Yang S, Feng Z, et al. Fast SAR autofocus based on ensemble convolutional extreme learning machine[J]. Remote Sensing, 2021, 13 (14): 2683. doi: 10.3390/rs13142683
22	Liu Z, Yang S, Gao Q, et al. AFnet and PAFnet: fast and accurate SAR autofocus based on deep learning[J]. IEEE Trans. on Geoscience and Remote Sensing, 2022, 60
23	Goodfellow I, Pouget-Abadie J, MIRZA M, et al. Generative adversarial nets[C]//27th International Conference on Neural Information Systems, 2014, 2: 2672.
24	胡铭菲, 左信, 刘建伟. 深度生成模型综述[J]. 自动化学报, 2022, 48 (1): 40.
25	Lamberti P W, Majtey A P. Non-logarithmic Jensen-Shannon divergence[J]. Physica A: Statistical Mechanics and its Applications, 2003, 329 (1/2): 81. doi: 10.1016/S0378-4371(03)00566-1
26	Hu J, Shen L, Sun G. Squeeze and excitation networks[C]//IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7132.
27	Kupyn O, Budzan V, Mykhailych M, et al. DeblurGAN: blind motion deblurring using conditional adversarial networks[C]//IEEE Conference on Computer Vision and Pattern Recognition, 2018: 8183.
28	Zhang H, Zu K K, Lu J, et al. EPSANet: an efficient pyramid squeeze attention block on convolutional neural network [C]//Asian Conference on Computer Vision, 2022: 1161.
29	Rjovsky M, Chintala S, Bottou L. Wasserstein generative adversarial networks[C]//International Conference on Machine Learning, 2017: 214.
30	Johnson J, Alahi A, Li F F. Perceptual losses for real-time style transfer and super-resolution[C]//14th European Conference on Computer Vision, 2016: 694.

参数	值
基频 /谐波载频/GHz	5.5/11
基频 /谐波带宽/ MHz	50/100
脉冲重复频率/Hz	100

自聚焦方法	有效速度/（m/s）
自聚焦方法	0.35		0.5		0.7		0.9
GPGA	1.7559	1.3628		1.5249		1.7178
SG-MEA	1.5973	1.3505		1.4321		1.5232
AF-Net	1.6181	1.3572		1.4469		1.5319
PAF-Net	1.6025	1.3519		1.4395		1.5283
AF-GAN	1.5629	1.3375		1.3629		1.4102

自聚焦方法	熵
自聚焦方法	场景1	场景2	场景3
GPGA	2.8221	2.8473	2.9124
SG-MEA	2.6180	2.4548	2.6508
AF-Net	2.6387	2.4612	2.6636
PAF-Net	2.6213	2.4582	2.6595
AF-GAN	2.4286	2.3223	2.4652

[1]	吉家欣, 李汀, 李飞. 一种学习多域数据联合分布的量子耦合生成对抗网络[J]. 系统工程与电子技术, 2026, 48(3): 779-786.
[2]	张振兴, 杨任农, 李永林, 左家亮, 胡利平, 陈双艳. 基于双重注意力生成对抗网络的文本到图像生成[J]. 系统工程与电子技术, 2026, 48(1): 34-43.
[3]	黄丰卓, 冯东, 华洋晟, 黄晓涛. 一种改进的全息SAR相位梯度自聚焦算法[J]. 系统工程与电子技术, 2025, 47(6): 1786-1795.
[4]	张兰, 张彪, 梁天一, 朱辉杰. 面向电磁信息智能控制的生成对抗网络研究进展[J]. 系统工程与电子技术, 2025, 47(3): 730-744.
[5]	贾钰嘉, 张思乾, 唐涛, 匡纲要. 强化散射特征的机载SAR实传图像盲超分辨重建[J]. 系统工程与电子技术, 2025, 47(3): 753-767.
[6]	唐曦, 李文海, 唐贞豪, 李睿峰, 李根. 基于DBSCAN和CGAN的不平衡数据过采样方法[J]. 系统工程与电子技术, 2025, 47(11): 3739-3753.
[7]	薛丽莎, 葛瑞星, 朱宇轩, 鲍雁飞. 基于SABEGAN的通信干扰信号生成与效能分析[J]. 系统工程与电子技术, 2024, 46(6): 2155-2163.
[8]	汤寓麟, 王黎明, 余德荧, 李厚朴, 刘敏, 张卫东. 基于CSLS-CycleGAN的侧扫声纳水下目标图像样本扩增法[J]. 系统工程与电子技术, 2024, 46(5): 1514-1524.
[9]	马兰, 孟诗君, 吴志军. 基于BERT与生成对抗的民航陆空通话意图挖掘[J]. 系统工程与电子技术, 2024, 46(2): 740-750.
[10]	庞世燕, 郝京京, 邢立宁, 谭旭. 融合域自适应和弱监督策略的遥感影像建筑物提取算法[J]. 系统工程与电子技术, 2023, 45(6): 1616-1623.
[11]	陈丽, 方梓涵, 梅立泉. 基于GAN的直扩信号生成算法[J]. 系统工程与电子技术, 2023, 45(5): 1544-1552.
[12]	孟智超, 张磊, 卢景月, 李军. 随机样本选择的合成孔径雷达距离空变相位梯度自聚焦算法[J]. 系统工程与电子技术, 2023, 45(12): 3828-3835.
[13]	曾浩然, 冯蕴雯, 路成, 潘维煌. 基于改进生成对抗网络的国产民机航材消耗预测方法[J]. 系统工程与电子技术, 2023, 45(10): 3132-3138.
[14]	秦博伟, 蒋磊, 许华, 牛伟宇. 基于RE-GAN的调制信号开集识别算法[J]. 系统工程与电子技术, 2023, 45(10): 3321-3328.
[15]	刘戎翔, 吴琳, 谢智歌, 刘虹麟. 基于生成对抗网络的防空体系态势辅助分析[J]. 系统工程与电子技术, 2022, 44(8): 2522-2529.

基于生成对抗网络的谐波SAR自聚焦方法

Autofocus method based on generative adversarial networks for harmonic SAR

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 24

参考文献 30

相关文章 15

编辑推荐

Metrics

本文评价