具有稳定支撑集的位移测量MIMO雷达稀疏成像方法

晋良念; 黄渤孙

doi:10.11999/JEIT260470

具有稳定支撑集的位移测量MIMO雷达稀疏成像方法

doi: 10.11999/JEIT260470 cstr: 32379.14.JEIT260470

晋良念^{1, 2, ,},
黄渤孙¹

1.
桂林电子科技大学信息与通信学院桂林 541004
2.
广西无线宽带通信与信号处理重点实验室桂林 541004

基金项目: 国家自然科学基金(62661017)，广西重点研发项目(桂科FN2600640261，桂科AB23075161)，广西无线宽带通信与信号处理重点实验室基金项目基金项目资助(GXKL06240110)

详细信息

作者简介:
晋良念：男，教授，研究方向为雷达信号处理，邮箱　jing@guet.edu.cn

黄渤孙：男，硕士生，研究方向为雷达信号处理

通讯作者:
晋良念　jing@guet.edu.cn

中图分类号: TN957.51
计量
- 文章访问数: 10
- HTML全文浏览量: 2
- PDF下载量: 0
- 被引次数: 0
出版历程
- 修回日期: 2026-10-16
- 录用日期: 2026-10-16
- 网络出版日期: 2025-10-25

A Support Set Stabilization Method for MIMO Radar Sparse Imaging in Displacement Measurement

JIN Liangnian^{1, 2
, ,},
HUANG Bosun¹

1.
Institute of Information and Communication, Guilin University of Electronic Technology, Guilin 541004, China
2.
XXXXXXXXXX, Guilin 541004, China

Funds: The National Natural Science Foundation of China(62661017), Guangxi Key Research and Development Program (Guike FN2600640261, Guike AB23075161), Supported by the Project of Guangxi Wireless Broadband Communication and Signal Processing Key Laboratory(GXKL06240110)

摘要

摘要: 调频连续波和多输入多输出(FMCW-MIMO)毫米波雷达通过稀疏成像和相位差分技术，可实现全场位移测量。然而，现有的稀疏成像方法容易导致多帧重建图像的目标位置集合(支撑集)出现波动，破环跨帧相位一致性，从而影响位移测量的精度。为此，提出一种结合深度学习UNet增益模块与稀疏优化交替方向乘子法(UNetE-ADMM)的MIMO雷达成像方法。首先，将UNet目标感知模型嵌入ADMM近端算子中，使近端算子能够识别目标主瓣与旁瓣区域并进行差异化加权，以增强主瓣、抑制旁瓣；其次，在ADMM迭代框架下，利用结构先验引导迭代优化过程，快速抑制旁瓣干扰并纠正偏离主瓣的位置，以获得稳定的支撑集；最后，对多帧成像结果进行相位差分处理，提取得到目标径向位移信息。实验结果表明，该方法较于L1正则化交替乘子法(L1-ADMM)和即插即用(PnP)方法在主瓣漂移概率和位移均方误差根均有提升效果，在人体特征测量实验的心率谱尖峰更加清晰，表明该方法在稳定支撑集和位移测量精度具备优势。
- 位移测量 /
- MIMO雷达 /
- 稀疏成像 /
- 交替方向乘子法 /
- UNet
Abstract: Objective FMCW-MIMO radar provides a non-contact approach for full-field displacement measurement by combining sparse imaging with inter-frame phase-difference estimation. Compared with conventional point-based measurement methods, radar imaging enables displacement information to be obtained over multiple spatial targets simultaneously. However, the accuracy of phase-based displacement estimation strongly depends on the consistency of the reconstructed target locations between consecutive frames. Existing sparse imaging methods may produce fluctuations in the reconstructed support set because of sidelobe leakage, target omission, and mainlobe drift, particularly under low-SNR conditions. Such support-set instability introduces phase errors and causes discontinuities or fluctuations in the estimated displacement. Therefore, a sparse imaging method capable of maintaining a stable support set while suppressing sidelobe interference is developed for high-precision MIMO radar displacement measurement.Methods A UNet-enhanced ADMM (UNetE-ADMM) algorithm is proposed by embedding a UNet gain module into the proximal operator of ADMM. Spatially adaptive normalized weights are generated from imaging features to distinguish target mainlobes from sidelobes. The mainlobes are preserved while sidelobes are progressively suppressed, allowing the support set to be stabilized during iterative reconstruction. The reconstructed complex scattering coefficients are subsequently used for inter-frame phase-difference estimation and radial displacement measurement. A support-set stability criterion and displacement-error model are further established to analyze the effects of mainlobe drift, grid mismatch, and inter-Tx motion.Results and Discussions Simulations are conducted using a cascaded AWR2243 MIMO radar configuration with a 77-GHz carrier frequency, 3.65-GHz bandwidth, and 50-Hz frame rate (Table 2). For distributed point targets under SNRs from −10 to 10 dB, the proposed method produces more stable target locations and lower sidelobe levels than L1-ADMM and PnP-UNet (Fig. 9). At −10 dB, the mainlobe drift probability is reduced from 37.5% for L1-ADMM to 6.3%, corresponding to an 83.2% reduction, while the displacement RMSE decreases from 44.62 to 1.54 μm (Table 3). In the vibration experiment with two corner reflectors, the proposed method achieves displacement RMSEs of 185 and 18 μm for the moving and stationary targets, respectively, compared with 2247 and 122 μm for L1-ADMM (Table 4). In the human vital-sign experiment, the reconstructed chest displacement is smoother (Fig. 16), and the heartbeat spectral peak is more distinguishable (Fig. 17), while the measured heart rates remain consistent with the reference measurements (Table 5). The convergence and target-background-ratio results further demonstrate effective sidelobe suppression and stable iterative reconstruction (Fig. 18).Conclusions The proposed UNetE-ADMM method integrates learned target structural priors into ADMM sparse imaging and improves support-set stability through adaptive mainlobe preservation and sidelobe suppression. Experimental results demonstrate reduced mainlobe drift and improved displacement accuracy compared with conventional L1-ADMM and PnP-UNet, particularly under low-SNR conditions. The method is also applicable to both discrete targets and continuous structures. Grid mismatch, computational cost associated with dense imaging grids, and generalization to more complex scenes remain to be further investigated.
- Displacement measurement /
- MIMO radar /
- Sparse imaging /
- ADMM /
- UNet

HTML全文

图 1 MIMO雷达虚拟阵元几何分布

下载: 全尺寸图片幻灯片

图 2 UNetE-ADMM迭代流程图

下载: 全尺寸图片幻灯片

图 3 UNet网络结构

下载: 全尺寸图片幻灯片

图 4 训练数据构建示意图

下载: 全尺寸图片幻灯片

图 5 UNet网络内部结构

下载: 全尺寸图片幻灯片

图 6 正则化逆矩阵的热力图

下载: 全尺寸图片幻灯片

图 7 支撑集不稳定对位移测量的影响

下载: 全尺寸图片幻灯片

图 8 本文算法能量变化示意图

下载: 全尺寸图片幻灯片

图 9 成像结果对比

下载: 全尺寸图片幻灯片

图 10 所有目标的平均位移测量结果对比

下载: 全尺寸图片幻灯片

图 11 振动位移测量实验环境

下载: 全尺寸图片幻灯片

图 12 振动位移测量实验迭代结果对比

下载: 全尺寸图片幻灯片

图 13 位移曲线测量结果对比

下载: 全尺寸图片幻灯片

图 14 生命体征测量实验环境

下载: 全尺寸图片幻灯片

图 15 生命体征测量实验迭代结果对比

下载: 全尺寸图片幻灯片

图 16 左胸腔位移曲线测量结果对比

下载: 全尺寸图片幻灯片

图 17 两种状态的心率谱对比

下载: 全尺寸图片幻灯片

图 18 收敛曲线和目标背景比曲线

下载: 全尺寸图片幻灯片

表 1 训练集结构

	目标类型	数量
	离散结构目标(1–50个)	5000
	连续块状目标(1–10块)	5000
	随机组合	10000
	总计	20000

下载: 导出CSV

表 2 AWR2243硬件参数

	参数类型	参数值
	载频(GHz)	77
	带宽(GHz)	3.65
	调制斜率(MHz/us)	44.5
	采样频率(MHz)	15
	发射帧率(Hz)	50

下载: 导出CSV

表 3 主瓣漂移概率与平均位移RMSE对比

算法类型	主瓣漂移概率(%)					RMSE (μm)
算法类型	10 dB	5 dB	0 dB	–5 dB	–10 dB	RMSE (μm)
L1-ADMM	12.0	15.5	17.5	25.0	37.5	44.62
PnP-UNet	4.6	5.9	7.1	8.5	10.8	5.77
CIDT	3.2	3.7	5.3	6.4	8.2	0.43
本文方法	2.2	3.3	5.0	5.1	6.3	1.54

下载: 导出CSV

表 4 位移测量RMSE对比

算法类型	RMSE(μm)
算法类型	目标1	目标2
L1-ADMM	2247	122
PnP-UNet	879	60
CIDT	115	35
本文方法	185	18

下载: 导出CSV

表 5 平均心率测量结果对比

算法类型	静息心率(bpm)		运动心率(bpm)
算法类型	雷达	手环	雷达	手环
L1-ADMM	79	74	158	152
PnP-UNet	76	74	159	152
CIDT	76	74	155	152
本文方法	76	74	149	152

下载: 导出CSV

参考文献(25)

[1]	JANELIUKSTIS R and CHEN Xiao. Review of digital image correlation application to large-scale composite structure testing[J]. Composite Structures, 2021, 271: 114143. doi: 10.1016/j.compstruct.2021.114143.
[2]	MARDANSHAHI A, SREEKUMAR A, YANG Xin, et al. Sensing techniques for structural health monitoring: A state-of-the-art review on performance criteria and new-generation technologies[J]. Sensors, 2025, 25(5): 1424. doi: 10.3390/s25051424.
[3]	刘青旭, 陈海峰, BRYANSKY A, 等. 航天复合材料结构健康监测技术应用进展[J]. 复合材料学报, 2024, 41(9): 4563–4588. doi: 10.13801/j.cnki.fhclxb.20240606.002. LIU Qingxu, CHEN Haifeng, BRYANSKY A, et al. Progress in application on health monitoring technology for aerospace composite structures[J]. Acta Materiae Compositae Sinica, 2024, 41(9): 4563–4588. doi: 10.13801/j.cnki.fhclxb.20240606.002.
[4]	EHRHARDT D A, ALLEN M S, YANG Shifei, et al. Full-field linear and nonlinear measurements using Continuous-Scan Laser Doppler Vibrometry and high speed Three-Dimensional Digital Image Correlation[J]. Mechanical Systems and Signal Processing, 2017, 86: 82–97. doi: 10.1016/j.ymssp.2015.12.003.
[5]	WANG Yuchao, HU Weihua, TENG Jun, et al. Full-field displacement measurement of long-span bridges using one camera and robust self-adaptive complex pyramid[J]. Mechanical Systems and Signal Processing, 2024, 215: 111451. doi: 10.1016/j.ymssp.2024.111451.
[6]	王俊松, 张吴记, 潘志伟, 等. 基于激光散斑数字图像相关法的热应变测量[J]. 激光技术, 2023, 47(2): 171–177. doi: 10.7510/jgjs.issn.1001-3806.2023.02.003. WANG Junsong, ZHANG Wuji, PAN Zhiwei, et al Thermal strain measurement based on laser speckle digital image correlation method[J]. Laser Technology, 2023, 47(2): 171–177. doi: 10.7510/jgjs.issn.1001-3806.2023.02.003.
[7]	林赟, 时清, 王彦平, 等. 基于RPCA的地基SAR近距强耦合信号抑制算法研究[J]. 电子与信息学报, 2023, 45(4): 1321–1329. doi: 10.11999/JEIT220883. LIN Yun, SHI Qing, WANG Yanping, et al. Near-range strong coupled signal suppression algorithm based on RPCA for GBSAR[J]. Journal of Electronics & Information Technology, 2023, 45(4): 1321–1329. doi: 10.11999/JEIT220883.
[8]	方鑫, 何敏, 黄大荣, 等. 毫米波雷达微弱行人轨迹跟踪-预测一体化方法[J]. 仪器仪表学报, 2023, 44(11): 300–309. doi: 10.19650/j.cnki.cjsi.J2311824. FANG Xin, HE Min, HUANG Darong, et al. Integrated trajectory tracking and prediction method for weak pedestrian with millimeter wave radar[J]. Chinese Journal of Scientific Instrument, 2023, 44(11): 300–309. doi: 10.19650/j.cnki.cjsi.J2311824.
[9]	杜兰, 李逸明, 薛世鲲, 等. 结合相似度预测和阈值自动求解的开集条件下毫米波雷达点云步态识别方法[J]. 电子与信息学报, 2025, 47(6): 1850–1863. doi: 10.11999/JEIT241034. DU Lan, LI Yiming, XUE Shikun, et al. Millimeter-wave radar point cloud gait recognition method under open-set conditions based on similarity prediction and automatic threshold estimation[J]. Journal of Electronics & Information Technology, 2025, 47(6): 1850–1863. doi: 10.11999/JEIT241034.
[10]	MA Zhanxiong, HAI Lu, ZHANG Tongtong, et al. Accurate structural displacement measurement via enhanced millimeter-wave radar interferometry using multi-chirp-based adaptive phase unwrapping[J]. Measurement, 2026, 262: 120029. doi: 10.1016/j.measurement.2025.120029.
[11]	熊玉勇, 李松旭, 吴高阳, 等. 基于毫米波感知的形变及振动多点同步测量理论与方法[J]. 中国科学: 技术科学, 2021, 51(9): 998–1010. doi: 10.1360/SST-2020-0489. XIONG Yuyong, LI Songxu, WU Gaoyang, et al. Theory and method of multi-point synchronous deformation and vibration measurement based on millimeter-wave sensing[J]. Scientia Sinica Technologica, 2021, 51(9): 998–1010. doi: 10.1360/SST-2020-0489.
[12]	DONG Xuefei, LIU Yunxue, WANG Jinwei, et al. Towards heart rate estimation in complex multi-target scenarios: A high-precision FMCW radar scheme integrating HDBS and VLW[J]. Sensors, 2025, 25(24): 7629. doi: 10.3390/S25247629.
[13]	LI Songxu, XIONG Yuyong, SHEN Xiangtian, et al. Multi-scale and full-field vibration measurement via millimetre-wave sensing[J]. Mechanical Systems and Signal Processing, 2022, 177: 109178. doi: 10.1016/J.YMSSP.2022.109178.
[14]	XIONG Yuyong, LIU Zhaoyu, LI Songxu, et al. Scanning microwave vibrometer: Full-field vibration measurement via microwave sensing with phase-encoded beam scanning[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 3507611. doi: 10.1109/TIM.2023.3244835.
[15]	HE Xiaole, LIU Ping, and WANG Junling. Performance boundaries and tradeoffs in super-resolution imaging technologies for space targets[J]. Remote Sensing, 2025, 17(4): 696. doi: 10.3390/rs17040696.
[16]	程伊婷, 董涛, 苏昱玮, 等. 面向通信信号高效接收处理的压缩感知技术综述[J]. 电子与信息学报, 2026, 48(1): 168–182. doi: 10.11999/JEIT250855. CHENG Yiting, DONG Tao, SU Yuwei, et al. A review of compressed sensing technology for efficient receiving and processing of communication signal[J]. Journal of Electronics & Information Technology, 2026, 48(1): 168–182. doi: 10.11999/JEIT250855.
[17]	ARRIDGE S, MAASS P, ÖKTEM O, et al. Solving inverse problems using data-driven models[J]. Acta Numerica, 2019, 28: 1–174. doi: 10.1017/S0962492919000059.
[18]	LIU Zhaoyu, XIONG Yuyong, WU Gaoyang, et al. Super-resolution and accurate full-field displacement measurement with millimeter-wave radars[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 8507011. doi: 10.1109/TIM.2023.3327467.
[19]	TANG Junkui, LIU Zheng, RAN Lei, et al. Enhancing forward-looking image resolution: Combining low-rank and sparsity priors[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5100812. doi: 10.1109/TGRS.2023.3237332.
[20]	YUAN Yubing, JI Yicai, YE Shengbo, et al. Low-sidelobe imaging algorithm based on joint low-rank and sparse priors for small-array UWB radar[J]. IEEE Sensors Journal, 2025, 25(14): 26883–26896. doi: 10.1109/jsen.2025.3569231.
[21]	PHAM T H and HONG I P. Fast adaptive plug-and-play ADMM framework for short-range 3-D SAR imaging[J]. IEEE Transactions on Computational Imaging, 2025, 11: 764–778. doi: 10.1109/TCI.2025.3573587.
[22]	方震, 简璞, 张浩, 等. 基于FMCW雷达的非接触式医疗健康监测技术综述[J]. 雷达学报, 2022, 11(3): 499–516. doi: 10.12000/JR22019. FANG Zhen, JIAN Pu, ZHANG Hao, et al. Review of noncontact medical and health monitoring technologies based on FMCW radar[J]. Journal of Radars, 2022, 11(3): 499–516. doi: 10.12000/JR22019.
[23]	杨春玲, 梁梓文. 静态与动态域先验增强的两阶段视频压缩感知重构网络[J]. 电子与信息学报, 2024, 46(11): 4247–4258. doi: 10.11999/JEIT240295. YANG Chunling and LIANG Ziwen. Static and dynamic-domain prior enhancement two-stage video compressed sensing reconstruction network[J]. Journal of Electronics & Information Technology, 2024, 46(11): 4247–4258. doi: 10.11999/JEIT240295.
[24]	余礼苏, 钟润, 吕欣欣, 等. 压缩感知辅助的低复杂度SCMA系统优化设计[J]. 电子与信息学报, 2024, 46(5): 2011–2017. doi: 10.11999/JEIT231226. YU Lisu, ZHONG Run, LV Xinxin, et al. Optimized design of low complexity SCMA system assisted by compressed sensing[J]. Journal of Electronics & Information Technology, 2024, 46(5): 2011–2017. doi: 10.11999/JEIT231226.
[25]	李俊辉, 侯兴松. 基于伪监督注意力短期记忆与多尺度去伪影网络的图像分块压缩感知[J]. 电子与信息学报, 2024, 46(2): 472–480. doi: 10.11999/JEIT231069. LI Junhui and HOU Xingsong. Pseudo supervised attention short-term memory and multi-scale deartifacting network based on image block compressed sensing[J]. Journal of Electronics & Information Technology, 2024, 46(2): 472–480. doi: 10.11999/JEIT231069.