在阵列信号处理快速发展的当下，移动阵列是一项关键技术，它通过波达方向估计 （Direction of Arrival，DOA）以及波束形成技术精确估计目标的来波角度，为雷达、通 信等应用领域提供技术支撑。由于移动阵列能够灵活调整位置和姿态，灵活适应复杂的 动态环境，成为近年来的研究重点。然而，现有的传统波束形成估计算法和多重信号分 类（Multiple Signal Classification，MUSIC）算法在平台运动场景中，因DOA时变出现 测量误差以及协方差矩阵发生时变，从而导致精度下降的问题。为此本文研究了基于移 动阵列对静止目标的DOA估计方法，主要研究工作包括以下几个方面： （1）针对平台在运动场景下因DOA时变导致目标信号出现测量误差的问题，本文 提出了一种基于快速傅里叶变换的长时间时变 DOA 精确估计。该方法首先对常规波束 形成技术进行相位补偿，然后补偿长时间接收信号的 DOA 时变量，消除线性波达方向 变化，通过方向图叠加搜索最大谱峰，得到准确的 DOA 估计。最后在不同信噪比、快 拍数、辐射源数量的条件下进行蒙特卡洛实验，统计不同算法的均方根误差。仿真结果 表明，本文提出的补偿方向图叠加算法相比其他三种 DOA 估计算法，在低信噪比下定 位精度提升了30%左右。 （2）针对平台在运动场景下因时变DOA导致目标信号出现精度下降的问题，本文 进一步提出了一种基于旋转阵列的增强型 MUSIC 算法。该方法首先对阵列接收到的信 号进行预处理，将旋转周期划分为多个时间片，构建扩展协方差矩阵，然后引入角速度 总变分约束，消除运动误差，抑制运动突变引起的伪谱抖动，最后通过谱峰搜索得到准 确 DOA。在不同信噪比、快拍数、辐射源数量的条件下进行蒙特卡洛实验，仿真结果 表明，本文提出的增强型 MUSIC 算法相比其他三种 DOA 估计算法在定位精度方面提 升约20%~30%。 关 键 词：移动阵列；DOA估计；波束形成；多重信号分类； 研究类型：理论研究

At present, array signal processing is developing rapidly. The mobile array is a key technology. It precisely estimates the incoming wave angle through direction of arrival (DOA) estimation and beamforming technology, offering technical support to fields like radar and communication. In recent years, due to its flexible position and posture adjustment and its adaptation to complex dynamic environments, the mobile array has become a research focus. Traditional beamforming estimation algorithms often see accuracy drop in array movement scenarios. Measurement errors occur due to the time varying direction of arrival (DOA). covariance matrices from the multiple signal classification (MUSIC) algorithm also become time-varying. To address this issue, this thesis explores DOA estimation methods for stationary targets based on mobile arrays. The major research efforts fall into the following aspects: (1)In scenarios where the platform is in motion, the time varying DOA can cause measurement errors in target signals. To address this, this thesis proposes a long term time varying DOA precise estimation method based on the fast Fourier transform (FFT). The method first compensates the phase of conventional beamforming technology. Then, it offsets the long term time varying DOA of the received signals to eliminate linear DOA changes. Subsequently, it conducts direction - finding through pattern - overlay search for the maximum spectral peak to achieve accurate DOA estimation.Monte Carlo experiments are conducted under various conditions of signal-to-noise ratio (SNR), snapshot number, and number of radiation sources. The results show that the proposed compensated pattern superposition algorithm in this paper improves the positioning accuracy by about 30% compared with the other three DOA estimation algorithms under low SNR conditions. (2) In scenarios where the platform is in motion, the long-term time - varying DOA can lead to a reduction in the accuracy of target signals. To address this, this thesis puts forward an enhanced MUSIC algorithm based on a rotating array. Initially, the received signals are preprocessed, dividing the rotation cycle into multiple time slots to construct an extended covariance matrix. Then, angular velocity total variation constraints are introduced to eliminate motion errors and suppress pseudo - spectrum jitter caused by abrupt movements. Accurate DOA is obtained via spectrum peak search.Monte Carlo experiments are conducted under different SNR, snapshot number, and radiation source number conditions. The results show that the proposed enhanced MUSIC algorithm improves positioning accuracy by 20% - 30% over the other three DOA estimation algorithms. Key words: Mobile Array; Direction of Arrival estimation; Beamforming; Multiple Signal Classification Thesis : Theoretical Research

[1] SYEDA R Z, SAVELYEV T G, van BEURDEN M C, et al. Sparse MIMO Array for

Improved 3D mm- Wave Imaging Radar: 2020 17th European Radar Conference

(EuRAD)[C]: EuMA, 2021.

[2] 彭木根,刘喜庆,刘子乐,等.6G 通信感知一体化理论与技术[J].控制与决策,2023,38(1):22-38.

[3] Guo Y J, Ziolkowski R W. Advanced antenna array engineering for 6G and beyond wireless

communications[M]. John Wiley & Sons, 2021.

[4] Qin G ,Zhang D Y ,Amin G M .DOA Estimation Exploiting Moving Dilated Nested

Arrays.[J].IEEE Signal Process. Lett,2019,26(3):490-494.

[5] Montaruli M F, Facchini L, Di Lizia P, et al. Adaptive track estimation on a radar array

system for space surveillance[J]. Acta Astronautica, 2022, 198: 111-123.

[6] Bakhuraisa Y, Aziz A A, Geok T K, et al. A survey of ranging techniques for vehicle

localization in intelligence transportation system: challenges and opportunities[J]. Int. J.

Electr. Comput. Eng, 2022, 12: 6248-6260.

[7] KIRKEBO J E, AUSTENG A. Amplitude Tolerance Analysis of Curved Sonar arrays:

OCEANS 2021: San Diego–Porto[C]: MTS, 2021.

[8] ZHOU L, YE K, QI J, et al. DOA Estimation Based on Pseudo-Noise Subspace for Relocating Enhanced Nested Array[J]. IEEE Signal Processing Letters, 2022, 29: 1858

1862.

[9] Li J, Stoica P, Wang Z. On robust Capon beamforming and diagonal loading[J]. IEEE

transactions on signal processing, 2003, 51(7): 1702-1715.

[10] ZHANG M, CHEN X, ZHANG A. A simple tridiagonal loading method for robust

adaptivebeamforming[J]. Signal Processing, 2019, 157: 103-107.

[11] 徐常伟,孙希超,罗选,等.一种改进的 MUSIC 算法在干扰源定位中的应用[J].舰船电子

工程,2024,44(07):186-189.

[12] 陆虹宇,李晓曼,陈鸿运,等.不同 DOA 算法在水声信号中的应用对比分析[J].云南民族

大学学报(自然科学版),2024,33(04):486-495.

[13] FENG M, CUI Z, YANG Y, et al. A Reduced-Dimension MUSIC Algorithm for Monostatic

FDA-MIMO Radar[J]. IEEE Communications Letters, 2021, 25(4): 1279-1282.

[14] 李晓璇,孙闽红.降维长矢量 MUSIC 算法的高速并行 FPGA 实现[J].探测与控制学

报,2024,46(05):50-56.

58

参考文献

[15] Haardt, M and Nossek, J. A . Unitary ESPRIT: How to Obtain Increased Estimation

Accuracy with AReducedComputational Burden[J]. IEEE Trans. Signal Processing, 1995,

43(5): P.1232–1242.

[16] 刘帅,许媛媛,闫锋刚,等.泰勒展开与交替投影最大似然结合的离网格 DOA 估计算法

[J].电子与信息学报,2024,46(08):3219-3227.

[17] 窦慧晶,谢东旭,郭威,等.基于改进的正交匹配跟踪算法的波达方向估计[J].吉林大学

学报(工学版),2024,54(12):3568-3576.

[18] 杨义芳,王金平.块稀疏信号重构的广义正交匹配追踪算法研究[J].武汉大学学报(理学

版),2024,70(06):680-686.

[19] 赵卫鹏,毛锦,刘凯,等.分段取阈值正交匹配追踪反卷积声源识别算法[J].振动与冲

击,2023,42(15):268-276.

[20] 郭艺夺,宫健,高乾,等.基于改进 FOCUSS 算法的距离模糊杂波抑制方法[J].现代雷

达,2020,42(09):32-38.

[21] 孙长伟,成谢锋.混响环境下改进的球谐波域 L1-SVD 声源定位算法[J].武汉大学学报

(理学版),2018,64(05):459-470.

[22] 郭业才,田佳佳,胡国乐.基于稀疏贝叶斯学习在未知噪声场的欠定宽带信号DOA估计

[J].实验室研究与探索,2021,40(06):5-10.

[23] 王燕,赵磊,郝宇等.观测平台转向时稀疏贝叶斯学习方位估计[J].声学学报,2022.

[24] Sheinvald J, Wax M, Meiss A J. Localization of multiple sources with moving arrays[J].

IEEE transactions on signal processing, 1998, 46(10): 2736-2743.

[25] Ai W, Huang Y, Zhang S. New results on Hermitian matrix rank-one decomposition[J].

Mathematical programming, 2011, 128(1): 253-283.

[26] Friedlander B, Zeira A. Eigenstructure-based algorithms for direction finding with time-

varying arrays[J]. IEEE transactions on aerospace and electronic systems, 1996, 32(2): 689

701.

[27] Bilik I. Spatial compressive sensing for direction-of-arrival estimation of multiple sources

using dynamic sensor arrays[J]. IEEE Transactions on Aerospace and Electronic Systems,

2011, 47(3): 1754-1769.

[28] 张德明,郭良浩,张仁和.基于聚焦变换的宽带信号方位稳健估计[J].声学学

报,2005,30(4):303-308.

[29] 朱维杰,孙进才.基于快速富氏变换插值的宽带信号方位估计[J].声学学

报,2002,27(6):513-517.

[30] Hu Y, Lu J, Qiu X. Direction of arrival estimation of multiple acoustic sources using a

maximum likelihood method in the spherical harmonic domain[J]. Applied Acoustics, 2018,

59

西安科技大学全日制专业硕士学位论文

135: 85-90.

[31] Meng Z, Shen F, Zhou W, et al. Robust adaptive beamforming based on subspace method[J].

Journal of Electromagnetic Waves and Applications, 2018, 32(11): 1369-1378.

[32] Bilik I. Spatial compressive sensing for direction-of-arrival estimation of multiple sources

using dynamic sensor arrays[J]. IEEE Transactions on Aerospace and Electronic Systems,

2011, 47(3): 1754-1769.

[33] 冯成,龚晓峰,雒瑞森.基于 FFT 和黄金分割的快速 DOA 估计[J].计算机仿

真,2021,38(09):159-163+172.

[34] 刘姿琪,李吉祥,李伟,等.基于函数波束形成的传声器阵列声成像性能分析[J].信息技

术与信息化,2024(01):135-140.

[35] Dou Y ,Sun G ,Wang Y,et al.High-Precision DOA Estimation Method Based on Synthetic

Aperture Technique[J].Journal of Beijing Institute of Technology,2024,33(02):111-118.

[36] 吴鑫炜,胡明华,毛继志,等.基于空间谱估计的无人机协同目标方位感知算法[J].西南

交通大学学报,2024,59(04):898-906+932.

[37] 曹泽坤,怀园园,张昕,等.基于 GPU 加速的双向平滑MUSIC相干信号DOA估计算法

[J].火控雷达技术,2024,53(04):31-36+41.

[38] 张海瑛,崔宇晗,陆东,等.一种基于接收数据重构的 DOA 估计技术[J].无线电通信技

术,2024,50(06):1223-1228.

[39] 孙思琦,郝琨,李成.多信息融合的水下传感器阵列网络移动节点定位[J].传感器与微系

统,2022,41(07):143-146+151.

[40] 唐灿,张波,李风华.移动强干扰条件下大孔径阵列稀疏空间谱估计[J].声学学

报,2024,49(04):696-708.

[41] 朱根村,魏爽.基于载波变频稀疏采样与阵列移动的 TOA-DOA 联合估计[J].中国电子

科学研究院学报,2023,18(12):1079-1085.

[42] 林钊希,赵宏志.面向大容量海洋移动通信的阵列级全双工自干扰抑制[J].移动通

信,2024,48(11):92-97.

[43] Lebret H, Boyd S. Antenna array pattern synthesis via convex optimization[J]. IEEE

transactions on signal processing,1997,45(3):526-532.

[44] 邓欣,熊林森,董志飞,等.用于移动机器人听觉导航的光纤麦克风阵列研制[J].光学与

光电技术,2023,21(01):118-128.

[45] 方晓强,刘元庆,张啸天,等.基于移动麦克风阵列的换流站异常声源定位方法[J].浙江

电力,2024,43(01):108-116.

[46] 李铭時,杨佳敏,冀贞海,等.移动三平行互质阵列改进设计方法[J].航天电子对

抗,2024,40(04):52-57+64.DOI:10.16328/j.htdz8511.2024.04.003.

60

参考文献

[47] 韦进强.一种基于前向空间平滑的分组相干信号测向算法[J].中国科技信

息,2024,(19):131-134.

[48] Buckley K, Griffiths L. An adaptive generalized sidelobe canceller with derivative

constraints[J]. IEEE Transactions on antennas and propagation, 1986, 34(3): 311-319.

[49] Yan S, Ma Y. Robust supergain beamforming for circular array via second-order cone

programming[J]. Applied Acoustics, 2005, 66(9): 1018-1032.

[50] LI J, HE Y, MA P, et al. Direction of Arrival Estimation Using Sparse Nested Arrays With

Coprime Displacement[J]. IEEE Sensors Journal, 2021, 21(4): 5282-5291.

[51] LI S, ZHANG X, RUAN D, et al. Increasing the Uniform Degrees of Freedom for Moving

q-Dilated Arrays[J]. IEEE Internet of Things Journal, 2023, 10(1): 194-209.

[52] SUN L. Improved Coprime Array Designs Based on Moving Platform for Direction of

Arrival Estimation[J]. IEEE Sensors Journal, 2023, 23(5): 5288-5298.

[53] QIN G, AMIN M G, ZHANG Y D. DOA Estimation Exploiting Sparse Array Motions[J].

IEEE Transactions on Signal Processing, 2019, 67(11): 3013-3027.

[54] LI S, ZHANG X. Dilated Arrays: A Family of Sparse Arrays with Increased Uniform

Degrees of Freedom and Reduced Mutual Coupling on a Moving Platform[J]. IEEE

Transactions on Signal Processing, 2021, 69: 3367-3382.

[55] Li J, et al. Rotating array DOA estimation with motion compensation[J]. IEEE Sensors

Journal, 2019.

[56] Tao H, Hickman G, Krolik J. Field directionality synthesis from multiple array orientations:

A least squares approach[C]. OCEANS 2008. IEEE, 2008: 1-6.

[57] 全鹏.稳健波束形成与阵列旋转时的方位估计技术研究[D].哈尔滨工程大学,2024.

[58] He Q, et al. Time-varying DOA estimation using Kalman filter[JJ. Signal Processing, 2021.

[59] Liu Z, et al. Temporal smoothing for rotating arrays[C]. ICASSP 2020.

[60] 马健钧.阵列雷达复杂场景高分辨 DOA 估计技术研究[D].西安电子科技大

学,2023.DOI:10.27389/d.cnki.gxadu.2023.003852.