匹配场处理是水声被动定位领域的常用算法，它根据水域环境条件对水下声场进行建模，进而获取搜索区域中每一个位置到阵列之间的拷贝场向量，然后通过设计好的处理器将阵列接收数据和拷贝场向量进行匹配，即可实现水下目标的被动定位。声源位置与拷贝场向量之间形成一种非线性的映射关系，深度学习模型具有强大的非线性学习能力，因此非常适用于学习这种映射关系。本文提取由拷贝场向量组成的图片特征作为深度学习模型的训练样本，将水下目标被动定位问题转换为分类问题。此外，为了避免小样本容易导致网络过拟合的问题，利用预训练网络Res-net18作为模型对数据进行迁移学习。在水声仿真定位试验中，通过与常规匹配场处理和基于稀疏表征的匹配场处理方法相比，验证了本文方法具有更好的定位能力。
Matching field processing is a common algorithm in the field of underwater acoustic passive positioning. It models the underwater acoustic field according to the water environment conditions, and then obtains the copy field vector between each position in the search area and the array. Then, the data received by the array and the copy field vector are matched by the designed processor to realize the passive positioning of underwater targets. There is a nonlinear mapping relationship between the source position and the copy field vector, and the deep learning model has strong nonlinear learning ability, so it is very suitable for learning this mapping relationship. In this paper, image features composed of copy field vectors are extracted as training samples of deep learning models, and the passive location problem of underwater targets is transformed into a classification problem. In addition, in order to avoid the problem of network overfitting caused by small samples, the pre-trained network Res-net18 is used as the model for data transfer learning. Compared with conventional matching field processing and sparse representation based matching field processing, the proposed method is proved to have better locating ability in underwater acoustic simulation positioning experiments.
2025,47(20): 151-155 收稿日期：2024-11-20
DOI：10.3404/j.issn.1672-7649.2025.20.023
分类号：U666.7；TB566
基金项目：中国船舶集团有限公司第七二六研究所单位自筹项目（24-5323-1019）
作者简介：王豪(1996-),男,博士,工程师,研究方向为水声信号处理
参考文献：
[1] BAGGEROER A. B, KUPERMAN W. A. An overview of matched field methods in ocean acoustics[J]. IEEE Journal of Oceanic Engineering, 1993, 18(4): 401-424.
[2] LI N, HAN D, ZHANG S C, et al. The MVDR processor in underwater acoustic source location method based on modal amplitude correlation[C]//International Conference on Signal Processing, Computer Networks, and Communications (SPCNC 2022), 2023.
[3] JOMON G, JOJISH J V, SANTHANAKRISHNAN T. MVDR beamformer with subband peak energy detector for detection and tracking of fast moving underwater targets using towed array sonars[J]. Acta Acustica united with Acustica, 2019, 105(1): 220-225.
[4] LIANG G L, FU J, ZHANG K, et al. Modified MVDR algorithm for DOA estimation using acoustic vector hydrophone[C]//2011 IEEE International Conference on Computer Science and Automation Engineering, 2011.
[5] WONG K T, ZOLTOWSKI M D. Self-initiating MUSIC-based direction finding in underwater acoustic particle velocity-field beamspace[J]. IEEE Journal of Oceanic Engineering, 2000, 25(2): 262-273.
[6] LAN C F, CHEN H, ZHANG L, et al. Underwater acoustic DOA estimation of incoherent signal based on improved GA-MUSIC[J]. IEEE Access, 2023: 69474-69485.
[7] ISHI C T, CHATOT O, ISHIGURO H, et al. Evaluation of a MUSIC-based real-time sound localization of multiple sound sources in real noisy environments[C]//2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2009.
[8] LIANG J L, LIU D. Passive localization of mixed near-field and far-field sources using two-stage MUSIC algorithm[J]. IEEE Transactions on Signal Processing, 2009, 58(1): 108-120.
[9] BERGEAUD F, MALLAT S. Matching pursuit of images[C]// Proceedings., International Conference on Image Processing, 1995.
[10] LAURA R N, DAVID L. Optimized orthogonal matching pursuit approach[J]. IEEE Signal Processing Letters, 2002, 9(4): 137-140.
[11] WANG J, KWON S, SHIM B. Generalized orthogonal matching pursuit[J]. IEEE Transactions on signal processing, 2012, 60(12): 6202-6216.
[12] CAI T T, WANG L. Orthogonal matching pursuit for sparse signal recovery with noise[J]. IEEE Transactions on Information theory, 2011, 57(7): 4680-4688.
[13] SHI J, YANG D S, SHI S G, et al. Compressive focused beamforming based on vector sensor array[J]. Acta Phys, 2016, 65(2): 194-204.
[14] DAVID P WIPF, BHASKAR D RAO. Sparse bayesian learning for basis selection[J]. IEEE Transactions on Signal Processing, 2004, 52(8): 2153-2164.
[15] GEMBA K L, NANNURU S, GERSTOFT P. Robust ocean acoustic localization with sparse bayesian learning[J]. IEEE Journal of Selected Topics in Signal Processing, 2019, 13(1): 49-60.
[16] LIU Y N, NIU H Q, YANG S S, et al. Multiple source localization using learning-based sparse estimation in deep ocean[J]. The Journal of the Acoustical Society of America, 2021, 150(5): 3773-3786.
[17] 孙霆, 王威, 高婧洁, 等. 传感器运动效应下水声信号传播速度未知的TOA/TOA-AOA水下无源定位方法[J]. 电子学报, 2025, 53(2): 409-419.
[18] 余杰, 王平波, 周彬, 蔡志明. 一种基于冲激脉冲回波检测的主动目标定位技术[J]. 舰船科学技术, 2024, 46(10): 120-125.
[19] NIU H Q, OZANICH E, GERSTOFT P. Ship localization in santa barbara channel using machine learning classifiers[J]. The journal of the acoustical society of America, 2017, 142(5): EL455-EL460.
[20] NIU H Q, REEVES E, GERSTOFT P. Source localization in an ocean waveguide using supervised machine learning[J]. The Journal of the Acoustical Society of America, 2017, 142(3): 1176-1188.
[21] WANG Y, PENG H. Underwater acoustic source localization using generalized regression neural network[J]. The Journal of the Acoustical Society of America, 2018, 143(4): 2321-2331.
[22] LIU Y N, NIU H Q, LI Z L. Source ranging using ensemble convolutional networks in the direct zone of deep water[J]. Chinese Physics Letters, 2019, 36(4): 044302.
[23] 张嘉平. 基于卷积神经网络和迁移学习的水下声源定位算法研究[D]. 杭州: 浙江大学, 2021.
