Range estimation of few-shot underwater sound source in shallow water based on transfer learning and residual CNN

doi:10.23919/JSEE.2023.000095

Journal of Systems Engineering and Electronics ›› 2023, Vol. 34 ›› Issue (4): 839-850.doi: 10.23919/JSEE.2023.000095

• ELECTRONICS TECHNOLOGY • Previous Articles

Range estimation of few-shot underwater sound source in shallow water based on transfer learning and residual CNN

Qihai YAO¹^,²(), Yong WANG¹^,²^,*(), Yixin YANG¹^,²()

¹ School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an 710072, China
² Shaanxi Key Laboratory of Underwater Information Technology, Xi’an 710072, China

Received:2022-07-05 Accepted:2023-06-06 Online:2023-08-18 Published:2023-08-28
Contact: Yong WANG E-mail:2019260659@mail.nwpu.edu.cn;yongwang@nwpu.edu.cn;yxyang@nwpu.edu.cn
About author:
YAO Qihai was born in 1997. He received his Master ’s degree in ship and ocean engineering from Northwestern Polytechnical University in 2022. He is pursuing his Ph.D. degree at Northwestern Polytechnical University. In 2020, he completed an exchange project of the Oxford prospects program of the University of Oxford, UK, and obtained an excellent graduation certificate. His main research direction is the application of machine learning in array signal processing. E-mail: 2019260659@mail.nwpu.edu.cn

WANG Yong was born in 1987. He received his B.S. degree in applied electronic engineering and M.S. and Ph.D. degrees in underwater acoustic engineering from Northwestern Polytechnical University (NPU), Xi ’an, China, in 2009, 2011, and 2015, respectively. From 2016 to 2017, he was a post-doctoral fellow with Xi ’an Jiaotong University, Xi’an, China. He is currently an associate professor with the School of Marine Science and Technology, NPU. His research interests include array signal processing, parameter estimation, and sonar signal processing. E-mail: yongwang@nwpu.edu.cn

YANG Yixin was born in 1975. He received his B.S. degree in applied electronic engineering and M.S. and Ph.D. degrees in underwater acoustic engineering from Northwestern Polytechnical University (NPU), Xi ’an, China, in 1997, 1999, and 2002, respectively. From June 2002 to June 2004, he was a research fellow with the School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore. Since July 2004, he has been with the School of Marine Science and Technology, NPU, where he became a professor in 2006. He is currently the Vice President of NPU, the Chair of the Underwater Acoustics Committee of the Acoustical Society of China, and a member of the Acoustical Society of America. His research interests include acoustic array signal processing, spectral estimation, and their applications. E-mail: yxyang@nwpu.edu.cn
Supported by:
This work was supported by the National Natural Science Foundation of China (11974286;11904274) and the Shaanxi Young Science and Technology Star Program (2021KJXX-07), and the fundamental research funding for characteristic disciplines (G2022WD0235)

Abstract

Abstract:

Taking the real part and the imaginary part of complex sound pressure of the sound field as features, a transfer learning model is constructed. Based on the pre-training of a large amount of underwater acoustic data in the preselected sea area using the convolutional neural network (CNN), the few-shot underwater acoustic data in the test sea area are retrained to study the underwater sound source ranging problem. The S5 voyage data of SWellEX-96 experiment is used to verify the proposed method, realize the range estimation for the shallow source in the experiment, and compare the range estimation performance of the underwater target sound source of four methods: matched field processing (MFP), generalized regression neural network (GRNN), traditional CNN, and transfer learning. Experimental data processing results show that the transfer learning model based on residual CNN can effectively realize range estimation in few-shot scenes, and the estimation performance is remarkably better than that of other methods.

Key words: transfer learning, residual convolutional neural network (CNN), few shot, vertical array, range estimation

Qihai YAO, Yong WANG, Yixin YANG. Range estimation of few-shot underwater sound source in shallow water based on transfer learning and residual CNN[J]. Journal of Systems Engineering and Electronics, 2023, 34(4): 839-850.

Figures/Tables 14

Fig 1

Fig 2

Fig 3

Fig 4

Fig 5

Fig 6

Fig 7

Fig 8

Fig 9

Fig 10

Fig 11

Fig 12

Table 1

Table 2

References 31

1	BUCKER H P Use of calculated sound fields and matched-field detection to locate sound sources in shallow water. The Journal of the Acoustical Society of America, 1976, 59 (2): 368- 373. doi: 10.1121/1.380872
2	KLEMM R Range and depth estimation by line arrays in shallow water. Signal Processing, 1981, 3 (4): 333- 344. doi: 10.1016/0165-1684(81)90003-7
3	BAGGEROER A B Matched field processing: source localization in correlated noise as an optimum parameter estimation problem. The Journal of the Acoustical Society of America, 1988, 83 (2): 571- 587. doi: 10.1121/1.396151
4	MICHALOPOULOU Z H, PORTER M B Matched-field processing for broad-band source localization. IEEE Journal of Oceanic Engineering, 1996, 21 (4): 384- 392. doi: 10.1109/48.544049
5	SOARES C, JESUS S M Broadband matched-field processing: coherent and incoherent approaches. The Journal of the Acoustical Society of America, 2003, 113 (5): 2587- 2598. doi: 10.1121/1.1564016
6	YANG K D, MA Y L, ZOU S X, et al Linear matched field processing based on environmental perturbation. Acta Acustica, 2006, 31 (6): 496- 505.
7	STEINBERG B Z, BERAN M J, CHIN S H, et al A neural network approach to source localization. The Journal of the Acoustical Society of America, 1991, 90 (4): 2081- 2090. doi: 10.1121/1.401635
8	NIU H Q, OZANICH E, GERSTOFT P Ship localization in Santa Barbara Channel using machine learning classifiers. The Journal of the Acoustical Society of America, 2017, 142 (5): EL455- EL460. doi: 10.1121/1.5010064
9	WANG Y, PENG H Underwater acoustic source localization using generalized regression neural network. The Journal of the Acoustical Society of America, 2018, 143 (4): 2321- 2331. doi: 10.1121/1.5032311
10	FERGUSON E L, WILLIAMS S B, JIN C T. Sound source localization in a multipath environment using convolutional neural networks. Proc. of the IEEE International Conference on Acoustics, Speech and Signal Processing, 2018: 2386−2390.
11	LIU Y N, NIU H Q, LI Z L Source ranging using ensemble convolutional networks in the direct zone of deep water. Chinese Physics Letters, 2019, 36 (4): 044302. doi: 10.1088/0256-307X/36/4/044302
12	NIU H Q, GONG Z X, OZANICH E, et al Deep-learning source localization using multi-frequency magnitude-only data. The Journal of the Acoustical Society of America, 2019, 146 (1): 211- 222. doi: 10.1121/1.5116016
13	HOWARTH K, KOMEN D, NEILSEN T B, et al Effect of signal to noise ratio on a convolutional neural network for source ranging and environmental classification. The Journal of the Acoustical Society of America, 2019, 146 (4): 2961- 2962.
14	KOMEN D, NEILSEN T B, KNOBLES D P, et al A convolutional neural network applied to measured time series for source range and ocean seabed classification. The Journal of the Acoustical Society of America, 2019, 146 (4): 2930.
15	KOMEN D, NEILSEN T B, HOWARTH K, et al Seabed and range estimation of impulsive time series using a convolutional neural network. The Journal of the Acoustical Society of America, 2020, 147 (5): EL403- EL408. doi: 10.1121/10.0001216
16	OZANICH E, GERSTOFT P, NIU H Q A feedforward neural network for direction-of-arrival estimation. The Journal of the Acoustical Society of America, 2020, 147 (3): 2035- 2048. doi: 10.1121/10.0000944
17	LIU Y N, NIU H Q, LI Z L A multi-task learning convolutional neural network for source localization in deep occan. The Journal of the Acoustical Society of America, 2020, 148 (2): 873- 883. doi: 10.1121/10.0001762
18	CHEN R, SCHMIDT H Model-based convolutional neural network approach to underwater source-range estimation. The Journal of the Acoustical Society of America, 2021, 149 (1): 405- 420. doi: 10.1121/10.0003329
19	WEISS K, KHOSHGOFTAAR T M, WANG D D A survey of transfer learning. Journal of Big Data, 2016, 3 (1): 1- 40. doi: 10.1186/s40537-015-0036-x
20	DAI W Y, CHEN Y Q, XUE G R, et al. Translated learning: transfer learning across different feature spaces. Proc. of the Advances in Neural Information Processing Systems, 2008: 353−360.
21	ZHU Y, CHEN Y Q, LU Z Q, et al. Heterogeneous transfer learning for image classification. Proc. of the 25th Association for the Advancement of Artificial Intelligence Conference on Artificial Intelligence, 2011: 1304−1309.
22	NAKAMURA A, HARADA T. Revisiting fine-tuning for few-shot learning. https://arxiv.org/abs/1910.00216.
23	JANG Y H, LEE H K, HWANG S J, et al. Learning what and where to transfer. Proc. of the International Conference on Machine Learning, 2019: 3030–3039.
24	GERACE F, SAGLIETTU L, MANNELLI S S, et al Probing transfer learning with a model of synthetic correlated datasets. Machine Learning: Science and Technology, 2022, 3 (1): 015030. doi: 10.1088/2632-2153/ac4f3f
25	WANG W B, NI H Y, SU L, et al Deep transfer learning for source ranging: deep-sea experiment results. The Journal of the Acoustical Society of America, 2019, 146 (4): EL317- EL322. doi: 10.1121/1.5126923
26	CAO H G, WANG W B, SU L, et al Deep transfer learning for underwater direction of arrival using one vector sensor. The Journal of the Acoustical Society of America, 2021, 149 (3): 1699- 1711. doi: 10.1121/10.0003645
27	PORTER M B. The KRAKEN normal mode program. Washington D.C.: Naval Research Lab, 1992.
28	YANG K D. The matched field processing of underwater acoustic array signals. Xi’an: Northwestern Polytechnical University Press, 2008. (in Chinese)
29	FUKUSHIMA K, MIYAKE S, ITO T Neocognitron: a neural network model for a mechanism of visual pattern recognition. IEEE Trans. on Systems, Man, and Cybernetics, 1982, 13 (5): 826- 834.
30	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition. Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770−778.
31	MUARRY J, ENSBERG D. The swellex-96 experiment. http://www.swellex96.ucsd.edu/.

Method	MFP	GRNN			CNN			Transfer learning
Method	MFP	T1	T2	T1/T2	T1	T2	T1/T2	Transfer learning
Narrowband	55.6925	19.0144	12.9857	10.3732	17.5382	7.5285	5.8026	4.0886
Broadband	20.2007	17.4170	10.3911	8.1141	16.7749	6.9176	5.0352	2.9136

Method	GRNN			CNN			Transfer learning
Method	T1	T2	T1/T2	T1	T2	T1/T2	Transfer learning
Training time/s	20.5	9.7	29.2	50.8	32.9	89.3	95.2
Testing time/s	2.3	2.1	2.4	2.8	2.5	3.1	2.9

Range estimation of few-shot underwater sound source in shallow water based on transfer learning and residual CNN

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

Share this article

Figures/Tables 14

References 31

Related Articles 1

Recommended Articles

Metrics

Comments