Deep learning-based DOA estimation using CRNN for underwater acoustic arrays
In the marine environment, estimating the direction of arrival (DOA) is challenging because of the multipath signals and low signal-to-noise ratio (SNR). In this paper, we propose a convolutional recurrent neural network (CRNN)-based method for underwater DOA estimation using an acoustic array. The...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-11-01
|
| Series: | Frontiers in Marine Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmars.2022.1027830/full |
| _version_ | 1828336039014432768 |
|---|---|
| author | Xiaoqiang Li Jianfeng Chen Jisheng Bai Muhammad Saad Ayub Dongzhe Zhang Mou Wang Qingli Yan |
| author_facet | Xiaoqiang Li Jianfeng Chen Jisheng Bai Muhammad Saad Ayub Dongzhe Zhang Mou Wang Qingli Yan |
| author_sort | Xiaoqiang Li |
| collection | DOAJ |
| description | In the marine environment, estimating the direction of arrival (DOA) is challenging because of the multipath signals and low signal-to-noise ratio (SNR). In this paper, we propose a convolutional recurrent neural network (CRNN)-based method for underwater DOA estimation using an acoustic array. The proposed CRNN takes the phase component of the short-time Fourier transform of the array signals as the input feature. The convolutional part of the CRNN extracts high-level features, while the recurrent component captures the temporal dependencies of the features. Moreover, we introduce a residual connection to further improve the performance of DOA estimation. We train the CRNN with multipath signals generated by the BELLHOP model and a uniform line array. Experimental results show that the proposed CRNN yields high-accuracy DOA estimation at different SNR levels, significantly outperforming existing methods. The proposed CRNN also exhibits a relatively short processing time for DOA estimation, extending its applicability. |
| first_indexed | 2024-04-13T21:55:24Z |
| format | Article |
| id | doaj.art-1fef786e206d4bc4ac989370aa2085cc |
| institution | Directory Open Access Journal |
| issn | 2296-7745 |
| language | English |
| last_indexed | 2024-04-13T21:55:24Z |
| publishDate | 2022-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Marine Science |
| spelling | doaj.art-1fef786e206d4bc4ac989370aa2085cc2022-12-22T02:28:17ZengFrontiers Media S.A.Frontiers in Marine Science2296-77452022-11-01910.3389/fmars.2022.10278301027830Deep learning-based DOA estimation using CRNN for underwater acoustic arraysXiaoqiang Li0Jianfeng Chen1Jisheng Bai2Muhammad Saad Ayub3Dongzhe Zhang4Mou Wang5Qingli Yan6Joint Laboratory of Environmental Sound Sensing, School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an, ChinaJoint Laboratory of Environmental Sound Sensing, School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an, ChinaJoint Laboratory of Environmental Sound Sensing, School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an, ChinaJoint Laboratory of Environmental Sound Sensing, School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an, ChinaJoint Laboratory of Environmental Sound Sensing, School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an, ChinaJoint Laboratory of Environmental Sound Sensing, School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an, ChinaSchool of Computer Science and Technology, Xi’an University of Posts & Telecommunications, Xi’an, ChinaIn the marine environment, estimating the direction of arrival (DOA) is challenging because of the multipath signals and low signal-to-noise ratio (SNR). In this paper, we propose a convolutional recurrent neural network (CRNN)-based method for underwater DOA estimation using an acoustic array. The proposed CRNN takes the phase component of the short-time Fourier transform of the array signals as the input feature. The convolutional part of the CRNN extracts high-level features, while the recurrent component captures the temporal dependencies of the features. Moreover, we introduce a residual connection to further improve the performance of DOA estimation. We train the CRNN with multipath signals generated by the BELLHOP model and a uniform line array. Experimental results show that the proposed CRNN yields high-accuracy DOA estimation at different SNR levels, significantly outperforming existing methods. The proposed CRNN also exhibits a relatively short processing time for DOA estimation, extending its applicability.https://www.frontiersin.org/articles/10.3389/fmars.2022.1027830/fullDOA estimationarray signal processingunderwater acousticconvolutional recurrent neural networkdeep learning |
| spellingShingle | Xiaoqiang Li Jianfeng Chen Jisheng Bai Muhammad Saad Ayub Dongzhe Zhang Mou Wang Qingli Yan Deep learning-based DOA estimation using CRNN for underwater acoustic arrays Frontiers in Marine Science DOA estimation array signal processing underwater acoustic convolutional recurrent neural network deep learning |
| title | Deep learning-based DOA estimation using CRNN for underwater acoustic arrays |
| title_full | Deep learning-based DOA estimation using CRNN for underwater acoustic arrays |
| title_fullStr | Deep learning-based DOA estimation using CRNN for underwater acoustic arrays |
| title_full_unstemmed | Deep learning-based DOA estimation using CRNN for underwater acoustic arrays |
| title_short | Deep learning-based DOA estimation using CRNN for underwater acoustic arrays |
| title_sort | deep learning based doa estimation using crnn for underwater acoustic arrays |
| topic | DOA estimation array signal processing underwater acoustic convolutional recurrent neural network deep learning |
| url | https://www.frontiersin.org/articles/10.3389/fmars.2022.1027830/full |
| work_keys_str_mv | AT xiaoqiangli deeplearningbaseddoaestimationusingcrnnforunderwateracousticarrays AT jianfengchen deeplearningbaseddoaestimationusingcrnnforunderwateracousticarrays AT jishengbai deeplearningbaseddoaestimationusingcrnnforunderwateracousticarrays AT muhammadsaadayub deeplearningbaseddoaestimationusingcrnnforunderwateracousticarrays AT dongzhezhang deeplearningbaseddoaestimationusingcrnnforunderwateracousticarrays AT mouwang deeplearningbaseddoaestimationusingcrnnforunderwateracousticarrays AT qingliyan deeplearningbaseddoaestimationusingcrnnforunderwateracousticarrays |