Page 28 - 《应用声学》2023年第2期
P. 28
216 2023 年 3 月
Journal of Oceanic Engineering, 2019, 44(1): 156–166. gression: experimental validation and comparison with
[13] Bianco M J, Gerstoft P, Traer J, et al. Machine learning in MUSIC[J]. IEEE Antennas and Wireless Propagation Let-
acoustics: theory and applications[J]. The Journal of the ters, 2007, 6(11): 379–382.
Acoustical Society of America, 2019, 146(5): 3590–3628. [21] Ozanich E, Gerstoft P, Niu H. A feedforward neural
[14] Huang H, Yang J, Huang H, et al. Deep learning for super- network for direction-of-arrival estimation[J]. The Jour-
resolution channel estimation and DOA estimation based nal of the Acoustical Society of America, 2020, 147(3):
massive MIMO system[J]. IEEE Transactions on Vehicu- 2035–2048.
lar Technology, 2018, 67(9): 8549–8560. [22] Trabelsi C, Bilaniuk O, Zhang Y, et al. Deep complex
[15] Ahmed A M, Thanthrige U, Gamal A E, et al. Deep networks[J]. arXiv Preprint, 2017, arXiv: 1705.09792.
learning for DOA estimation in MIMO radar systems via
[23] Cao Y, Lyu T, Lin Z, et al. Complex ResNet aided
emulation of large antenna arrays[J]. IEEE Communica-
DOA estimation for near-field MIMO systems[J]. IEEE
tions Letters, 2021, 29(5): 1559–1563.
Transactions on Vehicular Technology, 2020, 69(10):
[16] Jha S, Durrani T. Direction of arrival estimation using ar-
11139–11151.
tificial neural networks[J]. IEEE Transactions on Systems,
[24] Liu Z M, Zhang C, Yu P S. Direction-of-arrival estima-
Man, and Cybernetics, 1991, 21(5): 1192–1201.
tion based on deep neural networks with robustness to
[17] 李洪升, 赵俊渭, 王峰, 等. 一种基于径向基函数网络的盲
array imperfections[J]. IEEE Transactions on Antennas
波束形成方法研究 [J]. 系统工程与电子技术, 2003, 25(6):
and Propagation, 2018, 66(12): 7315–7327.
661–663, 687.
[25] Zhu W, Zhang M, Li P, et al. Two-dimensional DOA esti-
Li Hongsheng, Zhao Junwei, Wang Feng, et al. Study of a
mation via deep ensemble learning[J]. IEEE Access, 2020,
blind beamforming method based on RBFNN[J]. Journal
8: 124544–124552.
of Air Force Engineering University, 2003, 25(6): 661–663,
[26] Xiang H, Chen B, Yang M, et al. Improved direction-
687.
of-arrival estimation method based on LSTM neural net-
[18] 于斌, 尹成友, 黄冶. 阵列误差影响下的 RBF 神经网络波达
works with robustness to array imperfections[J]. Applied
方向估计 [J]. 微波学报, 2007, 23(6): 21–25, 31.
Intelligence, 2021, 51(2): 1–14.
Yu Bin, Yin Chengyou, Huang Ye. Direction of arrival
(DOA) estimation for an array with errors using RBF neu- [27] Biggs D, Andrews M. Acceleration of iterative image
ral network[J]. Journal of Microwaves, 2007, 23(6): 21–25, restoration algorithms[J]. Applied Optics, 1997, 36(8):
31. 1766–1775.
[19] Pastorino M, Randazzo A. A smart antenna system for [28] Ronneberger O, Fischer P, Brox T. U-Net: convolutional
direction of arrival estimation based on a support vector networks for biomedical image segmentation[J]. arXiv
regression [J]. IEEE Transactions on Antennas and Prop- Preprint, 2015, arXiv: 1505.04597v1.
agation, 2005, 53(7): 2161–2168. [29] Dumoulin V, Visin F. A guide to convolution arith-
[20] Randazzo A, Abou-Khousa M A, Pastorino M, et al. Di- metic for deep learning[J]. arXiv Preprint, 2016, arXiv:
rection of arrival estimation based on support vector re- 1603.07285v2.