Deep Mutual Learning-Based Mode Recognition of Orbital Angular Momentum
Due to its orbital angular momentum (OAM), optical vortex has been widely used in communications and LIDAR target detection. The OAM mode recognition based on deep learning is mostly based on the basic convolutional neural network. To ensure high-precision OAM state detection, a deeper network struc...
Published in: | Photonics |
---|---|
Main Authors: | , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
MDPI AG
2023
|
Subjects: | |
Online Access: | https://doi.org/10.3390/photonics10121357 https://doaj.org/article/8a89f8e66f844d6482612652531e8908 |
id |
ftdoajarticles:oai:doaj.org/article:8a89f8e66f844d6482612652531e8908 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:8a89f8e66f844d6482612652531e8908 2024-01-21T10:05:45+01:00 Deep Mutual Learning-Based Mode Recognition of Orbital Angular Momentum Tan Qu Zhiming Zhao Yan Zhang Jiaji Wu Zhensen Wu 2023-12-01T00:00:00Z https://doi.org/10.3390/photonics10121357 https://doaj.org/article/8a89f8e66f844d6482612652531e8908 EN eng MDPI AG https://www.mdpi.com/2304-6732/10/12/1357 https://doaj.org/toc/2304-6732 doi:10.3390/photonics10121357 2304-6732 https://doaj.org/article/8a89f8e66f844d6482612652531e8908 Photonics, Vol 10, Iss 12, p 1357 (2023) deep mutual learning orbital angular momentum mode recognition knowledge distillation Applied optics. Photonics TA1501-1820 article 2023 ftdoajarticles https://doi.org/10.3390/photonics10121357 2023-12-24T01:36:29Z Due to its orbital angular momentum (OAM), optical vortex has been widely used in communications and LIDAR target detection. The OAM mode recognition based on deep learning is mostly based on the basic convolutional neural network. To ensure high-precision OAM state detection, a deeper network structure is required to overcome the problem of similar light intensity distribution of different superimposed vortex beams and the effect of atmospheric turbulence disturbance. However, the large number of parameters and the computation of the OAM state detection network conflict with the requirements of deploying optical communication system equipment. In this paper, an online knowledge distillation scheme is selected to achieve an end-to-end single-stage training and the inter-class dark knowledge of similar modes are fully utilized. An optical vortex OAM state detection technique based on deep mutual learning (DML) is proposed. The simulation results show that after mutual learning training, a small detection network with higher accuracy can be obtained, which is more suitable for terminal deployment. Based on the scalability of the number of networks in the DML queue, it provides a new possibility to further improve the detection accuracy of the optical communication. Article in Journal/Newspaper DML Directory of Open Access Journals: DOAJ Articles Photonics 10 12 1357 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
deep mutual learning orbital angular momentum mode recognition knowledge distillation Applied optics. Photonics TA1501-1820 |
spellingShingle |
deep mutual learning orbital angular momentum mode recognition knowledge distillation Applied optics. Photonics TA1501-1820 Tan Qu Zhiming Zhao Yan Zhang Jiaji Wu Zhensen Wu Deep Mutual Learning-Based Mode Recognition of Orbital Angular Momentum |
topic_facet |
deep mutual learning orbital angular momentum mode recognition knowledge distillation Applied optics. Photonics TA1501-1820 |
description |
Due to its orbital angular momentum (OAM), optical vortex has been widely used in communications and LIDAR target detection. The OAM mode recognition based on deep learning is mostly based on the basic convolutional neural network. To ensure high-precision OAM state detection, a deeper network structure is required to overcome the problem of similar light intensity distribution of different superimposed vortex beams and the effect of atmospheric turbulence disturbance. However, the large number of parameters and the computation of the OAM state detection network conflict with the requirements of deploying optical communication system equipment. In this paper, an online knowledge distillation scheme is selected to achieve an end-to-end single-stage training and the inter-class dark knowledge of similar modes are fully utilized. An optical vortex OAM state detection technique based on deep mutual learning (DML) is proposed. The simulation results show that after mutual learning training, a small detection network with higher accuracy can be obtained, which is more suitable for terminal deployment. Based on the scalability of the number of networks in the DML queue, it provides a new possibility to further improve the detection accuracy of the optical communication. |
format |
Article in Journal/Newspaper |
author |
Tan Qu Zhiming Zhao Yan Zhang Jiaji Wu Zhensen Wu |
author_facet |
Tan Qu Zhiming Zhao Yan Zhang Jiaji Wu Zhensen Wu |
author_sort |
Tan Qu |
title |
Deep Mutual Learning-Based Mode Recognition of Orbital Angular Momentum |
title_short |
Deep Mutual Learning-Based Mode Recognition of Orbital Angular Momentum |
title_full |
Deep Mutual Learning-Based Mode Recognition of Orbital Angular Momentum |
title_fullStr |
Deep Mutual Learning-Based Mode Recognition of Orbital Angular Momentum |
title_full_unstemmed |
Deep Mutual Learning-Based Mode Recognition of Orbital Angular Momentum |
title_sort |
deep mutual learning-based mode recognition of orbital angular momentum |
publisher |
MDPI AG |
publishDate |
2023 |
url |
https://doi.org/10.3390/photonics10121357 https://doaj.org/article/8a89f8e66f844d6482612652531e8908 |
genre |
DML |
genre_facet |
DML |
op_source |
Photonics, Vol 10, Iss 12, p 1357 (2023) |
op_relation |
https://www.mdpi.com/2304-6732/10/12/1357 https://doaj.org/toc/2304-6732 doi:10.3390/photonics10121357 2304-6732 https://doaj.org/article/8a89f8e66f844d6482612652531e8908 |
op_doi |
https://doi.org/10.3390/photonics10121357 |
container_title |
Photonics |
container_volume |
10 |
container_issue |
12 |
container_start_page |
1357 |
_version_ |
1788696210510446592 |