Improved neural network based demodulation of orbital angular momentum fiber-optic communication system

SHEN Yuanxun; CHEN Xinjie; MA Xuelian

您当前的位置：

首页 >

文章列表页 >

Improved neural network based demodulation of orbital angular momentum fiber-optic communication system

Optical Communication Technology | 更新时间：2026-03-24

- Improved neural network based demodulation of orbital angular momentum fiber-optic communication system
- OPTICAL TECHNIQUE Vol. 52, Issue 2, Pages: 190-198(2026)
- 作者机构：
  
  东北大学计算机与通信工程学院，河北秦皇岛 066004
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP394.1;TH691.9
- Received：05 January 2026，
  
  Revised：2026-03-02，
  
  Accepted：03 March 2026，
  
  Published：20 March 2026
- 稿件说明：
移动端阅览
沈袁勋,陈鑫杰,马雪莲.基于改进神经网络的轨道角动量光纤通信解调[J].光学技术,2026,52(02):190-198.

SHEN Yuanxun,CHEN Xinjie,MA Xuelian.Improved neural network based demodulation of orbital angular momentum fiber-optic communication system[J].OPTICAL TECHNIQUE,2026,52(02):190-198.
沈袁勋,陈鑫杰,马雪莲.基于改进神经网络的轨道角动量光纤通信解调[J].光学技术,2026,52(02):190-198. DOI：

SHEN Yuanxun,CHEN Xinjie,MA Xuelian.Improved neural network based demodulation of orbital angular momentum fiber-optic communication system[J].OPTICAL TECHNIQUE,2026,52(02):190-198. DOI：

摘要

针对轨道角动量光纤通信系统误码率高的问题，提出一种基于改进神经网络的轨道角动量（OAM）解调方法。对于神经网络架构，使用卷积神经网络捕捉OAM模式图像的局部特征，使用Swin Transformer捕捉OAM模式图像的全局信息，以增强OAM模式图像特征的判别力，进而提升OAM模式解调的准确性。对于神经网络训练，采用广义互信息代替传统交叉熵作为损失函数，使神经网络在学习OAM模式识别的同时，学习包含信道软信息的概率分布，以提升低密度奇偶校验码的纠错效果。实验结果表明，所提模型通过网络架构与训练两个方面的改进，有效提升了OAM模式的解调准确性，并且显著降低了OAM光纤通信系统的误码率。

Abstract

Aiming at the problem of high bit error rate of orbital angular momentum（OAM） fiber-optic communication system

a demodulation method of orbital angular momentum based on improved neural network is proposed. For neural network architecture

the convolutional neural network is used to capture the local features of the OAM mode image

and the Swin Transformer is adopted to capture the global information of the OAM mode image

in order to enhance the discriminative power of the OAM mode image features. Consequently the accuracy of OAM demodulation is improved. For neural network training

the Generalized Mutual Information（GMI） is used to replace the traditional cross entropy loss

in order to enable the neural network not only to learn OAM mode recognition

but also to learn the probability distribution containing channel soft information

thus improving the error correction effect of Low Density Parity Check code. Experimental results show that

through improving the model architecture and model training

the proposed method effectively increases the demodulation accuracy of OAM mode

and the bit error rate of the OAM Fiber-optic communication system decreases significantly.

关键词

Keywords

references

田野 , 蒋睿 , 张嘉明 , 等 . 面向多芯光纤通信的完美垂直光栅耦合器（特邀） [J ] . 激光与光电子学进展 , 2025 , 62 （ 17 ）: 330 — 335 .

Tian Ye , Jiang Rui , Zhang Jiaming , et al . Perfect vertical grating coupler for multi-core fiber communication（Invited） [J ] . Laser & Optoelectronics Progress , 2025 , 62 （ 17 ）: 330 — 335 .

Varshney A , Goyal V . Fiber optics communication: evolution of guided media [J ] . International Journal for Research in Applied Science and Engineering Technology , 2024 , 12 （ 1 ）: 959 — 969 .

Qiao M , Wang Y , Li Z , et al . Novel dispersion and timing estimation for weakly-coupled OAM fiber transmission systems [J ] . IEEE Photonics Technology Letters , 2024 , 36 （ 14 ）: 913 — 916 .

Liu J , Mo S , Xu Z , et al . High spectral-efficiency, ultra-low MIMO SDM transmission over a field-deployed multi-core OAM fiber [J ] . Photonics Research , 2025 , 13 （ 1 ）: 18 — 30 .

Wang F , Gao R , Li Z , et al . OAM mode-division multiplexing IM/DD transmission at 4.32Tbit/s with a low-complexity adaptive-network-based fuzzy inference system nonlinear equalizer [J ] . Optics letters , 2024 , 49 （ 3 ）: 430 — 433 .

Zhang J , Lin Z , Liu J . SDM transmission of orbital angular momentum mode channels over a multi-ring-core fibre [J ] . Nanophotonics , 2021 , 11 （ 4 ）: 873 — 884 .

Liu J , Zhang J , Liu J , et al . 1-Pbps orbital angular momentum fibre-optic transmission [J ] . Light: Science&Applications , 2022 , 11 （ 8 ）: 1823 — 1833 .

Bai Y , Lv H , Fu X , et al . Vortex beam: generation and detection of orbital angular momentum[Invited] [J ] . Chinese Optics Letters , 2022 , 20 （ 1 ）: 133 — 147 .

Saitoh K , Hasegawa Y , Hirakawa K , et al . Measuring the orbital angular momentum of electron vortex beams using a forked grating [J ] . Physical Review Letters , 2013 , 111 （ 7 ）: 74801 . 1—74801. 5 .

Lei T , Zhang M , Li Y , et al . Massive individual orbital angular momentum channels for multiplexing enabled by Dammann gratings [J ] . Light Science&Applications , 2015 , 4 （ 3 ）: 257 . 1—257. 7 .

Fu S , Wang T , Zhang S , et al . Integrating 5×5 Dammann gratings to detect orbital angular momentum states of beams with the range of 24 to +24 [J ] . Applied Optics , 2016 , 55 （ 7 ）: 1514 — 1520 .

Li B , Luan H , Li K , et al . Orbital angular momentum optical communications enhanced by artificial intelligence [J ] . Journal of Optics , 2022 , 24 （ 9 ）, 094003 .

Hu J , Guo Z , Fu Y , et al . How convolutional-neural-network detects optical vortex scattering fields [J ] . Optics and Lasers in Engineering , 2023 , 160 （ 1 ）: 107246 .

Zhu J , Fan M , Pu Y , et al . 1024-ary composite OAM shift keying for free-space optical communication system decoded by a two-step neural network [J ] . Optics Letters , 2023 , 48 （ 10 ）: 2692 — 2695 .

Wang Z , Lai X , Huang H , et al . Recognizing the orbital angular momentum（OAM） of vortex beams from speckle patterns [J ] . Science China Physics , Mechanics & Astronomy, 2022 , 65 （ 4 ）: 1 — 7 .

Liu Y , Zhang Z , Yu P , et al . Learning-enabled recovering scattered data from twisted light transmitted through a long standard multimode fiber [J ] . Applied Physics Letters , 2022 , 120 （ 13 ）: 131101 .

Wang Z , Chen M , Wan M , et al . Coherent demodulated underwater wireless optical communication system based on convolutional neural network [J ] . Optics Communications , 2023 , 534 （ 1 ）: 129316 .

Tian Q , Li Z , Hu K , et al . Turbo-coded 16-ary OAM shift keying FSO communication system combining the CNN-based adaptive demodulator [J ] . Optics Express , 2018 , 26 （ 21 ）: 27849 — 27864 .

雎洲 , 王晨 , 王凯辉 , 等 . 32.768 Tbit/s净速率1000 km长少模光纤波分复用系统 [J ] . 光学学报 , 2023 , 43 （ 10 ）: 38 — 45 .

Ju Zhou , Wang Chen , Wang Kaihui , et al . 1000-km-Long Few-Mode Fiber Wavelength Division Multiplexing Transmission System with Net Rate of 32.768 Tbit/s [J ] . Acta Optica Sinica , 2023 , 43 （ 10 ）: 38 — 45 .

Elmeadawy S , Shalaby H M H , Ismail N A , et al . A free space 16-ary OAM coded optical communication system based on chaotic-interleavingand CNNs [J ] . Applied Optics , 2020 , 59 （ 23 ）: 6966 — 6976 .

Chen W , Hou Z , Yang L , et al . Performance investigation of an OAM-SK optical communication system based on a PCNN-LDPC under atmospheric turbulence [J ] . Optics Express , 2025 , 33 （ 3 ）: 5369 — 5381 .

Jiang W , Zhao X , Fang F , et al . End-to-end learning of constellation shaping for optical fiber communication systems [J ] . IEEE Photonics Journal , 2023 , 15 （ 6 ）: 1 — 7 .

王千帆 , 杨佳仪 , 王寅楚 , 等 . 面向流式通信的耦合LDPC码研究综述 [J ] . 电子学报 , 2024 , 52 （ 8 ）: 2913 — 2932 .

Wang Qianfan , Yang Jiayi , Wang yinchu , et al . A review of coupled LDPC codes for streaming communications [J ] . Acta Electronica Sinica , 2024 , 52 （ 8 ）: 2913 — 2932 .

Arbab A , Habibi A , Rabbani H , et al . From image to sequence: exploring vision transformers for optical coherence tomography classification [J ] . Journal of Medical Signals & Sensors , 2025 , 15 （ 6 ）: 1 — 18 .

Liu Ze , Lin Yutong , Cao Yue , et al . Swin transformer: hierarchical vision transformer using shifted windows [C ] // 2021 IEEE/CVF International Conference on Computer Vision . Montreal, QC, Canada : IEEE , 2021 : 9992 — 10002 .

Alvarado A , Agrell E , Lavery D , et al . Replacing the soft-decision FEC limit paradigm in the design of optical communication systems [J ] . Journal of Lightwave Technology , 2015 , 33 （ 20 ）: 4338 — 4352 .

Hu X Y , Eleftheriou E , Arnold D M , et al . Efficient implementations of the sum-product algorithm for decoding LDPC codes [C ] // IEEE Global Telecommunications Conference . San Antonio, TX, USA : IEEE , 2001 : 1036A — 1036E .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

OAM mode recognition system of vortex beams based on deep learning and FPGA

Icing type classification via Near-infrared spectral irradiance and deep convolutional neural networks

Nonlinear impairment compensation method for Long-haul Fiber-optic communication system

Related Author

ZHANG Chen

LV Hong

WANG Kunpeng

LIU Yike

WU Tongqiao

SHAO Ruikang

HUANG Dingjin

WU Jun

Related Institution

College of Photoelectric Engineering, Xi 'an University of Technology, Xi 'an

School of Aeronautical Engineering, Civil Aviation University of China

School of Electronic Information and Automation, Civil Aviation University of China

Engineering and Technical Training Center, Civil Aviation University of China

School of Information Engineering, Hubei Polytechnic Institute

⁰