Power and Modulation Format Transfer Learning for Neural Network Equalizers in Coherent Optical Transmission Systems

Transfer learning is proposed to adapt an NN-based nonlinear equalizer across different launch powers and modulation formats using a 450km TWC-fiber transmission. The result shows up to 92% reduction in epochs or 90% in the training dataset.

Engineering Physics Computer Science

Source

Paper

Transfer Learning Aided Neural Networks for Nonlinear Equalization in Short-Reach Direct Detection SystemsZhaopeng Xu, Chuanbowen Sun, Tonghui Ji et al.2020 · Optical Fiber Communications Conference and Exhibition · 17 citations In Library

Multi-Task Learning to Enhance Generalizability of Neural Network Equalizers in Coherent Optical SystemsS. Srivallapanondh, Pedro J. Freire, A. Alam et al.2023 · European Conference on Optical Communication · 4 citations In Library

Multi-Task Learning Aided Neural Networks for Equalization in Coherent Optical SystemFangfang Huang, Xiatao Huang, J. Zhang et al.2022 · 2022 IEEE/CIC International Conference on Communications in China (ICCC) · 0 citations In Library

Low Complexity Convolutional Neural Networks for Equalization in Optical Fiber TransmissionMohannad Abu-Romoh, N. Costa, A. Napoli et al.2022 · OSA Advanced Photonics Congress 2021 · 1 citation In Library

Convolutional Neural Network Equalizer for Short-reach Optical Communication SystemsJing Zhang, Lianshan Yan, Lin Jiang et al.2020 · Asia Communications and Photonics conference and Exhibition · 4 citations In Library

Neural Turbo Equalization: Deep Learning for Fiber-Optic Nonlinearity CompensationT. Koike-Akino, Ye Wang, D. Millar et al.2019 · Journal of Lightwave Technology · 39 citations In Library

Neural Networks based Equalization of Experimental Transmission using the Nonlinear Fourier TransformationJ. Koch, Ken Chan, S. Kühl et al.2020 · European Conference on Optical Communication · 2 citations In Library

Experimental Study of Deep Neural Network Equalizers Performance in Optical LinksPedro J. Freire, Yevhenii Osadchuk, B. Spinnler et al.2021 · Optical Fiber Communications Conference and Exhibition · 9 citations In Library

Convolutional long short-term memory neural network equalizer for nonlinear Fourier transform-based optical transmission systems.O. Kotlyar, M. Kamalian-Kopae, M. Pankratova et al.2021 · Optics Express · 28 citations In Library

Patents near this paper

ADAPTIVE EQUALIZER USING SELF-LEARNING NEURAL NETWORKUS 5,504,780 · BELL COMMUNICATIONS RESEARCH, INC. · 1996 NEURAL NETWORK TRANSFER LEARNING FOR QUALITY OF TRANSMISSION PREDICTIONUS 11,461,635 · NEC LABORATORIES AMERICA, INC. · 2022 OPTICAL FIBER NONLINEARITY COMPENSATION USING NEURAL NETWORKSUS 10,833,770 · NEC CORPORATION · 2020 SYSTEM AND METHOD FOR OPTICAL CHANNEL RECONFIGURATIONUS 10,630,416 · HUAWEI TECHNOLOGIES CO., LTD. · 2020 METHOD AND APPARATUS FOR ELECTRONIC EQUALIZATION IN OPTICAL COMMUNICATION SYSTEMSUS 7,471,904 · LUCENT TECHNOLOGIES INC. · 2008

Source of record

arXiv

Power and Modulation Format Transfer Learning for Neural Network Equalizers in Coherent Optical Transmission Systems

Paper

Similar papers

Patents near this paper

Welcome.

Patents near this paper