Abstract

We improve an inverse regular perturbation (RP) model using a machine learning (ML) technique. The proposed learned RP (LRP) model jointly optimizes step-size, gain and phase rotation for individual RP branches. We demonstrate that the proposed LRP can outperform the corresponding learned digital back-propagation (DBP) method based on a split-step Fourier method (SSFM), with up to 0.75 dB gain in a 800 km standard single mode fiber link. Our LRP also allows a fractional step-per-span (SPS) modeling to reduce complexity while maintaining superior performance over a 1-SPS SSFM-DBP.

© 2022 Optica Publishing Group

Design of fully interpretable neural networks for digital coherent demodulation

Xiatao Huang, Wenshan Jiang, Xingwen Yi, Jing Zhang, Taowei Jin, Qianwu Zhang, Bo Xu, and Kun Qiu
Opt. Express 30(20) 35526-35538 (2022)

Intelligent and adaptive intra-channel and inter-channel fiber nonlinearity compensation in coherent optical transmission systems

Xiansong Fang, Xinyu Chen, Xiang Cai, Chuanchuan Yang, and Fan Zhang
Opt. Lett. 48(15) 4093-4096 (2023)

Learning-based digital back propagation to compensate for fiber nonlinearity considering self-phase and cross-phase modulation for wavelength-division multiplexed systems

Takashi Inoue, Ryosuke Matsumoto, and Shu Namiki
Opt. Express 30(9) 14851-14872 (2022)

Adaptive digital back propagation exploiting adjoint-based optimization for fiber-optic communications

Mahmoud M. T. Maghrabi, Mohamed H. Bakr, and Shiva Kumar
Opt. Express 30(10) 16264-16288 (2022)

Comparison of DSP-based nonlinear equalizers for intra-channel nonlinearity compensation in coherent optical OFDM

Elias Giacoumidis, Sofien Mhatli, Tu Nguyen, Son T. Le, Ivan Aldaya, Mary E. McCarthy, Andrew D. Ellis, and Benjamin J. Eggleton
Opt. Lett. 41(11) 2509-2512 (2016)