Model-Based Machine Learning for Joint Digital Backpropagation and PMD Compensation

@article{Hger2020ModelBasedML,
  title={Model-Based Machine Learning for Joint Digital Backpropagation and PMD Compensation},
  author={Christian H{\"a}ger and Henry D. Pfister and Rick M. B{\"u}tler and Gabriele Liga and Alex Alvarado},
  journal={2020 Optical Fiber Communications Conference and Exhibition (OFC)},
  year={2020},
  pages={1-3}
}
We propose a model-based machine-learning approach for polarization-multiplexed systems by parameterizing the split-step method for the Manakov-PMD equation. This approach performs hardware-friendly DBP and distributed PMD compensation with performance close to the PMD-free case. 

Figures from this paper

Model-Based Machine Learning for Joint Digital Backpropagation and PMD Compensation
TLDR
A model-based machine-learning approach for dual-polarization systems by parameterizing the split-step Fourier method for the Manakov-PMD equation, which combines hardware-friendly time-domain nonlinearity mitigation via the recently proposed learned digital backpropagation (LDBP) with distributed compensation of polarization-mode dispersion (PMD).
Learned Digital Back-Propagation for Dual-Polarization Dispersion Managed Systems
Digital back-propagation (DBP) and learned DBP (LDBP) are proposed for nonlinearity mitigation in WDM dual-polarization dispersion-managed systems. LDBP achieves Q -factor improvement of 1.8 dB and
Physics-Based Deep Learning for Fiber-Optic Communication Systems
TLDR
The proposed algorithm, dubbed learned DBP (LDBP), uses the physics-based model with trainable filters in each step and its complexity is reduced by progressively pruning filter taps during gradient descent, illustrating that combining data-driven optimization with existing domain knowledge can generate new insights into old communications problems.
Revisiting Efficient Multi-Step Nonlinearity Compensation With Machine Learning: An Experimental Demonstration
TLDR
This paper considers the recently proposed learned digital backpropagation (LDBP) approach, where the linear steps in the split-step method are re-interpreted as general linear functions, similar to the weight matrices in a deep neural network, and shows how LDBP can be integrated into a coherent receiver DSP chain and successfully trained in the presence of various hardware impairments.
Few-bit Quantization of Neural Networks for Nonlinearity Mitigation in a Fiber Transmission Experiment
A neural network is quantized for the mitigation of nonlinear and components’ distortions in a 16-QAM 9x50km dual-polarization fiber transmission experiment. Post-training additive power-of-two
Experimental demonstration of low complexity nonlinear compensation algorithm based on logistic regression
In this paper, a novel algorithm based on logistic regression digital backpropagation (LR‐DBP) is proposed to mitigate the effect of dispersion and nonlinearity in a long‐distance optical
Advanced Convolutional Neural Networks for Nonlinearity Mitigation in Long-Haul WDM Transmission Systems
TLDR
This work investigates the application of convolutional neural networks for compensating nonlinear signal distortions in a 3200 km fiber-optic 11x400-Gb/s WDM PDM-16QAM transmission link with a focus on the optimization of the corresponding algorithmic complexity.
Machine Learning Applications for Short Reach Optical Communication
TLDR
A comprehensive review of various ML methods and their applications in short-reach optical communications are presented and discussed, focusing on existing and potential advantages, limitations and prospective trends.
Towards FPGA Implementation of Neural Network-Based Nonlinearity Mitigation Equalizers in Coherent Optical Transmission Systems
For the first time, recurrent and feedforward neural network-based equalizers for nonlinearity compensation are implemented in an FPGA, with a level of complexity comparable to that of a dispersion

References

SHOWING 1-10 OF 43 REFERENCES
Model-Based Machine Learning for Joint Digital Backpropagation and PMD Compensation
TLDR
A model-based machine-learning approach for dual-polarization systems by parameterizing the split-step Fourier method for the Manakov-PMD equation, which combines hardware-friendly time-domain nonlinearity mitigation via the recently proposed learned digital backpropagation (LDBP) with distributed compensation of polarization-mode dispersion (PMD).
Fiber nonlinearity compensation using extreme learning machine for DSP-based coherent communication systems
  • T. S. Shen, A. Lau
  • Physics
    16th Opto-Electronics and Communications Conference
  • 2011
TLDR
Simulations show comparable compensation results to backpropagation (BP) while the computation complexity is significantly reduced in extreme learning machine.
Nonlinear Interference Mitigation via Deep Neural Networks
TLDR
A neural-network-based approach is presented to efficiently implement digital backpropagation (DBP) and the resulting "learned" DBP significantly reduces the complexity compared to conventional DBP implementations.
Revisiting Multi-Step Nonlinearity Compensation with Machine Learning
TLDR
It is shown that carefully designed multi-step approaches can lead to better performance-complexity trade-offs than their few-step counterparts.
ASIC Implementation of Time-Domain Digital Backpropagation with Deep-Learned Chromatic Dispersion Filters
TLDR
This work considers time-domain digital backpropagation with chromatic dispersion filters jointly optimized and quantized using machine-learning techniques and shows improved BER performance and >40% power dissipation reductions in 28-nm CMOS.
Polarization-mode dispersion aware digital backpropagation
TLDR
A modified DBP algorithm is studied that distributively compensates for PMD in the reverse propagation and outperforms the conventional approach by up to 2.1 dB.
Wideband Time-Domain Digital Backpropagation via Subband Processing and Deep Learning
TLDR
This work proposes a low-complexity sub-banded DSP architecture for digital backpropagation where the walk-off effect is compensated using simple delay elements and achieves a 2.8 dB SNR improvement over linear equalization.
A PMD-adaptive DBP Receiver Based on SNR Optimization
TLDR
A novel adaptive digital backpropagation (DBP) scheme that tracks the fiber polarization-mode dispersion via the optimization of the signal-to-noise ratio is proposed, achieving gains of up to 1.4 dB over conventional DBP.
Time-domain digital back propagation: Algorithm and finite-precision implementation aspects
TLDR
A nonlinear mitigation algorithm designed from an ASIC perspective, given 9 signal and 11 coefficient bits, reach is increased by 105% compared to linear compensation in single-channel 16-QAM transmission.
Artificial Neural Network Nonlinear Equalizer for Coherent Optical OFDM
We propose a novel low-complexity artificial neural network (ANN)-based nonlinear equalizer (NLE) for coherent optical orthogonal frequency-division multiplexing (CO-OFDM) and compare it with the
...
...