Towards FPGA Implementation of Neural Network-Based Nonlinearity Mitigation Equalizers in Coherent Optical Transmission Systems

TL;DR

This paper demonstrates the first FPGA implementation of neural network-based equalizers for nonlinearity mitigation in coherent optical systems, showing they can outperform traditional digital backpropagation methods.

Contribution

It introduces FPGA-compatible neural network equalizers for nonlinearity mitigation, achieving comparable complexity to dispersion equalizers and outperforming 1 step-per-span digital backpropagation.

Findings

Neural network equalizers outperform traditional methods.

FPGA implementation is feasible with manageable complexity.

Achieved real-time processing capability.

Abstract

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 equalizer. We demonstrate that the NN-based equalizers can outperform a 1 step-per-span DBP.