Revisiting Multi-Step Nonlinearity Compensation with Machine Learning

04/22/2019

∙

For the efficient compensation of fiber nonlinearity, one of the guiding principles appears to be: fewer steps are better and more efficient. We challenge this assumption and show that carefully designed multi-step approaches can lead to better performance-complexity trade-offs than their few-step counterparts.

READ FULL TEXT

Revisiting Multi-Step Nonlinearity Compensation with Machine Learning

Optimal Robustness-Consistency Trade-offs for Learning-Augmented Online Algorithms

Profinet vs Profibus

Inflationary Constant Factors and Why Python is Faster Than C++

On the Fundamental Trade-offs in Learning Invariant Representations

Reliability-Performance Trade-offs in Neuromorphic Computing

Dynamic Selection of Perception Models for Robotic Control

Analysis of Trade-offs in RF Photonic Links based on Multi-Bias Tuning of Silicon Photonic Ring-Assisted Mach Zehnder Modulators

Revisiting Multi-Step Nonlinearity Compensation with Machine Learning

Related Research

Optimal Robustness-Consistency Trade-offs for Learning-Augmented Online Algorithms

Profinet vs Profibus

Inflationary Constant Factors and Why Python is Faster Than C++

On the Fundamental Trade-offs in Learning Invariant Representations

Reliability-Performance Trade-offs in Neuromorphic Computing

Dynamic Selection of Perception Models for Robotic Control

Analysis of Trade-offs in RF Photonic Links based on Multi-Bias Tuning of Silicon Photonic Ring-Assisted Mach Zehnder Modulators