Revisiting Multi-Step Nonlinearity Compensation with Machine Learning
arXiv:1904.09807v110 citations
Originality Incremental advance
AI Analysis
This work addresses efficiency in fiber optic communication systems, presenting an incremental improvement over existing methods.
The paper challenges the assumption that fewer steps are better for fiber nonlinearity compensation, showing that carefully designed multi-step approaches can achieve better performance-complexity trade-offs than few-step methods.
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.