Reduced Complexity Nonlinearity Compensation via Principal Component Analysis and Deep Neural Networks

作者：Yuliang Gao, Ziad A. El-Sahn, Ahmed Awadalla, Demin Yao, Han Sun, Pierre Mertz, Kuang‐Tsan Wu · 年份：2019 · DOI：10.1364/ofc.2019.th2a.49 · 被引用次数：10 · 研究领域：Optical Network Technologies、Advanced Photonic Communication Systems、Advanced Optical Network Technologies

We demonstrate a novel fiber nonlinearity post-equalization algorithm using principal component analysis and neural networks. We achieve ∼0.46 dBQ improvement for 21 Gbaud DP-8QAM transmission over ∼13,000 km deployed fiber with over 90% complexity reduction.