Deep Convolutional Recurrent Neural Network For Fiber Nonlinearity Compensation
- Resource Type
- Conference
- Authors
- Jain, Prasham; Lampe, Lutz; Mitra, Jeebak
- Source
- 2022 European Conference on Optical Communication (ECOC) Optical Communication (ECOC), 2022 European Conference on. :1-4 Sep, 2022
- Subject
- Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Photonics and Electrooptics
Signal Processing and Analysis
Q-factor
Optical losses
Recurrent neural networks
Artificial neural networks
Optical fiber networks
Complexity theory
Iterative methods
- Language
An iterative deep convolutional recurrent neural network is proposed to mitigate fiber non-linearity with distributed compensation of polarization mode dispersion, demonstrating 1.3 dB Q-factor gain over previous neural network based techniques for dual-polarized 960 km 32 Gbaud 64QAM transmission.