Transceiver-Impairment Mitigation Enabled by Adaptive Symbol Decision with Neural Networks

Transceiver-Impairment Mitigation Enabled by Adaptive Symbol Decision with Neural Networks

We propose to use symbol decision based on neural networks to alleviate the impacts of quantization noise, laser phase noise, IQ impairments, and their interactions. Our DSP-assisted transceiver design allows the use of low-cost transceiver hardware while satisfying the performance requirements. Ext...

Full description

Saved in:
Bibliographic Details
Published in:2023 International Conference on Photonics in Switching and Computing (PSC) pp. 1 - 3
Main Authors: Ochiai, Takuro, Higuchi, Reiji, Kuno, Takuma, Shiraki, Ryuta, Mori, Yojiro, Hasegawa, Hiroshi
Format: Conference Proceeding
Language:English
Published: IEEE 26-09-2023
Subjects:
Artificial neural networks
digital coherent receiver
IQ impairments
Laser noise
machine learning
Phase noise
Quantization (signal)
quantization noise
Receivers
Switches
Symbols
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose to use symbol decision based on neural networks to alleviate the impacts of quantization noise, laser phase noise, IQ impairments, and their interactions. Our DSP-assisted transceiver design allows the use of low-cost transceiver hardware while satisfying the performance requirements. Extensive computer simulations show the effectiveness of neural networks when higher-order QAM signals are contaminated by the above impairments.
ISSN:2166-8892
DOI:10.1109/PSC57974.2023.10297285