Efficient Training of Volterra Series-Based Pre-distortion Filter Using Neural Networks
Conference Paper
(2022)
Author(s)
V. Bajaj (TU Delft - Mechanical Engineering, Nokia Bell Labs, Stuttgart)
Mathieu Chagnon (Nokia Bell Labs, Stuttgart)
S. Wahls (TU Delft - Mechanical Engineering)
Vahid Aref (Nokia Bell Labs, Stuttgart)
Research Group
Team Gabriel Gleizer
DOI related publication
https://doi.org/10.1364/OFC.2022.M1H.3
Final published version
To reference this document use
https://resolver.tudelft.nl/uuid:4b610b59-15eb-4b84-bbef-fd9c3473b92b
More Info
expand_more
expand_more
Publication Year
2022
Language
English
Research Group
Team Gabriel Gleizer
Article number
M1H.3
ISBN (electronic)
978-1-55752-466-9
Event
2022 Optical Fiber Communications Conference and Exhibition (OFC) (2022-03-06 - 2022-03-10), San Diego, United States
Downloads counter
213
Abstract
We present a simple, efficient “direct learning” approach to train Volterra series-based pre-distortion filters using neural networks. We show its superior performance over conventional training methods using a 64-QAM 64 GBaud simulated transmitter with varying transmitter nonlinearity and noisy conditions.