Recurrent neural network equalizer to extend input power dynamic range of SOA in 100Gb/s/λ PON

Murphy, Stephen; Townsend, Paul D.; Antony, Cleitus

Recurrent neural network equalizer to extend input power dynamic range of SOA in 100Gb/s/λ PON

Files

CLEO22_Extended_Deadline_FinalDraft.pdf(669.52 KB)

Accepted Version

Date

2022-09-23

Authors

Murphy, Stephen

Townsend, Paul D.

Antony, Cleitus

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Abstract

We propose a novel equalization scheme for 100Gb/s/λ PAM4 PON based on Gated Recurrent Neural Network to increase SOA preamplifier input power dynamic range tolerance to 30 dB below hard-decision FEC BER limit of 3.8×10 −3.

Keywords

Recurrent neural networks , Equalizers , Dynamic range , Logic gates , Lasers and electrooptics , Preamplifiers , Electrooptical waveguides

Citation

Murphy, S., Townsend, P. D. and Antony, C. (2022) 'Recurrent neural network equalizer to extend input power dynamic range of SOA in 100Gb/s/λ PON', 2022 Conference on Lasers and Electro-Optics (CLEO), San Jose, CA, USA, 15-20 May. Available at: https://ieeexplore.ieee.org/document/9889990 (Accessed: 29 November 2022)