Attention-Based Neural Network Equalization in Fiber-Optic Communications - Télécom Paris Access content directly
Journal Articles Asia Communications and Photonics Conference 2021 Year : 2021

Attention-Based Neural Network Equalization in Fiber-Optic Communications

Abstract

An attention mechanism is integrated into neural network-based equalizers to prune the fully-connected output layer. For a 100 GBd 16-QAM 20 × 100 km SMF transmission, this approach reduces the computational complexity by ∼15% in a CNN+LSTM model.
Fichier principal
Vignette du fichier
APC2021__open_access_version_.pdf (885.17 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03585118 , version 1 (22-02-2022)

Identifiers

Cite

Abtin Shahkarami, Mansoor Yousefi, Yves Jaouën. Attention-Based Neural Network Equalization in Fiber-Optic Communications. Asia Communications and Photonics Conference 2021, 2021, ⟨10.1364/acpc.2021.m5h.3⟩. ⟨hal-03585118⟩
93 View
124 Download

Altmetric

Share

Gmail Facebook X LinkedIn More