Adapting Spiking Neural Networks for Enhanced Optical Signal Equalization in Communication Systems

Conference: ECOC 2024 - 50th European Conference on Optical Communication
09/22/2024 - 09/26/2024 at Frankfurt, Germany

Proceedings: ITG-Fb. 317: ECOC 2024

Pages: 4Language: englishTyp: PDF

Authors:
Ron, D. Arguello

Abstract:
A version of the Receptance Weighted Key Value (RWKV) language model, that integrates spiking linear transformations and reconfigured for the regression task of signal equalization is employed to compensate the nonlinearity from the transmission of PDM 34.4 Gbaud 16QAM over 10x50km of TWC fiber.