Low Complexity Convolutional Neural Networks for Equalization in Optical Fiber Transmission
Pedro, J. M.
; Costa, N.
Low Complexity Convolutional Neural Networks for Equalization in Optical Fiber Transmission, Proc OSA Signal Processing in Photonic Communications - SPPCom, Burlingame, United States, Vol. , pp. - , July, 2021.
Digital Object Identifier:
Abstract
A convolutional neural network is proposed to mitigate fiber transmission ef- fects, achieving a five-fold reduction in trainable parameters compared to alternative equal- izers, and 3.5 dB improvement in MSE compared to DBP with comparable complexity.
