Concept and Experimental Demonstration of Optical IM/DD End-to-End System Optimization using a Generative Model

TL;DR

This paper demonstrates an experimental optical IM/DD system optimized end-to-end using deep learning with a generative adversarial network, enabling channel approximation without explicit channel modeling.

Contribution

It introduces a novel end-to-end optimization method for optical IM/DD systems using a generative adversarial network to model the channel.

Findings

Successful experimental implementation of the optimized system

Effective channel approximation without explicit models

Improved system performance through deep learning optimization

Abstract

We perform an experimental end-to-end transceiver optimization via deep learning using a generative adversarial network to approximate the test-bed channel. Previously, optimization was only possible through a prior assumption of an explicit simplified channel model.