Neural-network-based MDG and Optical SNR Estimation in SDM Transmission

Abstract

We propose a neural network model for MDG and optical SNR estimation in SDM transmission. We show that the proposed neural-network-based solution estimates MDG and SNR with high accuracy and low complexity from features extracted after DSP.

MDG and SNR Estimation in SDM Transmission Based on Artificial Neural Networks

“…The results are validated in a back-to-back 32.5m 3-mode FMF link. This paper extends the results in [19], discussing the method in detail, and validating it in an experimental short-haul 73km 3-mode FMF link with controlled MDG and optical SNR. In addition, we apply the ANN estimator in a case study of an experimental long-haul 6-mode transmission setup with unknown MDG and SNR.…”

“…To circumvent the limitations of conventional methods, we propose in [19] an ANN-based method to estimate σ mdg and SNR from features extracted after DSP. The block diagram of the proposed solution is depicted in Fig.…”

“…Currently, machine learning (ML) techniques are being extensively investigated for optical performance monitoring in both single-mode [17] and mode-multiplexed systems [18]. In [19], we propose an artificial neural network (ANN)-based solution to estimate both MDG and optical SNR in coupled SDM transmission. The results are validated in a back-to-back 32.5m 3-mode FMF link.…”