公开(公告)号：US11133865B2

公开(公告)日：2021-09-28

申请号：US17106143

申请日：2020-11-29

Applicant: NEC Laboratories America, Inc.

Inventor： Yue-Kai Huang ,  Shaoliang Zhang ,  Ezra Ip ,  Jiakai Yu

IPC: H04B10/079 ,  G06N3/04 ,  H04L27/34 ,  H04B10/07

Abstract: Aspects of the present disclosure describe systems, methods. and structures in which a hybrid neural network combining a CNN and several ANNs are shown useful for predicting G-ONSR for Ps-256QAM raw data in deployed SSMF metro networks with 0.27 dB RMSE. As demonstrated, the CNN classifier is trained with 80.96% testing accuracy to identify channel shaping factor. Several ANN regression models are trained to estimate G-OSNR with 0.2 dB for channels with various constellation shaping. Further aspects include the tuning of existing optical networks and the characterization of retrofit/upgraded optical networks to estimate capacity—both aspects employing our inventive hybrid neural network methodology.

公开(公告)号：US10887009B2

公开(公告)日：2021-01-05

申请号：US16596582

申请日：2019-10-08

Applicant: NEC Laboratories America, Inc.

Inventor： Yue-Kai Huang ,  Shaoliang Zhang ,  Ezra Ip ,  Jiakai Yu

IPC: G06N3/08 ,  G06N3/04 ,  H04L27/34 ,  H04B10/07 ,  H04B10/079

Abstract: Aspects of the present disclosure describe systems, methods and structures in which a hybrid neural network combining a CNN and several ANNs are shown useful for predicting G-ONSR for Ps-256QAM raw data in deployed SSMF metro networks with 0.27 dB RMSE. As demonstrated, the CNN classifier is trained with 80.96% testing accuracy to identify channel shaping factor. Several ANN regression models are trained to estimate G-OSNR with 0.2 dB for channels with various constellation shaping.