EVM Estimation for Performance Monitoring in Coherent Optical Systems: An Approach of Linear RegressionShow others and affiliations
2022 (English)In: Optics InfoBase Conference Papers, Optica Publishing Group (formerly OSA) , 2022Conference paper, Published paper (Refereed)
Abstract [en]
We experimentally demonstrate the effectiveness of a simple linear regression scheme for optical performance monitoring when applied after modulation format identification. It outperforms the FFNN-based benchmark scheme providing 0.2% mean absolute error for EVM estimation., © 2022 The Author(s)
Place, publisher, year, edition, pages
Optica Publishing Group (formerly OSA) , 2022.
Keywords [en]
Coherent optical systems, Mean absolute error, Modulation formats, Optical performance monitoring, Performance-monitoring, Simple linear regression
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:ri:diva-60188Scopus ID: 2-s2.0-85136804299ISBN: 9781557528209 (print)OAI: oai:DiVA.org:ri-60188DiVA, id: diva2:1701928
Conference
CLEO: Science and Innovations, S and I 2022, 15 May 2022 through 20 May 2022
Note
Conference code: 181730; Export Date: 22 September 2022; Conference Paper; Correspondence Address: Ozolins, O.; School of Engineering Sciences, Isafjordsgatan 22, Sweden; email: ozolins@kth.se; Funding details: 1.1.1.2/VIAA/4/20/660; Funding details: Vetenskapsrådet, VR, 2016-04510, 2019-05197, P109599; Funding details: China Scholarship Council, CSC, 201807930003; Funding details: Latvijas Zinātnes Padome, lzp-2021/1-0062; Funding text 1: This work was supported in part by the China Scholarship Council 201807930003, Swedish Research Council (VR) projects 2019-05197 and 2016-04510, in part by the RISE SK funded project Optical Neural Networks (P109599), in part by the ERDF-funded CARAT project (No. 1.1.1.2/VIAA/4/20/660), and in part by the Latvian Council of Science funded ONN-AI project No. lzp-2021/1-0062.
2022-10-072022-10-072024-03-05Bibliographically approved