Supervised Machine Learning for Refractive Index Structure Parameter Modeling
-
Published:2023-06-01
Issue:2
Volume:7
Page:18
-
ISSN:2412-382X
-
Container-title:Quantum Beam Science
-
language:en
-
Short-container-title:QuBS
Author:
Lionis Antonios1ORCID, Peppas Konstantinos1ORCID, Nistazakis Hector E.2ORCID, Tsigopoulos Andreas3ORCID, Cohn Keith4, Drexler Kyle R.5
Affiliation:
1. Information and Telecommunications Department, University of Peloponnese, 22100 Tripoli, Greece 2. Section of Electronic Physics and Systems, Department of Physics, National and Kapodistrian University of Athens, Panepistimiopolis Zografou, 15784 Athens, Greece 3. Division of Combat Systems, Naval Operations, Sea Sciences, Navigation, Electronics & Telecommunications Sector, Hellenic Naval Academy, 18538 Pireas, Greece 4. Physics Department, Naval Postgraduate School, Monterey, CA 93943, USA 5. Atmospheric Propagation, Naval Information Warfare Center Pacific, San Diego, CA 92152, USA
Abstract
The Hellenic Naval Academy (HNA) reports the latest results from a medium-range, near-maritime, free-space laser-communications-testing facility, between the lighthouse of Psitalia Island and the academy’s laboratory building. The FSO link is established within the premises of Piraeus port, with a path length of 2958 m and an average altitude of 35 m, mainly above water. Recently, the facility was upgraded through the addition of a BLS450 scintillometer, which is co-located with the MRV TS5000/155 FSO system and a WS-2000 weather station. This paper presents the preliminary optical turbulence measurements, collected from 24 to 31 of May 2022, alongside the macroscopic meteorological parameters. Four machine-learning algorithms (random forest (RF), gradient boosting regressor (GBR), single layer (ANN), and deep neural network (DNN)) were utilized for refractive-index-structural-parameter regression modeling. Additionally, another DNN was used to classify the strength level of the optical turbulence, as either strong or weak. The results showed very good prediction accuracy for all the models. Specifically, the ANN algorithm resulted in an R-squared of 0.896 and a mean square error (MSE) of 0.0834; the RF algorithm also gave a highly acceptable R-squared of 0.865 and a root mean square error (RMSE) of 0.241. The Gradient Boosting Regressor (GBR) resulted in an R-squared of 0.851 and a RMSE of 0.252 and, finally, the DNN algorithm resulted in an R-squared of 0.79 and a RMSE of 0.088. The DNN-turbulence-strength-classification model exhibited a very acceptable classification performance, given the highly variability of our target value (Cn2), since we observed a predictive accuracy of 87% with the model.
Subject
Nuclear and High Energy Physics,Atomic and Molecular Physics, and Optics
Reference39 articles.
1. Survey on Free Space Optical Communication: A Communications Theory Perspective;Khalingi;IEEE Commun. Surv. Tutor.,2014 2. Valley, M.T., and Vorontsov, M. (2004). Atmospheric Tracking, Imaging and Compensation, Proceedings of the SPIE 49th Annual Meeting, Denver, CO, USA, 2–6 August 2004, SPIE. 3. Lionis, A., Peppas, K., Nistazakis, H.E., Tsigopoulos, A.D., and Cohn, K. (2020). Experimental Performance Analysis of an Optical Communication Channel over Maritime Environment. Electronics, 9. 4. Statistical Modeling of Received Signal Strength for an FSO Channel over Maritime Environment;Lionis;Opt. Commun.,2021 5. Lionis, A., Peppas, K., Nistazakis, E., Tsigkopoulos, A., and Cohn, K. (2021). RSSI probability density functions comparison using Jenshen-Shannon divergence and Pearson distribution. Technologies, 9.
|
|