Using Machine Learning Algorithms for Accurate Received Optical Power Prediction of an FSO Link over a Maritime Environment

The performance prediction of an optical communications link over maritime environments has been extensively researched over the last two decades. The various atmospheric phenomena and turbulence effects have been thoroughly explored, and long-term measurements have allowed for the construction of s...

Full description

Bibliographic Details
Main Authors:	Antonios Lionis, Konstantinos Peppas, Hector E. Nistazakis, Andreas Tsigopoulos, Keith Cohn, Athanassios Zagouras
Format:	Article
Language:	English
Published:	MDPI AG 2021-06-01
Series:	Photonics
Subjects:	free space optical communication received signal strength indicator atmospheric turbulence refractive index structure parameter machine learning
Online Access:	https://www.mdpi.com/2304-6732/8/6/212

id	doaj-fb919e4c697e4799837398b19fe1da79
record_format	Article
spelling	doaj-fb919e4c697e4799837398b19fe1da792021-06-30T23:50:16ZengMDPI AGPhotonics2304-67322021-06-01821221210.3390/photonics8060212Using Machine Learning Algorithms for Accurate Received Optical Power Prediction of an FSO Link over a Maritime EnvironmentAntonios Lionis0Konstantinos Peppas1Hector E. Nistazakis2Andreas Tsigopoulos3Keith Cohn4Athanassios Zagouras5Information and Telecommunications Department, University of Peloponnese, 22131 Tripoli, GreeceInformation and Telecommunications Department, University of Peloponnese, 22131 Tripoli, GreeceSection of Electronic Physics and Systems, Department of Physics, National and Kapodistrian University of Athens, Panepistimiopolis Zografou, 15784 Athens, GreeceDivision of Combat Systems, Naval Operations, Sea Sciences, Navigation, Electronics & Telecommunications Sector, Hellenic Naval Academy, 18539 Pireas, GreecePhysics Department, Naval Postgraduate School, Monterey, CA 93943, USACue Health Inc., San Diego, CA 92121, USAThe performance prediction of an optical communications link over maritime environments has been extensively researched over the last two decades. The various atmospheric phenomena and turbulence effects have been thoroughly explored, and long-term measurements have allowed for the construction of simple empirical models. The aim of this work is to demonstrate the prediction accuracy of various machine learning (ML) algorithms for a free-space optical communication (FSO) link performance, with respect to real time, non-linear atmospheric conditions. A large data set of received signal strength indicators (RSSI) for a laser communications link has been collected and analyzed against seven local atmospheric parameters (i.e., wind speed, pressure, temperature, humidity, dew point, solar flux and air-sea temperature difference). The k-nearest-neighbors (KNN), tree-based methods-decision trees, random forest and gradient boosting- and artificial neural networks (ANN) have been employed and compared among each other using the root mean square error (RMSE) and the coefficient of determination (R<sup>2</sup>) of each model as the primary performance indices. The regression analysis revealed an excellent fit for all ML models, indicative of their ability to offer a significant improvement in FSO performance modeling as compared to traditional regression models. The best-performing R<sup>2</sup> model found to be the ANN approach (0.94867), while random forests achieved the most optimal RMSE result (7.37).https://www.mdpi.com/2304-6732/8/6/212free space optical communicationreceived signal strength indicatoratmospheric turbulencerefractive index structure parametermachine learning
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	Antonios Lionis Konstantinos Peppas Hector E. Nistazakis Andreas Tsigopoulos Keith Cohn Athanassios Zagouras
spellingShingle	Antonios Lionis Konstantinos Peppas Hector E. Nistazakis Andreas Tsigopoulos Keith Cohn Athanassios Zagouras Using Machine Learning Algorithms for Accurate Received Optical Power Prediction of an FSO Link over a Maritime Environment Photonics free space optical communication received signal strength indicator atmospheric turbulence refractive index structure parameter machine learning
author_facet	Antonios Lionis Konstantinos Peppas Hector E. Nistazakis Andreas Tsigopoulos Keith Cohn Athanassios Zagouras
author_sort	Antonios Lionis
title	Using Machine Learning Algorithms for Accurate Received Optical Power Prediction of an FSO Link over a Maritime Environment
title_short	Using Machine Learning Algorithms for Accurate Received Optical Power Prediction of an FSO Link over a Maritime Environment
title_full	Using Machine Learning Algorithms for Accurate Received Optical Power Prediction of an FSO Link over a Maritime Environment
title_fullStr	Using Machine Learning Algorithms for Accurate Received Optical Power Prediction of an FSO Link over a Maritime Environment
title_full_unstemmed	Using Machine Learning Algorithms for Accurate Received Optical Power Prediction of an FSO Link over a Maritime Environment
title_sort	using machine learning algorithms for accurate received optical power prediction of an fso link over a maritime environment
publisher	MDPI AG
series	Photonics
issn	2304-6732
publishDate	2021-06-01
description	The performance prediction of an optical communications link over maritime environments has been extensively researched over the last two decades. The various atmospheric phenomena and turbulence effects have been thoroughly explored, and long-term measurements have allowed for the construction of simple empirical models. The aim of this work is to demonstrate the prediction accuracy of various machine learning (ML) algorithms for a free-space optical communication (FSO) link performance, with respect to real time, non-linear atmospheric conditions. A large data set of received signal strength indicators (RSSI) for a laser communications link has been collected and analyzed against seven local atmospheric parameters (i.e., wind speed, pressure, temperature, humidity, dew point, solar flux and air-sea temperature difference). The k-nearest-neighbors (KNN), tree-based methods-decision trees, random forest and gradient boosting- and artificial neural networks (ANN) have been employed and compared among each other using the root mean square error (RMSE) and the coefficient of determination (R<sup>2</sup>) of each model as the primary performance indices. The regression analysis revealed an excellent fit for all ML models, indicative of their ability to offer a significant improvement in FSO performance modeling as compared to traditional regression models. The best-performing R<sup>2</sup> model found to be the ANN approach (0.94867), while random forests achieved the most optimal RMSE result (7.37).
topic	free space optical communication received signal strength indicator atmospheric turbulence refractive index structure parameter machine learning
url	https://www.mdpi.com/2304-6732/8/6/212
work_keys_str_mv	AT antonioslionis usingmachinelearningalgorithmsforaccuratereceivedopticalpowerpredictionofanfsolinkoveramaritimeenvironment AT konstantinospeppas usingmachinelearningalgorithmsforaccuratereceivedopticalpowerpredictionofanfsolinkoveramaritimeenvironment AT hectorenistazakis usingmachinelearningalgorithmsforaccuratereceivedopticalpowerpredictionofanfsolinkoveramaritimeenvironment AT andreastsigopoulos usingmachinelearningalgorithmsforaccuratereceivedopticalpowerpredictionofanfsolinkoveramaritimeenvironment AT keithcohn usingmachinelearningalgorithmsforaccuratereceivedopticalpowerpredictionofanfsolinkoveramaritimeenvironment AT athanassioszagouras usingmachinelearningalgorithmsforaccuratereceivedopticalpowerpredictionofanfsolinkoveramaritimeenvironment
_version_	1721350389580693504

Using Machine Learning Algorithms for Accurate Received Optical Power Prediction of an FSO Link over a Maritime Environment

Similar Items