An improved neural-network to estimate the inputs of Rino's ionospheric scintillation model

Abstract

Ionospheric scintillation is a well-known effect that occurs when electromagnetic waves pass through the ionosphere, leading to rapid fluctuations in the phase and intensity of the received signal. In 1979 Charles Rino introduced a theory to compute the expected ionospheric scintillation. However, Rino’s model requires knowing some input variables related to the physical properties of the ionosphere’s plasma density irregularities. WBMOD model was especially developed to provide these parameters from climatological data as a function of several environmental conditions; however, the use of this model requires a license. In this study, using large datasets from past studies, a neural network has been trained to estimate the main output parameters from WBMOD: the probability density function of CkL and the value of the p-slope (slope of power spectra of phase scintillation). This allows retrieving Rino’s input variable to compute the scintillation indices S4 and sf. The resulting software, called IonoSciNN, has been published as an open web application.

This work was supported in part by project GENESIS: GNSS Environmental and Societal Missions–Subproject UPC under Grant PID2021-126436OB-C21 sponsored by MCIN/AEI/10.13039/501100011033/ and in part by the IEEC INTREPID project, in which this study is contextualized.

Peer Reviewed

Postprint (published version)