The ionosphere, a region of geospace located 60 to 1,000 kilometers above Earth, disrupts radio signal transmission from Global Navigation Satellite Systems (GNSS) via its electrically charged particles. These obstacles pose challenges to the increased accuracy required by these systems in both research and practical applications such as autonomous driving or precise satellite orbit determination.
Ionosphere models and fluctuating dynamic charge distribution can be used to compensate for ionospheric delays, which are a major source of error in GNSS applications. The new model of the ionosphere was presented by researchers Artem Smirnov and Yuri Shprits from the German Geosciences Research Center GFZ. This model, based on artificial neural networks and 19 years of satellite measurement data, was published in the journal. Scientific Reports.
In particular, the upper ionosphere can reconfigure the electron-rich upper part of the ionosphere much more accurately than before. It is therefore an important basis for advances in ionospheric research, such as in the study of electromagnetic wave propagation or in the analysis of certain space weather phenomena.
Earth’s ionosphere is a region of the upper atmosphere about 60 to 1,000 kilometers high. It is governed by charged particles such as electrons and positive ions caused by the sun’s radiation activity – hence the name. The ionosphere is important for many scientific and industrial applications because charged particles affect the propagation of electromagnetic waves such as radio signals.
The so-called ionospheric delay in the propagation of radio signals is one of the most important sources of interference for satellite navigation. This is proportional to the density of electrons in the traveled space. Therefore, a good knowledge of electron density can help smooth out the signals. The upper region of the ionosphere above 600 kilometers is particularly interesting because 80 percent of the electrons are concentrated in this upper ionosphere.
The problem is that electron density varies greatly depending on longitude and latitude on Earth, time of day and year, and solar activity. This makes it difficult to reconstruct and predict, for example, the basis of rectification of radio signals.
previous models
There are several approaches for modeling electron density in the ionosphere, including the IRI International Ionosphere Reference Model, which has been recognized since 2014. It is an empirical model that establishes a relationship between input and output variables based on the statistical analysis of observations. . However, a significant region of the upper ionosphere still has weaknesses due to the limited scope of previously collected observations in this region.
Recently, however, a large amount of data has become available for this field. Therefore, machine learning (ML) approaches are particularly well suited to extract patterns from it for complex nonlinear relationships.
Source: Port Altele
