A method for spatial feature extraction of ionospheric total electron content using conditional generative adversarial network

Abstract

The invention discloses a method for realizing spatial feature extraction of ionospheric total electron content by using conditional generative adversarial network. In the method, a deep learning model of conditional encoding and decoding generative adversarial neural network with space considerations is firstly designed. The designed model Incorporating an encoder-decoder structure with the idea of ​​adversarial learning, the deep features of the input ionospheric total electron content sampling spatial data and their complex interactions with local structural patterns can be learned. The validity of the method is proved by the example analysis. Compared with the traditional method, this method successfully completes the extraction and restoration of the spatial characteristics of the ionospheric total electron content when faced with complex ionospheric spatial distribution characteristics, and has higher accuracy in realizing the spatial estimation of the ionospheric total electron content and a smaller mean square error.

Description

technical field [0001] The invention belongs to the field of the ionosphere, and in particular relates to a method for extracting spatial features of the total electron content of the ionosphere by using a conditional generation confrontation network. Background technique [0002] The ionosphere is the part of the Earth's thermosphere ionized by solar radiation, filled with a large number of charged particles. The ionosphere is the undulating scattering medium of space radio signals, which will cause signal amplitude attenuation, phase delay, etc., and is an important source of space ranging errors for satellite navigation signals. The number of charged particles in the ionosphere is described by the total electron content of the ionosphere. When the ionosphere is affected by solar activities, geomagnetic activities, etc., it will cause fluctuations in the total electron content of the ionosphere. At present, ionospheric disturbance is a hot and difficult research topic in ...

AI Technical Summary

Problems solved by technology

In fact, the spatial distribution of the total electron content in the ionosphere is relatively complex. Affected by factors such as the solar radiation area and the distribution of geomagnetic intensity, its spatial distribution has strong nonlinearity, especially when the ionosphere is disturbed, it will produce strong The spatial gradient of the ionosphere further weakens the spatial linear correlation of the ionosphere. At this time, if the traditional feature extraction method is still used, it will inevitably cause a large error

The error of the total electron content of the ionosphere may be further amplified in the application and become a navigation ranging error, reducing the use accuracy of related applications

Images