Using non-linear filters to identify critical frequencies based on vertical sounding

Abstract

© 2019 IEEE. This paper presents the method of estimation and filtration of time series of critical frequencies using a single quality criterion. The method involves processing of sets of ionograms over a long period of time and representing a series of critical frequencies in the form of a wavelet series with a small number of non-zero coefficients. Filtration of the foF2 series is performed with a nonlinear filter using genetic algorithms and L1 regularization. The method is tested on the ionograms obtained using the Cyclone ionosonde owned by Kazan Federal University in 2012.