In this study, a new multicomponent model (MCM) to determine the time variation of ionospheric parameters is suggested. The model was based on the combination of wavelets with autoregressive-integrated moving average model classes and allowed the study of the seasonal and diurnal variations of ionospheric parameters and the determination of anomalies occurring during ionospheric disturbances. To investigate in detail anomalous changes in the ionosphere, new computational solutions to detect anomalies of different scales and estimate their parameters (e.g., time of occurrence, duration, scale, and intensity) were developed based on a continuous wavelet transform. The MCM construction for different seasons and periods of solar activity was described using ionosphere critical frequency f o F2 data from Kamchatka (Paratunka Station, 52°58′ N, 158°15′ E, Institute of Cosmophysical Research and Radio Wave Propagation FEB RAS). A comparison of the MCM with the empiric International Reference Ionosphere (IRI) model and the moving median method for the analyzed region showed that the suggested method was promising for future research, since it had the advantage of providing quantitative estimates for the occurrence time, duration, and intensity of the anomalies, characterizing the ionospheric state and disturbance degree with a higher accuracy. Geomagnetic storms from 17 March and 2 October 2013 were analyzed using the suggested method, and it was shown that the ionospheric disturbances were at maximum during the strongest geomagnetic disturbances. An increase in the electron concentration in comparison with the background level, under calm or weakly disturbed geomagnetic field conditions, was identified before the analyzed magnetic storms.
The present paper discusses the empirical mode decomposition technique relative to signal denoising, which is often included in signal preprocessing. We provide some basics of the empirical mode decomposition and introduce intrinsic mode functions with the corresponding illustrations. The problem of denoising is described in the paper and we illustrate denoising using soft and hard thresholding with the empirical mode decomposition. Furthermore, we introduce a new approach to signal denoising in the case of heteroscedastic noise using a classification statistics. Our denoising procedure is shown for a harmonic signal and a smooth curve corrupted with white Gaussian heteroscedastic noise. We conclude that empirical mode decomposition is an efficient tool for signal denoising in the case of homoscedastic and heteroscedastic noise. Finally, we also provide some information about denoising applications in vibrational signal analysis.
The present paper describes the method of creating data mining algorithms from unified functional blocks. This method splits algorithms into independently functioning blocks. These blocks must have unified interfaces and implement pure functions. The method allows us to create new data mining algorithms from existing blocks and improves the existing algorithms by optimizing single blocks or the whole structure of the algorithms. This becomes possible due to a number of important properties inherent in pure functions and hence functional blocks.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.