Loading…

The Parameterization Method of Discrete VLF Chorus Emissions

We present a method for automatic search and parameterization of discrete elements of very low-frequency (VLF) chorus emissions. The method is based on the processing of dynamic spectrograms using a special scanning algorithm, which is intended for seeking the discrete elements of VLF chorus emissio...

Full description

Saved in:
Bibliographic Details
Published in:Radiophysics and quantum electronics 2019-08, Vol.62 (3), p.159-173
Main Authors: Larchenko, A. V., Demekhov, A. G., Kozelov, B. V.
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present a method for automatic search and parameterization of discrete elements of very low-frequency (VLF) chorus emissions. The method is based on the processing of dynamic spectrograms using a special scanning algorithm, which is intended for seeking the discrete elements of VLF chorus emissions and calculation of their parameters. We propose to create the optimal dynamic spectrograms for the scanning algorithm by using short-time Fourier transform. The paper gives general recommendations for calculation of spectrograms and their preprocessing. The scanning algorithm is based on processing of the dynamic-spectrogram images by using the methods of mathematical morphology. The developed method was tested for several cases of chorus emissions observed by the Van Allen Probes spacecraft. It is shown that the total errors related to “false positive” detection and missing the target are about 10% of the elements visible to the human eye when the optimal parameters of the scanning algorithm are used and intense discrete elements are processed. The method can be applied to both spacecraft and ground-based wave data. The results of using the method can be employed for verification of physical theories of the formation of chorus emissions and determining their statistical properties.
ISSN:0033-8443
1573-9120
DOI:10.1007/s11141-019-09964-z