Application of artificial neural networks in the problem of periodic signal detection
Vestnik of Saint Petersburg University. Mathematics. Mechanics. Astronomy · 2026 · 13 (2) · 321-333
DOI: 10.21638/spbu01.2026.213 ↗Kurz gefasst
Fachartikel in der Zeitschrift Vestnik of Saint Petersburg University. Mathematics. Mechanics. Astronomy, 2026. Verfasst von Vladislav V. Topinskiy und 1 weiteren Autor:innen.
Abstract (Originalsprache)
A convolutional neural network is constructed solving the task of detecting a sinusoidal signal in an uneven time series with periodic gaps, also containing Gaussian white noise. It is shown that this model approximates the posterior probability of the signal existence, and the neural network model works much faster than the classical Bayesian numerical modeling. Some pitfalls of this method are discussed, in particular the problem of specifying initial approximations for neural network weights, as well as possible development prospects.
Zitieren
APA
Topinskiy, V. V., & Baluev, R. V. (2026). Application of artificial neural networks in the problem of periodic signal detection. Vestnik of Saint Petersburg University. Mathematics. Mechanics. Astronomy, 13(2), 321-333. https://doi.org/10.21638/spbu01.2026.213
Vancouver
Topinskiy VV, Baluev RV. Application of artificial neural networks in the problem of periodic signal detection. Vestnik of Saint Petersburg University. Mathematics. Mechanics. Astronomy. 2026;13(2):321-333. doi:10.21638/spbu01.2026.213
BibTeX
@article{Topinskiy2026,
author = {Topinskiy, Vladislav V. and Baluev, Roman V.},
title = {Application of artificial neural networks in the problem of periodic signal detection},
journal = {Vestnik of Saint Petersburg University. Mathematics. Mechanics. Astronomy},
year = {2026},
volume = {13},
number = {2},
pages = {321--333},
doi = {10.21638/spbu01.2026.213}
}
EUVIMED