Detection of solar flare effects in the amplitude variations of VLF radio signals using an autoencoder
https://doi.org/10.55648/1998-6920-2025-19-1-28-44
Abstract
Detection of anomalies in changes in radio signal parameters is one of the tools for studying the dynamic properties of a radio communication channel. The development of machine learning technologies allows us to take a new approach to solving pressing problems in radio physics. The purpose of this work is to develop an algorithm for searching for the effects of solar flares in variations in the amplitude of VLF radio signals using machine learning
methods. The developed models of a neural network based on an autoencoder are presented for automatic detection of anomalous amplitude changes based on the propagation data of VLF
radio waves on the JJI (Japan, 22.2 kHz) – Irkutsk (receiving point in Teploenergetik) path. After training, the neural network reconstructs the daily variation of the signal amplitude so that the output data closely matches the propagation under quiet conditions. Anomalous signal
changes associated with ionospheric disturbances during solar flares are detected by determining the reconstruction error and comparing it with a calculated threshold value. The threshold is determined by calculating the root mean square value of the reconstruction error of the training data set. The results of using the autoencoders proposed by the authors to search for
methods. The developed models of a neural network based on an autoencoder are presented for automatic detection of anomalous amplitude changes based on the propagation data of VLF
radio waves on the JJI (Japan, 22.2 kHz) – Irkutsk (receiving point in Teploenergetik) path. After training, the neural network reconstructs the daily variation of the signal amplitude so that the output data closely matches the propagation under quiet conditions. Anomalous signal
changes associated with ionospheric disturbances during solar flares are detected by determining the reconstruction error and comparing it with a calculated threshold value. The threshold is determined by calculating the root mean square value of the reconstruction error of the training data set. The results of using the autoencoders proposed by the authors to search for
anomalies have shown their effectiveness (the overall accuracy was close to 90%). An important advantage of the method is the ability to perform an assessment using data only for the measurement day in question, which eliminates the influence of anomalies present on neighboring days.
About the Authors
Khac Hoang Duong Nguyen
Irkutsk National Research Technical University
Viet Nam
Viet Nam
postgraduate student of the Department of Radio Electronics and Telecommunication Systems,
Irkutsk National Research Technical University
Aleksandr Sergeevich Poletaev
Irkutsk National Research Technical University
Russian Federation
Russian Federation
Cand. Sci. (Phys.–Math.), associate professor of the Department of Radio Electronics and
Telecommunication Systems, Irkutsk National Research Technical University
References
1. References
2. Savitskiy D. E., Dunaev M. E., Zaitsev K. S. "Vyyavlenie anomalii pri obrabotke potokovykh dannykh v realnom vremeni" [Anomaly Detection in Real-Time Stream Data Processing]. International Journal of Open Information Technologies, 2022, Vol. 14, No. 6, pp. 70–75.
3. Sarvani A., Venugopal B., Devarakonda N. "Anomaly Detection Using K-means Approach and Outliers Detection Technique." In: Ray K., Sharma T., Rawat S., Saini R., Bandyopadhyay A. (eds) "Soft Computing: Theories and Applications." Advances in Intelligent Systems and Computing, 2019, Vol. 742.
4. Krascheninnikov V. R., Kuvaiskova Yu. E., Malenova O. E., Subbotin A. Yu. "Obnaruzhenie anomalii na kvaziperiodicheskikh protsessakh, tsilindricheskikh i krugovykh izobrazheniyakh i ikh sistemakh" [Detection of Anomalies in Quasiperiodic Processes, Cylindrical and Circular Images, and Their Systems]. Avtomatizatsiya protsessov upravleniya, 2022, No. 4, pp. 67–75.
5. Tokarev D. A., Pogreban S. V. "Anomalii trafika v IP setyakh i ikh obnaruzhenie" [Traffic Anomalies in IP Networks and Their Detection]. Izvestiya Instituta Inzhenernoy Fiziki, 2013, Vol. 4, pp. 13–37.
6. Tokmachev D. A., Poletaev A. S., Bezrukin A. G., Chensky A. G., Zasenko V. E., Gubin N. A. "Sistema sinkhronizatsii sverkhlinnovolnovogo interferometra" [Synchronization System of an Extremely Long-Wave Interferometer]. Pribory i Tekhnika Eksperimenta, 2014, No. 5, pp. 77–84.
7. Podgorny A. I., Podgorny I. M. "Magnitnyy potok aktivnoy oblasti i yego svyaz' so vspyshkami" [Magnetic Flux of the Active Region and Its Relationship with Flares]. Astronomicheskiy Zhurnal, 2011, Vol. 88, No. 7, pp. 684.
8. Podgorny I. M., Podgorny A. I. "Magnetic Field Distribution in the Flare Productive Active Region NOAA 10720." Journal of Atmospheric and Solar-Terrestrial Physics, 2013, Vol. 92, p. 59.
9. Podgorny A. I., Podgorny I. M., Meshalkina N. S. "Dinamika magnitnogo polya aktivnykh oblastey v predvspyshchnom sostoyanii i vo vremya solnechnykh vspyshk" [Dynamics of Magnetic Fields in Active Regions in the Pre-Flare State and During Solar Flares]. Astronomicheskiy zhurnal (Astronomy Reports), 2015, vol. 92, no. 8, pp. 669.
10. Kotik D. S., Polyakov S. V., Yashnov V. A. "Vozbuzhdenie volnovoda Zemlya-ionosfera nizkochastotnymi istochnikami, raspolozhennymi v neodnorodnoy ionosfere" [Excitation of Earth-Ionosphere Waveguide by Low-Frequency Sources Located in the Inhomogeneous Ionosphere]. Izvestiya Vuzov. Radiofizika, 1978, Vol. 21, No. 7, pp. 938–944.
11. Zhanaeva S. B. "K voprosu o podgotovke dannykh pri razrabotke modeli neyronnoy seti" [Data Preparation in Neural Network Model Development]. Vestnik SibGUTI, 2022, Vol. 16, No. 4, pp. 69-79.
12. Hastie T., Tibshirani R., Friedman J. "The Elements of Statistical Learning: Data Mining, Inference, and Prediction." Springer Series in Statistics, 2nd Edition. NY: Springer, 2017, 764 p.
13. Gras D. "Data Science. Nauka o dannykh s nulya: per. s angl." SPb: BKhV-Peterburg, 2017, 336 s.
14. Qu C. et al. "Predictive Anomaly Detection for Marine Diesel Engine based on Echo State Network and Autoencoder." Energy Reports, Elsevier Ltd, 2022, Vol. 8. P. 998–1003
15. Juan Terven, Wang H. et al. Anomaly detection for hydropower turbine unit based on variational modal decomposition and deep autoencoder // Energy Reports. Elsevier Ltd, 2021. Vol. 7. P. 938–946.
16. Diana M. Cordova-Esparza, Alfonso Ramirez-Pedraza, Edgar A. Chavez-Urbiola. Loss Functions and Metrics in Deep Learning [Электронный ресурс]. URL: https://arxiv.org/abs/2307.02694. (accessed: 28.09.2023).
Review
For citations:
Nguyen Kh., Poletaev A.S. Detection of solar flare effects in the amplitude variations of VLF radio signals using an autoencoder. The Herald of the Siberian State University of Telecommunications and Information Science. 2025;19(1):28-44. (In Russ.) https://doi.org/10.55648/1998-6920-2025-19-1-28-44
Views:
124
Email this article 