<div>
	Using machine learning technologies to solve the problem of classifying infrasound background monitoring signals</div>

Frolov Ivan N., Nikolai G. Kudryavtsev, Safonova Varvara Yu. , Kudin Dmitry V.

2024 , VOLUME 24, NUMBER 1 ( january-february )

ISSN 2226-1494 (print), ISSN 2500-0373 (online)

Publications

Editor-in-Chief

Nikiforov
Vladimir O.
D.Sc., Prof.

Partners

doi: 10.17586/2226-1494-2024-24-1-156-164

Using machine learning technologies to solve the problem of classifying infrasound background monitoring signals

I. N. Frolov, N. G. Kudryavtsev, V. Y. Safonova, D. V. Kudin

Read the full article

Article in Russian

For citation:

Frolov I.N., Kudryavtsev N.G., Safonova V.Yu., Kudin D.V. Using machine learning technologies to solve the problem of classifying infrasound background monitoring signals. Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2024, vol. 24, no. 1, pp. 156–164 (in Russian). doi: 10.17586/2226-1494-2024-24-1-156-164

Abstract

It is widely known that among sound signals generated by natural and anthropogenic phenomena, the most long-lived are waves of frequency less than 20 Hz, called infrasound. This property allows tracking at a distance by infrasound monitoring the occurrence of high-energy events on regional scales (up to 200–300 km). At the same time, the separation of useful infrasound signals from background noise is a non-trivial task in real-time and post-facto signal processing. In this paper we propose a new method for classification of specific signals in infrasound monitoring data using Shannon permutation entropy and vectors of frequency distribution of occurrence frequencies of permutations of consecutive sample values of rank 3 (number of permutation elements). To evaluate the validity of the proposed entropy-based classification method, two machine learning methods — random forest method and classical neural network approach — implemented in Python language using Scikit-lean, TensorFlow and Keras libraries were used. The classification quality was evaluated against the traditional frequency-based method of class extraction based on Fourier transform. Recognition was performed on the prepared infrasound monitoring data in the Altai Republic. The results of computational experiment on the separation of 5 classes of signals showed that classification by the proposed method gives the same results of recognition by neural network with in comparison with frequency classification of the original data; the recognition accuracy was 51–58 %. For the random forests method, the recognition accuracy of frequency classes was slightly higher: 51 % vs. 45 % for classes using the permutation entropy method. The analysis of the results of the computational experiment shows sufficient competitiveness of the method of classification by permutation entropy in the recognition of infrasound signals. In addition, the proposed method is much easier to implement for inline signal processing in low- consumption microcontroller systems. The next step is to test the method at infrasound signal registration points and as part of the infrasound monitoring data processing system for real-time event detection.

Keywords: machine learning, random forest model, artificial neural network, infrasound, permutation entropy, classification of time series fragments

Acknowledgements. The research was carried out using a grant from the Russian Science Foundation (RSF) and the Ministry of Education and Science of the Altai Republic No. 23-21-10087.

References

Sandheep P., Vineeth S., Poulose M., Subha D.P. Performance analysis of deep learning CNN in classification of depression EEG signals. Proc. of the TENCON 2019 - 2019 IEEE Region 10 Conference (TENCON), 2019, pp. 1339–1344. https://doi.org/10.1109/TENCON.2019.8929254
Heyden M. Classification of EEG data using machine learning techniques. Master's thesis. Department of Automatic Control, Lund University, 2016.
Guerrero M.C., Parada J.S., Espitia H.E. EEG signal analysis using classification techniques: Logistic regression, artificial neural networks, support vector machines, and convolutional neural networks. Heliyon, 2021, vol. 7, no. 6, pp. e07258. https://doi.org/10.1016/j.heliyon.2021.e07258
Siddharth T., Tripathy R., Pachori R. Discrimination of focal and non-focal seizures from EEG signals using sliding mode singular spectrum analysis. IEEE Sensors Journal, 2019, vol. 19, no. 24, pp. 12286–12296. https://doi.org/10.1109/JSEN.2019.2939908
Sharma R., Sircar P., Pachori R.B. A new technique for classification of focal and nonfocal EEG signals using higher-order spectra. Journal of Mechanics in Medicine and Biology, 2019, vol. 19, no. 1, pp. 1940010. https://doi.org/10.1142/S0219519419400104
Narayan Y. SEMG signal classification using KNN classifier with FD and TFD features. Materials Today: Proceedings, 2021, vol. 37, part 2, pp. 3219–3225. https://doi.org/10.1016/j.matpr.2020.09.089
Shi Y., Davaslioglu K., Sagduyu Y.E., Headley W.C., Fowler M., Green G. Deep learning for RF signal classification in unknown and dynamic spectrum environments. Proc. of the 2019 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN), 2019, pp. 1–10. https://doi.org/10.1109/DySPAN.2019.8935684
Shaker A.M., Tantawi M., Shedeed H.A., Tolba M.F. Generalization of convolutional neural networks for ECG classification using generative adversarial network. IEEE Access, 2020, vol. 8, pp. 35592–35605. https://doi.org/10.1109/ACCESS.2020.2974712
Donmez H., Ozkurt N. Emotion classification from EEG signals in convolutional neural networks. Proc. of the 2019 Innovations in Intelligent Systems and Applications Conference (ASYU), 2019, pp. 1–6. https://doi.org/10.1109/ASYU48272.2019.8946364
Kuzin D.A., Statsenko L.G., Anisimov P.N., Smirnova M.M. Applying machine learning methods to acoustic signal classification using spectrum characteristics. Proceedings of Saint Petersburg Electrotechnical University, 2021, no. 3, pp. 48–54. (in Russian)
Wu Y., Zhang J., Chen X., Zou S., Yang M. Research on the recognition of infrasound signal of nuclear explosion by SVM and CNN. IOP Conference Series: Earth and Environmental Science, 2020, vol. 610, pp. 012010. https://doi.org/10.1088/1755-1315/610/1/012010
Traversaro F., Redelico F.O., Risk M.R., Frery A.C., Rosso O.A. Bandt-Pompe symbolization dynamics for time series with tied values: A data-driven approach. Chaos, 2018, vol. 28, no. 7, pp. 075502. https://doi.org/10.1063/1.5022021
Konstantinov A.V., Leshtaev V.S. Preparing the feature space of a neural network model for the classification of seismoacoustic events of the PROGNOZ-ADS system. Proc. of the 5th conference of the International Scientific School of Academician K.N. Trubetskoy from the Russian Academy of Sciences “Problems and prospects for the integrated development and conservation of the earth interior”. Moscow, IPKON RAS, 2022, pp. 108–112. (in Russian)
Jiao S., Geng B., Li Y., Zhang Q., Wang Q. Fluctuation-based reverse dispersion entropy and its applications to signal classification. Applied Acoustics, 2021, vol. 175, pp. 107857. https://doi.org/10.1016/j.apacoust.2020.107857
Kuznetcov I.A. Machine learning methods and algorithms for preprocessing and classification of semi-structured text data in scientific recommendation systems. Dissertation for the degree of candidate of technical sciences. Moscow, MEPhI, 2019, 127 p. (in Russian)

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License