Lestenko N.A., Walshtein K.V., Verkhova A.A.
UDK 004.8
https://doi.org/10.56408/2412-8627.2026.1.12.004
Lestenko, N.A. Modern Methods of Analyzing the Acoustic Profile of Unmanned Aerial System Using Neural Network Technologies / N.A. Lestenko, K.V. Walshtein, A.A. Verkhova // Noise Theory and Practice. – 2026. – Vol. 12, No. 1 (44). – P. 36-46. – DOI: 10.56408/2412-8627.2026.1.12.004
artificial neural network, unmanned aerial system, acoustic profile, localization, detection
The article examines the primary tasks involved in analyzing and constructing the acoustic profile of an unmanned aircraft system. The applicability of neural network technologies for solving such problems is demonstrated, and a review of current research on this topic is conducted. Classes of tasks solved by onboard equipment, as well as by external surveillance systems, are identified. For tasks addressed by external surveillance systems, the main stages are defined, at which the application of neural network technologies yields a significant increase in accuracy compared to classical methods. The limitations of using neural network technologies in analyzing audio signals via an aircraft system's onboard equipment are determined. Existing signal pre-processing methods used for noise suppression by artificial neural network models, as well as methods for processing the acquired signal, are considered. Special attention is paid to the tasks of localizing an external sound source. An experiment was conducted to detect faults based on an audio signal, leading to a conclusion about the applicability and promise of using such technologies, as well as about future directions for research.
Lestenko N.A.
Baltic State Technical University "VOENMEH", Saint Petersburg, Russia
Walshtein K.V.
Baltic State Technical University "VOENMEH", Saint Petersburg, Russia
Verkhova A.A.
Baltic State Technical University "VOENMEH", Saint Petersburg, Russia
Вальштейн К. В., Верхова А. А., Енин Ю. Ю., Гладевич А. А. Использование современных моделей искусственного интеллекта на системах с ограниченными ресурсными возможностями // Информационные технологии в высокотехнологичных производствах (ВТП) : Сборник тезисов докладов III Всероссийской молодежной научной конференции (Санкт-Петербург, 13–14 марта 2025). – Санкт-Петербург: Балтийский государственный технический университет "ВОЕНМЕХ" им. Д.Ф. Устинова, 2025. – С. 162-163.;
Лестенко Н. А., Вальштейн К. В., Верхова А. А. Современные методы построения систем искусственного интеллекта для обработки аудиосигналов // Noise Theory and Practice. – 2025. – Vol. 11, N 1(40). – С. 26-42.; Jasim, Shahad. Real Time Drone Detection Based on Acoustics Using Hybrid Deep Learning Models // Journal of Internet Services and Information Security. – 2025. – N 15. – P. 673-693.;
Glüge S. et al. Robust low-cost drone detection and classification using convolutional neural networks in low SNR environments // IEEE Journal of Radio Frequency Identification. – 2024. – Vol. 8. – P. 821-830. DOI: https://doi.org/10.1109/JRFID.2024.3487303.;
Stefanescu R. et al. WAVE-DETR Multi-Modal Visible and Acoustic Real-Life Drone Detector // Computer Vision and Pattern Recognition – 2025. – [Электронный ресурс]. – URL: https://arxiv.org/pdf/2509.09859 (дата обращения 05.10.2025).;
Berg A. P., Zhang Q., Wang M. Y. 15,500 Seconds: Lean UAV Classification Leveraging PEFT and Pre-Trained Networks // Machine Learning. – 2025. – [Электронный ресурс].– URL: https://www.arxiv.org/pdf/2506.11049v2 (дата обращения 05.10.2025).;
Araz R. O. et al. Enhancing Neural Audio Fingerprint Robustness to Audio Degradation for Music Identification // Sound. – 2025. – [Электронный ресурс]. – URL: https://arxiv.org/abs/2506.22661 (дата обращения 05.10.2025).;
Hu F., Song X., He R., et al. Sound source localization based on residual network and channel attention module. // Scienti’c Reports. – 2023. – N 13 [Электронный ресурс]. – URL: https://www.nature.com/articles/s41598-023-32657-7 (дата обращения 07.10.2025).;
Youssef K., Barakat J. M. H., Said S., Kork S. A. and Beyrouthy T. An Approach for Single-Channel Sound Source Localization// IEEE Access. – 2024. – Vol. 14. – 12 p.;
Премачандра Чинтака [и др.] Подавление звукового шума на основе GAN для обнаружения жертв на местах стихийных бедствий с помощью БПЛА // IEEE Transactions on Services Computing. – 2023. – [Электронный ресурс]. – URL: https://www.researchgate.net/publication/376147388_GAN_Based_Audio_Noise_Suppression_for_Victim_Detection_at_Disaster_Sites_with_UAV (дата обращения 09.10.2025).;
Manamperi Wageesha N., Abhayapala Thushara D., Samarasinghe Prasanga N., Zhang Jihui (Aimee). Drone audition: Audio signal enhancement from drone embedded microphones using multichannel Wiener filtering and Gaussian-mixture based post-filtering //Applied Acoustics. – 2024. – Vol. 216, N 9. – 13 p. DOI: https://doi.org/10.1016/j.apacoust.2023.109818.;
Terwilliger A. M., Siegel J. E. The ai mechanic: Acoustic vehicle characterization neural networks // Sound. – 2022. – [Электронный ресурс]. – URL: https://arxiv.org/abs/2205.09667 (дата обращения 09.10.2025).;
Tuleski, B.L., Yamaguchi, C.K., Stefenon, S.F., Coelho, L.d.S., Mariani, V.C. Audio-Based Engine Fault Diagnosis with Wavelet, Markov Blanket, ROCKET, and Optimized Machine Learning Classifiers // Sensors. – 2024. – Vol. 24, N 22. – 23 p. DOI: https://doi.org/10.3390/s24227316;
Liu W., Chen Z., Zheng M. An audio-based fault diagnosis method for quadrotors using convolutional neural network and transfer learning //2020 American Control Conference (ACC). – 2020. – [Электронный ресурс]. – URL: https://arxiv.org/abs/2003.02649 (дата обращения 09.10.2025).;
Anidjar O. H., Barak A., Ben-Moshe B., Hagai E. and Tuvyahu S. A Stethoscope for Drones: Transformers-Based Methods for UAVs Acoustic Anomaly Detection // IEEE Access – 2023.– Vol. 11 – P. 33336-33353.;
Engine Acoustic Emissions. –– 2023. – [Электронный ресурс]. – URL: https://www.kaggle.com/datasets/julienjta/engine-acoustic-emissions/data (дата обращения 03.02.2026).;
Chevtchenko S. F. et al. Drone-Based Sound Source Localization: A Systematic Literature Review // IEEE Access. – 2025. – Vol. 13. – P. 94256-94274.