DEVELOPMENT OF A CONCEPT FOR AN ACTIVE INTEGRATED RADIO COVERT DEVICE DETECTION SYSTEM BASED ON CONVOLUTIONAL NEURAL NETWORKS AND SDR RECEIVERS
DOI:
https://doi.org/10.28925/2663-4023.2026.35.1459Keywords:
covert device detection, technical information protection, radio frequency monitoring, radio frequency fingerprint identification, software-defined radio, VOX activation, stimulus signals, convolutional neural networks, open-set classificationAbstract
Modern radio frequency means of detecting covert listening devices are predominantly passive: they register new emissions in the spectrum but cannot provoke transmission by devices that are activated by an event, such as voice (VOX activation), motion or a schedule, while commercial radio frequency fingerprint identification (RFFI) systems do not disclose their algorithms. The aim of the article is to develop a concept for the architecture of an integrated device detection system, based on affordable equipment, that combines passive RFFI with stimulus-signal emitter modules. The literature review shows that stimulation and identification of the responding source are still treated separately. A classification of detection targets by the domain of collected information and by behavioural characteristics (continuous transmission, internal, external, and time-based activation) is proposed, and the features of two generations of radio-frequency detection means are considered. A hardware architecture is developed that includes a software-defined radio (SDR) receiver with spectrum survey and targeted I/Q capture modes; acoustic, ultrasonic, infrared, and radio-frequency stimulus emitters with an interface module that records stimulus timestamps; and an antenna switch module. An inspection procedure is proposed in which stimuli are applied according to a pseudo-random schedule, and the decision is based on two features: the correlation between emission onsets and the stimulus schedule and the mismatch between the radio-frequency fingerprint and the database of legitimate devices. The software architecture uses a convolutional neural network as a feature extractor for open-set classification. Simulation shows that with six stimuli a VOX-activated device is detected with a probability of 0.93 at a false alarm probability of no more than 0.01 per session lasting about 2 minutes, provided that each frequency is revisited at least every 0.5 s, and that stimulus-synchronised capture improves the identification of analogue transmitters (area under the ROC curve of 0.75 versus 0.59 at a signal-to-noise ratio of 20 dB). However, given the model's current identification quality, fusing the two features does not reduce the number of false alarms; improving the reliability of RFFI is therefore identified as the key direction for further research.
Downloads
References
Samosud, Z. O. (2015). Bugs detection using acoustic location. Ukrainian Scientific Journal of Information Security, 21(3), 263—268. https://doi.org/10.18372/2225-5036.21.9689
Voronin, M. O. (2022). Methods of detecting radio broadcast at objects of information activity. Modern Information Security, 52(4). https://doi.org/10.31673/2409-7292.2022.040010
Klymenko, K. O., Kostenko, O. V., & Ilchenko, O. M. (2020). General classification of means of covert obtaining of information and methods of their detection. Scientific Notes of Taurida National V.I. Vernadsky University. Series: Technical Sciences, 1(6), 6–15. https://doi.org/10.32838/tnu-2663-5941/2020.6-1/02
Konotopets, M. M., Smolkov, O. Yu., & Pshenychna, Yu. O. (2025). Enhancing the capabilities of search complexes in detecting optoelectronic elements of covert information gathering devices at information activity sites. Modern Information Technologies in the Sphere of Security and Defence, 53(2), 60–69. https://doi.org/10.33099/2311-7249/2025-53-2-60-69
Sami, S., Tan, S. R. X., Sun, B., & Han, J. (2021). LAPD: Hidden Spy Camera Detection using Smartphone Time-of-Flight Sensors. Proceedings of the 19th ACM Conference on Embedded Networked Sensor Systems, 288—301. https://doi.org/10.1145/3485730.3485941
Chaman, A., Wang, J., Sun, J., Hassanieh, H., & Roy Choudhury, R. (2018). Ghostbuster: Detecting the Presence of Hidden Eavesdroppers. Proceedings of the 24th Annual International Conference on Mobile Computing and Networking, MobiCom ’18, 337—351. https://doi.org/10.1145/3241539.3241580
Jagannath, A., Jagannath, J., & Kumar, P. S. P. V. (2022). A comprehensive survey on radio frequency (RF) fingerprinting: Traditional approaches, deep learning, and open challenges. Computer Networks, 109455. https://doi.org/10.1016/j.comnet.2022.109455
Liu, Y., Wang, J., Li, J., Niu, S., & Song, H. (2021). Machine Learning for the Detection and Identification of Internet of Things (IoT) Devices: A Survey (Version 1). arXiv. https://doi.org/10.48550/ARXIV.2101.10181
Flak, P. (2022). Hardware-Accelerated real-time spectrum analyzer with a broadband fast sweep feature based on the cost-effective SDR platform. IEEE Access, 10, 110934–110946. https://doi.org/10.1109/access.2022.3215800
McTaggart, B., Maclellan, A., Atimati, E., Nyasulu, L. B. T., Crockett, L. H., Crawford, D., & Stewart, R. W. (2026). A Reconfigurable RFSoC-Based Spectrum Sensing Platform Using a Polyphase Channeliser and Distributed Publisher-Subscriber Framework. ICC 2026 - IEEE International Conference on Communications, 1—6. https://doi.org/10.1109/ICC59461.2026.11586939
Bae, Y., Lee, H., Park, H., Jang, W.-h., & Jang, B.-J. (2026). Edge CA-CFAR Data Reduction for Bandwidth-Efficient Real-Time Wideband Spectrum Sensing on Low-Cost SDRs. Sensors, 26(14), 4468. https://doi.org/10.3390/s26144468
Bhattacharya, A., Zhan, C., Maji, A., Gupta, H., Das, S. R., & Djuric, P. M. (2021). Selection of sensors for efficient transmitter localization. IEEE/ACM Transactions on Networking, 1–13. https://doi.org/10.1109/tnet.2021.3104000
Svanstrom, F., Englund, C., & Alonso-Fernandez, F. (2021). Real-Time Drone Detection and Tracking With Visible, Thermal and Acoustic Sensors. 2020 25th International Conference on Pattern Recognition (ICPR), 7265—7272. https://doi.org/10.1109/ICPR48806.2021.9413241
Acustek. (2026). SigDNA - universal RF TSCM & attribution platform | acustek. https://acustek.com/product/tscm/sigdna
Great Scott Gadgets. (2017). Release 2017.02.1 · greatscottgadgets/hackrf. GitHub. https://github.com/greatscottgadgets/hackrf/releases/tag/v2017.02.1
DigiScan-Labs. (2009). DigiScan (EX, version 2.4) [Computer software].
Sankhe, K., Belgiovine, M., Zhou, F., Angioloni, L., Restuccia, F., D'Oro, S., Melodia, T., Ioannidis, S., & Chowdhury, K. (2020). No radio left behind: Radio fingerprinting through deep learning of physical-layer hardware impairments. IEEE Transactions on Cognitive Communications and Networking, 6(1), 165–178. https://doi.org/10.1109/tccn.2019.2949308
Shen, G., Zhang, J., Marshall, A., & Cavallaro, J. R. (2022). Towards scalable and channel-robust radio frequency fingerprint identification for lora. IEEE Transactions on Information Forensics and Security, 17, 774–787. https://doi.org/10.1109/tifs.2022.3152404
Hanna, S., Karunaratne, S., & Cabric, D. (2020). Open set wireless transmitter authorization: Deep learning approaches and dataset considerations. IEEE Transactions on Cognitive Communications and Networking, 1. https://doi.org/10.1109/tccn.2020.3043332
Harris, C. R., Millman, K. J., van der Walt, S. J., Gommers, R., Virtanen, P., Cournapeau, D., Wieser, E., Taylor, J., Berg, S., Smith, N. J., Kern, R., Picus, M., Hoyer, S., van Kerkwijk, M. H., Brett, M., Haldane, A., del Río, J. F., Wiebe, M., Peterson, P., ... Oliphant, T. E. (2020). Array programming with NumPy. Nature, 585(7825), 357–362. https://doi.org/10.1038/s41586-020-2649-2
Virtanen, P., Gommers, R., Oliphant, T. E., Haberland, M., Reddy, T., Cournapeau, D., Burovski, E., Peterson, P., Weckesser, W., Bright, J., Van Der Walt, S. J., Brett, M., Wilson, J., Millman, K. J., Mayorov, N., Nelson, A. R. J., Jones, E., Kern, R., Larson, E., … Vázquez-Baeza, Y. (2020). SciPy 1.0: Fundamental algorithms for scientific computing in Python. Nature Methods, 17(3), 261–272. https://doi.org/10.1038/s41592-019-0686-2
PyTorch: An imperative style, high-performance deep learning library. (2019). In Proceedings of the 33rd International Conference on Neural Information Processing Systems. Curran Associates Inc.
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Іван Опірський, Артем Стефанків, Георгій Лобан, Олександр Гимза
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
