A.O. Kasyanov1, M.V. Potipak2, A.N. Evtushenko3

1 Southern Federal University (Rostov-on-Don, Russia)

1-3 FSUE «RNIIRS» FRPC (Rostov-on-Don, Russia)

The fundamental possibility of radar specific emitter identification (SEI) is due to the presence of unintentional intra-pulse modulation in the probing signal. Unintentional pulse modulation is a subtle change in the radar signal caused by a combination of non-functional modulations that form a unique set of radar signal features, called “fingerprint”. A review of the current state of signal identification approaches is made. It is noted that it is inappropriate to use neural networks for identification, since in real conditions the possibility of generating sufficient sets of experimental data is either absent or difficult, and the use of modeling to generate a training set is impractical due to the high complexity of model to build. It is shown that the selection of unique features in the probing radar signals using higher-order statistics (cumulants) does not allow stable radar identification. The paper proposes an identification method based on the assessment of the similarity of the radar probing signal bispectra by the criterion of structural similarity.

Kasyanov A.O., Potipak M.V., Evtushenko A.N. Radar identification based on unintentional intra-pulse modulation feature extraction. Radiotekhnika. 2023. V. 87. № 2. P. 73−83. DOI: https://doi.org/10.18127/j00338486-202302-11 (In Russian)

1. Akyon F.C., et al. Classification of intra-pulse modulation of radar signals by feature fusion based convolutional neural networks. 2018 26th European Signal Processing Conference (EUSIPCO). IEEE. 2018. С. 2290-2294.
2. Svarc M., Brnak L., Richterova M. Perspective methods of radar pulse classification. International Conference on Military Technologies (ICMT) 2015. IEEE. 2015. Р. 1-8.
3. Kauppi J.P., Martikainen K., Ruotsalainen U. Hierarchical classification of dynamically varying radar pulse repetition interval modulation patterns. Neural Networks. 2010. V. 23. № 10. Р. 1226-1237.
4. Svensson A. Classification of Radar Emitters Based on Pulse Repetition Interval using Machine Learning. 2022.
5. Perdoch J., Ochodnicky J., Matousek Z. ELINT objects identification based on intra-pulse modulation classification. Big Data and AI for Military Decision Making-IST-160 Specialists' Meeting. 2018.
6. Seddighi Z., Ahmadzadeh M.R., Taban M.R. Radar signals classification using energy‐time‐frequency distribution features. IET Radar, Sonar & Navigation. 2020. V. 14. № 5. Р. 707-715.
7. Nguyen H.Q., Ngo D.T., Van Long Do. Deep Learning for Radar Pulse Detection. ICPRAM. 2019. Р. 32-39.
8. Wang X., et al. Convolutional neural network applied to specific emitter identification based on pulse waveform images. IET Radar, Sonar & Navigation. 2020. V. 14. № 5. Р. 728-735.
9. Wang S., et al. Deep Feature Autoextraction Method for Intrapulse Data of Radar Emitter Signal. Mobile Information Systems. 2021. V. 2021. Р. 1-6.
10. Zhou Y., et al. Specific emitter identification via bispectrum-radon transform and hybrid deep model. Mathematical Problems in Engineering. 2020. V. 2020.
11. Dong X., et al. Radar specific emitter recognition based on DBN feature extraction. Journal of Physics: Conference Series. IOP Publishing. 2019. V. 1176. № 3. Р. 032025.
12. Dong P., et al. Study for classification and recognition of radar emitter intra-pulse signals based on the energy cumulant of CWD. Journal of Ambient Intelligence and Humanized Computing. 2021. V. 12. С. 9809-9823.
13. Bustos G.H., Segnorile H.H. Extended method for statistical signal characterization using moments and cumulants: Application to recognition of pattern alterations in pulse-like waveforms employing artificial neural networks. arXiv preprint arXiv:2212.14783. 2022.
14. Wang X., et al. Radar emitter recognition based on the energy cumulant of short time Fourier transform and reinforced deep belief network. Sensors. 2018. V. 18. № 9. Р. 3103.
15. Simic M., Stankovic M., Orlic V.D. Automatic Modulation Classification of Real Signals in AWGN Channel Based on Sixth-Order Cumulants. Radioengineering. 2021. V. 30. № 1.
16. Swami A., Mendel J.M., Nikias C.L. Higher-order spectral analysis toolbox. The Mathworks Inc. 1998. V. 3. Р. 22-26.
17. Chen T., Jin W., Li J. Feature extraction using surrounding-line integral bispectrum for radar emitter signal. 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence). IEEE. 2008. Р. 294-298.
18. Chen Y., Wu Z.L., Lei Y.K. Individual Identification of Radar Emitters Based on a One-Dimensional LeNet Neural Network. Symmetry. 2021. V. 13. № 7. Р. 1215.
19. Chen Y., et al. High-Performance Computational Recognition of Communication Signals Based on Bispectral Quadratic Feature Model. Computational Intelligence and Neuroscience. 2022. V. 2022.
20. Han J., et al. A new method for sorting radar signal based on entropy features. 2018 International Conference on Computer Science, Electronics and Communication Engineering (CSECE 2018). Atlantis Press. 2018. Р. 1-4.
21. Pinilla S., Mishra K.V., Sadler B.M. Unique Bispectrum Inversion for Signals with Finite Spectral/Temporal Support. arXiv preprint arXiv:2111.06479. 2021.
22. Shuhong J., Mingzhu L., Weisheng D. Evaluation of jamming effect on SAR based on textural feature. Proc. Int. Conf. Inf. Commun. Manage. 2012. Р. 13-18.
23. Lobach V.T., Kas'janov A.O., Potipak M.V., Sumatohin K.V. Osobennosti cifrovogo formirovanija diagrammy napravlennosti MIMO-radara. Antenny». 2016. №7. S. 47-52 (In Russian).
24. Beljaev A.V., Kas'janov A.O., Popov Ju.G. Izmerenie parametrov i identifikacija OFDM-signalov». Antenny. 2018. № 4. S. 42-50 (In Russian).
25. Bezverhij V., Olejnikov A., Margelov A., Surzhenko I., Bezverhij O. Sovremennye sredstva dlja passivnogo avtomaticheskogo pelengovanija i identifikacii RLS i ih nositelej. Sovremennye tehnologii avtomatizacii. 2005. № 2. S. 62-66 (In Russian).