Recognition of Low-Probability-of-Intercept Radar Signals Based on a Hybrid CNN-MLP Model
Article Sidebar
Main Article Content
Abstract
This paper proposes an intelligent recognition system for low-probability-of-intercept (LPI) radar signals based on a hybrid neural network model that combines a complex-valued convolutional neural network (Complex 1D-CNN) for processing in-phase and quadrature (IQ) components with a multilayer perceptron (MLP) for extracting parametric signal features. Unlike conventional approaches that rely on time-frequency transforms such as the short-time Fourier transform (STFT) and Choi–Williams distribution (CWD), the proposed model operates directly on complex-valued data while preserving the phase structure of the signal. To improve robustness under noisy conditions, a Hankel Cross-Attention mechanism and SNR-aware loss weighting are employed. Experiments conducted on the RadChar Baseline dataset (5 classes) achieved an accuracy of 90.0% and a macro F1-score of 0.90, outperforming single-branch CNN models and demonstrating stable performance at SNR levels down to -12 dB. The obtained results confirm the effectiveness of the proposed integrated approach for LPI radar signal recognition in noisy environments.
Article Details
Issue
Section
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
[1] L. Cohen, Time-frequency distributions – a review, Proceedings of the IEEE. 77 (1989) 941–981. https://doi.org/10.1109/5.30749
[2] H. Choi, W. Williams, Improved time-frequency representation of multicomponent signals using exponential kernels, IEEE Transactions on Acoustics, Speech, and Signal Processing. 37 (1989) 862–871. https://doi.org/10.1109/ASSP.1989.28057
[3] M. Zhang, L. Liu, M. Diao, LPI Radar Waveform Recognition Based on Time-Frequency Distribution, Sensors. 16 (2016) 1682. https://doi.org/10.3390/s16101682
[4] D.-H. Park, M.-W. Jeon, D.-M. Shin, H.-N. Kim, LPI Radar Detection Based on Deep Learning Approach with Periodic Autocorrelation Function, Sensors. 23 (2023) 8564. https://doi.org/10.3390/s23208564
[5] T. Wan, K.-L. Jiang, H. Ji, B. Tang, Deep learning-based LPI radar signals analysis and identification using a Nyquist Folding Receiver architecture, Defence Technology. 19 (2023) 196–209. https://doi.org/10.1016/j.dt.2021.09.019
[6] D.C. Schleher, LPI radar: fact or fiction, IEEE Aerospace and Electronic Systems Magazine. 21 (2006) 3–6. https://doi.org/10.1109/MAES.2006.1635166
[7] M. Zhang, M. Diao, L. Gao, L. Liu, Neural Networks for Radar Waveform Recognition, Symmetry. 9 (2017) 75. https://doi.org/10.3390/sym9050075
[8] J. Wan, X. Yu, Q. Guo, LPI Radar Waveform Recognition Based on CNN and TPOT, Symmetry. 11 (2019) 725. https://doi.org/10.3390/sym11050725
[9] Q. Guo, X. Yu, G. Ruan, LPI Radar Waveform Recognition Based on Deep Convolutional Neural Network Transfer Learning, Symmetry. 11 (2019) 540. https://doi.org/10.3390/sym11040540
[10] D. Quan, Z. Tang, X. Wang, W. Zhai, C. Qu, LPI Radar Signal Recognition Based on Dual-Channel CNN and Feature Fusion, Symmetry. 14 (2022) 570. https://doi.org/10.3390/sym14030570
[11] F. Xu, H. Hu, J. Mu, X. Wang, F. Zhou, D. Quan, Low Probability of Intercept Radar Signal Recognition Based on Semi-Supervised Support Vector Machine, Electronics. 13 (2024) 3248. https://doi.org/10.3390/electronics13163248
[12] Z. Qu, W. Wang, C. Hou, Radar Signal Intra-Pulse Modulation Recognition Based on Convolutional Denoising Autoencoder and Deep CNN, IEEE Access. 7 (2019) 112339–112347. https://doi.org/10.1109/ACCESS.2019.2935247
[13] W. Tao, K. Jiang, J. Li, T. Yan, T. Bin, Detection and recognition of LPI radar signals using visibility graphs, Journal of Systems Engineering and Electronics. 31 (2020) 1186–1192. https://doi.org/10.23919/JSEE.2020.000091
[14] O. Nazarevych, R. Peleshchak, V. Lytvyn, I. Peleshchak, V. Motyka, Recognition of Low Probability of Intercept Radar Signals in a Polar Coordinate System Based on a Hybrid CNN and MLP, in: 2025 IEEE 13th International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS), Gliwice, Poland, 2025, pp. 338–345. https://doi.org/10.1109/IDAACS68557.2025.11322024
[15] R.R. Yakkati, A.B. Boddu, B. Pardhasaradhi, S.R. Yeduri, L.R. Cenkeramaddi, Low Probability of Intercept Radar Signal Classification using Multi-Channel 1D-CNN, in: Proceedings of the 10th International Conference on Communication and Signal Processing (ICCSP), IEEE, 2024, pp. 1–5. https://doi.org/10.1109/ICCSP60870.2024.10543688
[16] RadChar Dataset, GitHub Repository, (2024). https://github.com/abcxyzi/RadChar (Accessed 24 April 2026)
[17] I. Lytvynenko, Method of segmentation of determined cyclic signals for the problems related to their processing and modeling, Scientific Journal of TNTU. 88 (2017) 153–169. https://doi.org/10.33108/visnyk_tntu2017.04.153