The Spectral Correlation Matrix as a Fingerprint for Robust RF Signal Identification and Jamming Mitigation
Keywords:
Adaptive filtering, Cyclostationary signals, Jamming mitigation, Modulation classification, Spectral correlation matrixAbstract
Background: In an era of spectral congestion and escalating electronic warfare threats, wireless communications face persistent challenges from non-stationary interference and jamming, compromising modulation recognition and signal integrity. Traditional power spectral density (PSD) methods falter against cyclostationary distortions, necessitating advanced tools for blind detection and mitigation. Purpose: This study aims to develop and validate a unified cyclostationary framework integrating spectral correlation matrices (SCMs), principal component analysis (PCA), deep neural networks (DNNs), and adaptive suppression techniques to enhance automatic modulation recognition (AMR) and anti-jamming resilience in contested environments. Methods: Leveraging the RML2016.10a dataset augmented with AWGN and diverse jammers (tone, multi-tone, sweep, noise, pulsed), we benchmark SCMs on sinusoidal signals, apply PCA for modulation separation, train CNN-DNNs for AMR, and evaluate frequency hopping (FH), adaptive filtering, and wavelet denoising via Monte Carlo simulations (n = 1000 trials, SNR = 0–20 dB). Performance metrics include BER, accuracy, coherence peaks, and inter-class distances, analyzed through ANOVA, t-tests, and power-law regressions. Findings: SCMs reveal harmonic ridges with 1.45e06 coherence peaks, 70 dB above noise; PCA achieves 97% k-NN accuracy with 68% variance in 2D; DNNs yield 100% confusion matrix diagonals (kappa = 1.0); adaptive filtering slashes tone-jam BER from 0.4492 to 0.0459 (9.79x improvement, Δ = 0.4033), outperforming wavelet (1.28x) and FH (1.00x) across jammer types, with SCM detection at 94.9% confidence. Novelty: The hybrid SCM-PCA-DNN pipeline introduces cyclic priors into end-to-end AMR, fusing dimensionality reduction with jamming typology-specific suppression, achieving 85% BER degradation tempering and 98% F1-scores, surpassing SOTA ResNets by 4.9% while halving latency. Conclusion: Cyclostationary intelligence fortifies spectrum awareness, enabling flawless AMR and robust mitigation in low-SNR regimes, pivotal for 5G/6G and EW applications. Recommendations: Deploy adaptive ensembles for tone/sweep threats, validate via over-the-air trials, and advance ML-α autotuning for MIMO extensions to ensure equitable spectral access.
References
B. S. Goshu, “5G technology deployment: A thorough examination of the potential and difficulties for environmental sustainability, human health, and sociotechnical systems,” Budapest International Research in Exact Sciences (BirEx) Journal, vol. 8, no. 1, pp. 1–19, 2025.
W. A. Gardner, “Exploitation of spectral redundancy in cyclostationary signals,” IEEE Signal Process. Mag., vol. 8, no. 2, pp. 14–36, 1991. doi: https://doi.org/10.1109/79.81007
C. M. Spooner and R. D. Nicholls, “Spectrum sensing based on spectral correlation,” in Cognitive Radio Technology, 2nd ed., Elsevier Academic Press, 2009, pp. 371–406.
T. J. O’Shea, J. Corgan, and T. C. Clancy, “Convolutional radio modulation recognition networks,” in Engineering Applications of Neural Networks, C. Jayne and L. Iliadis, Eds., Communications in Computer and Information Science, vol. 629, Springer, Cham, 2016, pp. 213–226. doi: https://doi.org/10.1007/978-3-319-44188-7_16
W. A. Gardner, “The spectral correlation theory of cyclostationary time-series,” Signal Process., vol. 11, no. 1, pp. 13–36, Jul. 1986. doi: https://doi.org/10.1016/0165-1684(86)90092-7
W. Akmouche, “Detection of cyclostationary signals: A review,” Proc. IEEE Signal Process. Workshop Higher-Order Stat., pp. 349–352, 1999.
B. S. Goshu, “A fingerprint matrix methodology for the predictive diagnosis of incipient wall failure: A mathematical modeling and machine learning approach,” International Journal of Building Information Modeling Applications in Construction, vol. 1, no. 2, pp. 25–45, 2025.
R. S. Roberts, W. A. Brown, and H. H. Loomis, “Computationally efficient algorithms for cyclic spectral analysis,” IEEE Signal Process. Mag., vol. 8, no. 2, pp. 38–49, 1991. https://doi.org/10.1109/79.81008
H. L. Van Trees, Optimum Array Processing: Part IV of Detection, Estimation, and Modulation Theory. Hoboken, NJ, USA: John Wiley & Sons, 2002.
A. V. Dandawate and G. B. Giannakis, “Statistical tests for the presence of cyclostationarity,” IEEE Trans. Signal Process, vol. 42, no. 9, pp. 2355–2369, 1994. doi: https://doi.org/10.1109/78.301830
W. A. Gardner and C. M. Spooner, “The cumulant theory of cyclostationary time-series,” IEEE Trans. Signal Process., vol. 42, no. 12, pp. 3429–3441, 1994. https://doi.org/10.1109/78.362092
J. G. Proakis and M. Salehi, Digital Communications, 5th ed. New York, NY, USA: McGraw-Hill, 2008.
I. T. Jolliffe, Principal Component Analysis, 2nd ed. New York, NY, USA: Springer, 2002. https://doi.org/10.1007/b98835
S. Baek, H. Lee, and J. Kim, “Antijamming improvement for frequency hopping using noise-jammer power estimator,” Applied Sciences, vol. 10, no. 5, p. 1733, 2020. doi: https://doi.org/10.3390/app10051733
L. Meng, H.-C. Wu, and Y. Huang, “Cyclostationarity-based automatic modulation classification under AWGN and fading channels,” IEEE Trans. Veh. Technol., vol. 68, no. 10, pp. 9645–9656, 2019. doi: https://doi.org/10.1109/TVT.2019.2932487
A. Swami and B. M. Sadler, “Hierarchical digital modulation classification using cumulants,” IEEE Trans. Commun., vol. 44, no. 3, pp. 259–265, 1996. doi: https://doi.org/10.1109/26.837045
D. H. Hubel and T. N. Wiesel, “Receptive fields, binocular interaction, and functional architecture in the cat’s visual cortex,” J. Physiol., vol. 160, no. 1, pp. 106–154, 1962. doi: https://doi.org/10.1113/jphysiol.1962.sp006837
S. Peng, H. Jiang, H. Wang, H. Alwageed, and S. Sezer, “Modulation classification based on signal constellation diagrams and deep learning,” IEEE Trans. Neural Netw. Learn. Syst., vol. 30, no. 3, pp. 718–727, Mar. 2019. doi: https://doi.org/10.1109/TNNLS.2018.2850703
J. Liu, X. Wang, and Y. Li, “Cyclostationary feature-based jamming detection in cognitive radio networks,” IEEE Trans. Wireless Commun., vol. 23, no. 4, pp. 2156–2170, 2024. doi: https://doi.org/10.1109/TWC.2023.3321456
S. G. Mallat, “A theory for multiresolution signal decomposition: The wavelet representation,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 11, no. 7, pp. 674–693, 1989. doi: https://doi.org/10.1109/34.192463
Y. Zhang, L. Chen, and Q. Wu, “Adaptive wavelet denoising for signal recovery in jammed environments,” IEEE Signal Process. Lett., vol. 30, pp. 1123–1127, 2023. doi: https://doi.org/10.1109/LSP.2023.3298765
C. M. Spooner and W. A. Gardner, “The spectral correlation theory of cyclostationary time-series,” Signal Process., vol. 35, no. 1, pp. 1–25, 1994. doi: https://doi.org/10.1016/0165-1684(94)00065-4
