ASSESSMENT OF THE IMPACT OF SPARSITY AND GEMAN-MCCLURE REGULARIZATION ON SIGNAL RECONSTRUCTION ACCURACY
DOI:
https://doi.org/10.26906/SUNZ.2025.2.239Keywords:
reconstruction, complex signals, Geman-McClure function, sparsity, regularization, nonlinearity, system, noise immunity, signal-to-noise ratio (SNR), Volterra series, optimization, signal parametersAbstract
The article proposes a method for spectral reconstruction of signals using the Geman-McClure Function and Lagrange multipliers to optimize parameters. The method combines approaches of adaptive filtering and sparsity-based parameter regularization, incorporating Geman-McClure Function and Lagrange multipliers to ensure orthogonality and optimal parameter selection. The proposed approach efficiently identifies significant signal parameters, minimizes mutual interference among nonlinear components, reduces computational complexity, and improves reconstruction accuracy. Experimental modeling demonstrated that the developed method achieves a significant reduction in mean square error (MSE) by 10–15% and enhances the robustness of signal reconstruction by 10–15% across a signal-to-noise ratio (SNR) range from -10 to 10 dB, confirming its effectiveness for practical applications in cognitive telecommunication networks operating under severe noise conditions and nonlinear distortions.Downloads
References
1. Araujo-Simon, Jake. Compositional Nonlinear Audio Signal Processing with Volterra Series. arXiv preprint, 2023. https://arxiv.org/abs/2308.07229
2. Birpoutsoukis, G., Csurcsia, P. Zoltán, S., Johan. Efficient Multidimensional Regularization for Volterra Series Estimation. IEEE Transactions on Instrumentation and Measurement, vol. 68, no. 10, 2019, РР. 3770-3782. DOI:10.1109/TIM.2019.2907240.
3. Birpoutsoukis, G., Marconato, A., Lataire, J., & Schoukens, J. Regularized nonparametric Volterra kernel estimation. Automatica, 82, Р. 324-327 https://doi.org/10.1016/j.automatica.2017.04.014.
4. Boyd S., Chua L. O., Desoer C. A. Analytical Foundations of Volterra Series//IMA Journal of Mathematical Control and Information, Oxford University Press, 1(3):243-282, (1984). analytical_volterra.pdf.
5. Franz, M. O., Schölkopf, B. A Unifying View of Wiener and Volterra Theory and Polynomial Kernel Regression. (2007) Neural Computation 18(12): Р.3097-118 DOI:10.1162/neco.2006.18.12.3097.
6. Geman, D., & McClure, D. E. (1987). Statistical Methods for Tomographic Image Reconstruction. Bulletin of the International Statistical Institute, 52(4), 5-21.
7. Gutleb, T. S. A Fast Sparse Spectral Method for Nonlinear Integro-Differential Volterra Equations with General Kernels. (2020) https://arxiv.org/abs/2005.06081
8. Hotz, Matthias, Vogel, Christian. Linearization of Time-Varying Nonlinear Systems Using A Modified Linear Iterative Method. arXiv preprint, 2014, https://arxiv.org/abs/1404.5901
9. Issa H. Al-Aidi and Ahmed Sh. Al-Atabi The Analytical Methods Of Volterra Integral Equations of The Second Kind. - WJCMS, Vol. 2, no. 3, РР. 39–45, (2023), DOI:10.31185/wjcm.119.
10. Kapgate, Sachin N., Gupta, Saurav, Sahoo, Ajit Kumar. Adaptive Volterra Modeling for Nonlinear Systems Based on LMS Variants. 2018 International Conference on Signal Processing and Integrated Networks (SPIN), 2018, РР. 414-419. DOI: 10.1109/SPIN.2018.8474036.
11. Lu, L., Zhang, J., & Wang, X. (2018). Recursive Geman-McClure Method for Implementing Second-Order Volterra Filter. Systems and Control. DOI: 10.48550/arXiv.1808.00613
12. Nmah, Benedict. Solving a Relaxed Min-Cost Redundancy Allocation Model Using Lagrange Multiplier and Newton’s Method. Journal of Advances in Mathematics and Computer Science. –2024. №39 (11):152-59. https://doi.org/10.9734/jamcs/2024/v39i111945.
13. Perets K., Lysechko V., Komar O. (2024) Modeling Nonlinear Signal Components Based on Volterra Series in the Frequency Domain during Spectral Reconstruction. Computer-integrated technologies: education, science, production. Telecommunications and radio engineering. – Луцьк. – 2024. № 57, С.192-201. DOI: https://doi.org/10.36910/6775-2524-0560-2024-57-23.
14. Pirogova, N.D., Neches, I.O. (2018). Compensation of Nonlinear Distortions in Telecommunication Systems with the Use of Functional Series of Volterra// Proc. of the Second International Scientific Conference «Intelligent Information Technologies for Industry». Advances in Intelligent Systems and Computing, Vol 680. https://doi.org/10.1007/978-3-319-68324-9_49/
15. Yazdanpanah H., Carini А., Lima М. V. S. L0-Norm Adaptive Volterra Filters 2019 27th European Signal Processing Conference (EUSIPCO)Р.1-5,DOI: 10.23919/EUSIPCO.2019.8903013.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Kostiantyn Perets , Oleksii Komar
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
