RADIO PHYSICS AND RADIO ASTRONOMY

Subject and Purpose. Designing antennas with desired operational features is a complex optimization problem addressing a large number of parameters and nonlinear relationships. Artificial neural networks (ANNs) have had strong potential in antenna engineering. The ability to approximate nonlinear functions and capture hidden relationships enables partial automation of antenna designing. However, it requires a detailed analysis of how ANN parameters affect predictive accuracy and computational efficiency. The present study investigates the influence of ANN architecture and training configurations on the predictive accuracy of resonant frequencies for rectangular microstrip patch antennas.

Methods and Methodology. A modular Python-based experimental platform is used to evaluate ANN performance for dif- ferent ANN configurations. The ANN training is on a synthetic dataset generated from a classical analytical antenna model. The number of hidden layers, neurons per layer, activation functions, and optimizers are systematically varied to assess their particular impacts on convergence, generalization performance, and execution time.

Results. It has been shown that a three-layer neural network [1024, 512, 256 neurons] with ReLU activation and the Adam optimizer strikes the best balance between predictive accuracy and training rate. Simpler or excessively deep architectures, non-adaptive optimizers, and saturating activation functions can slow convergence or cause unstable training. Further analysis indicated that a smaller batch size introduces useful stochasticity into training but might also destabilize the process.

Conclusions. The study has demonstrated that joint optimization of ANN architecture and training dynamics is essential for developing accurate and computationally efficient electromagnetic models. The practical results are recommendations for machine-learning-based antenna design.

Keywords:  artificial neural network (ANN), activation function, adaptive optimizer, microstrip patch antenna, automatic antenna design

Manuscript submitted 17.10.2025

Radio phys. radio astron. 2026, 31(2): 098-107

REFERENCES

1. James, J.R., Hall, P.S., Wood, C., 1981. Microstrip Antenna. Theory and Design, London: Peter Peregrinus LTD.

2. Kumar, G., Kamala, P.R., 2003. Broadband Microstrip Antennas. Boston, London: Artech House.
3. Waterhouse, R.B., 2013, Microstrip Patch Antennas: A Designer’s Guide. New York: Springer Science & Business Media.
4. Patnaik, A., Anagnostou, D.E., Mishra, R.K., Christodoulou, C.G., Lyke, J.C., 2004. Applications of neural networks in wire- less communications, IEEE Antennas and Propagation Magazine, 46(3), pp. 130—137. DOI: 10.1109/MAP.2004.1374125
5. Feng, F., Zhang, J., Na, W., Zhang, W., Jin, J., Zhang, Q.-J., 2022. Artificial neural networks for microwave computer-aided design: The state of the art. IEEE Trans. Microw. Theory Tech., 70(11), pp. 4597—4619. DOI: 10.1109/TMTT.2022.3197751
6. Yu, Y., Zhang, Z., Cheng, Q.S., Liu, B., Wang, Y., Guo, C., Ye, T.T., 2022. State-of-the-art: AI-assisted surrogate mod- eling and optimization for microwave filters. IEEE Trans. Microw. Theory Tech., 70(11), pp. 4635—4651. DOI:10.1109/ TMTT.2022.3208898
7. Qi, S., Sarris, C.D., 2022. Deep neural networks for rapid simulation of planar microwave circuits based on their layouts.
IEEE Trans. Microw. Theory Tech., 70(11), pp. 4805—4815. DOI: 10.1109/TMTT.2022.3210229
8. Swaminathan, M., Bhatti, O.W., Guo, Y., Huang, E., Akinwande, O., 2022. Bayesian Learning for Uncertainty Quanti-
fication, Optimization, and Inverse Design. IEEE Trans. Microw. Theory Tech., 70(11), pp. 4620—4634. DOI: 10.1109/TMTT.2022.3206455
9. Ma, J., Dang, S., Li, P., Watkins, G., Morris, K.A., Beach, M.A., 2023. A learning-based methodology for microwave passive component design. IEEE Trans. Microw. Theory Tech., 71(7), pp. 3037—3050. DOI: 10.1109/TMTT.2023.3238418
10. Thakare, V.V., Singhal, P., 2010. Microstrip antenna design using artificial neural networks. Int. J. RF Microw. Comput.-Aid-
ed Eng., 20(1), pp. 76—86. DOI: 10.1002/mmce.20414
11. Rawat, A., Yadav, R.N., Shrivastava, S.C., 2012. Neural network applications in smart antenna arrays: A review. AEU — Int. J. Electron. Commun., 66(11), pp. 903—912. DOI: 10.1016/j.aeue.2012.03.012
12. Khan, T., De, A., 2015. Modeling of microstrip antennas using neural networks techniques: a review. Int. J. RF Microw. Comput.-Aided Eng., 25(9), pp. 747—757. DOI: 10.1002/mmce.20910
13. Shi, D., Lian, W., Cui, K., Leong, D.S.H., 2022. An intelligent antenna synthesis method based on machine learning. IEEE Trans. Antennas Propag., 70(7), pp. 4965—4976. DOI: 10.1109/TAP.2022.3182693
14. Massa, A., Salucci, M., 2022. On the design of complex EM devices and systems through the system-by-design para-
digm: A framework for dealing with the computational complexity. IEEE Trans. Antennas Propag., 70(2), pp. 1328—1343. DOI: 10.1109/TAP.2021.3111417
15. Raya, M.B., Pal, S., Ali, K., 2019. Design of inset fed rectangular shaped microstrip patch antenna using deep neural net-
works. In: Proc. 2019 22nd Int. Conf. on Computer and Information Technology (ICCIT). Dhaka, Bangladesh, 18—20 Dec. 2019. Dhaka: IEEE, 2019. DOI: 10.1109/ICCIT48885.2019.9038284
16. Pal, S., Raya, M.B., Ali, K., 2019. Computation of resonant frequency and gain from inset fed rectangular shaped microstrip
patch antenna using deep neural network. In: Proc. 2019 4th Int. Con. on Electrical Information and Communication Tech-
nology (EICT). Khulna, Bangladesh, 20—22 Dec. 2019. Khulna. DOI: 10.1109/EICT48899.2019.9068758
17. Syahrial, S., Al Faqi, M.K., Meutia, E.D., Munadi, R., Roslidar R., 2024. Parameter estimation of two-element rectangular microstrip patch antenna using artificial neural network. In: Proc. 2024 FORTEI-Int. Conf. on Electrical Engineering (FORTEI-ICEE). Badung, Indonesia, 24—25 Oct. 2024. Badung: IEEE, pp. 216—221. DOI: 10.1109/FORTEI-ICEE64706.2024.10824537
18. Shereen, M.K., Liu, X., Wu, X., Niazi, S., Naseem, A., Khattak, M.I., 2025. Towards bi-directional deep learning approach in patch antenna design and optimization. In: Proc. 2025 IEEE Int. Workshop on Antenna Technology (iWAT). Cocoa Beach, FL, USA. 19—21 Feb. 2025. Cocoa Beach: IEEE. DOI: 10.1109/iWAT64079.2025.10931210
19. Derneryd, A., Lind, A., 1979. Extended analysis of rectangular microstrip resonator antennas. IEEE Trans. Antennas
Propag., 27(6), pp. 846—849. DOI: 10.1109/TAP.1979.1142206
20. Garg, R., Bhartia, P., Bahl, I.J., Ittipiboon, A., 2001. Microstrip Antenna Design Handbook. Boston, London: Artech House.
21. Steer, M., 2019. Fundamentals of Microwave and RF Design, 3rd ed. USA: NC State University.
22. Fesenko, V.I., Tkachenko, G.V., 2007. Modeling of 1-D photonic bandgap microstrip structures. In: Proc. 2007 Int. Work-
shop on Optoelectronic Physics and Technology (OPT’07), Kharkiv, Ukraine, 20—22 June 2007, pp. 42—43. DOI: 10.1109/OPT.2007.4298524.
23. Mishra, R.K., Patnaik, A., 2003. Designing rectangular patch antenna using the neurospectral method. IEEE Trans. Anten-
nas Propag., 51(8), pp. 1914—1921. DOI: 10.1109/TAP.2003.814748
24. Haykin, S., 2009. Neural Networks and Learning Machines. 3rd ed. NJ: Pearson Education.
25. Goodfellow, I., Bengio, Y., Courville, A., 2016. Deep Learning. Cambridge, Massachusetts London: MIT Press.
26. He, K., Zhang, X., Ren, S., Sun, J., 2015. Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification. In: Proc. 2015 IEEE Int. Conf. on Computer Vision (ICCV). Santiago, Chile, 7—13 Dec. 2015, pp. 1026—1034. DOI: 10.1109/ICCV.2015.123
27. Maity, B., Nayak, S.K., 2024. Artificial neural networks for optimal design parameters prediction of microstrip patch antenna with wideband harmonic suppression. In: Proc. 2024 Second Int. Conf. on Microwave, Antenna and Communication
(MAC). Dehradun, India, 04—06 Oct. 2024. DOI: 10.1109/MAC61551.2024.10837513