Journal of Radio Electronics. eISSN 1684-1719. 2026. ¹2
Full text in Russian (pdf)
DOI: https://doi.org/10.30898/1684-1719.2026.2.8
IMPLEMENTATION OF THE CALCULATION
OF THE GAIN OF AN ACTIVE PHASED ARRAY ANTENNA
IN SPECIFIED DIRECTIONS USING NEURAL NETWORKS
A.A. Komarov, S.Y. Pashaev, A.M. Mikhailov, V.A. Permyakov
National Research University «Moscow Power Engineering Institute»,
111250, Russia, Moscow, Krasnokazarmennaya st., 14, building 1
The paper was received November 14, 2025.
Abstract. Calculating the gain of an antenna array is a critical task in the design and operation of modern radio, radar, and electronic warfare systems. This parameter makes it possible to quantify the efficiency of electromagnetic energy concentration in a given direction, which directly determines the communication range, noise immunity and energy efficiency of the system. In this paper, we propose a method for calculating the AFAR gain using neural networks. The approach is based on training a model on synthesized data, where the input parameters are the lattice geometry, amplitude-phase distribution and scanning angles, and the target variable is the gain value. The trained model demonstrates the ability to instantly predict antenna characteristics for arbitrary configurations. The advantage of the method is a significant reduction in computational costs compared to traditional electrodynamic calculations while maintaining satisfactory accuracy. The developed approach opens up opportunities for creating adaptive directional pattern control systems in real time and effectively optimizing antenna array parameters under changing operating conditions.
Key words: phased antenna array, neural networks, application of neural networks for numerical calculations, approximation, gain, directional pattern, machine learning.
Financing: The work was carried out with the financial support of the Russian Science Foundation, project No. 23-19-00485 https://rscf.ru/project/23-19-00485/
Corresponding author: Pashaev Soltanbek Yusupovich soltanbek.pashaev@mail.ru
References
1. Litvinov O. S. i dr. Issledovanie kharakteristik podavleniya shirokopolosnykh pomekh v adaptivnykh antennykh reshetkakh s nejrosetevym upravleniem //Antenny. – 2020. – ¹. 4. – S. 48-53.
2. Litvinov O. S. i dr. Issledovanie kharakteristik podavleniya shumov v adaptivnykh antennykh reshetkakh na neirosetevom upravlenii //Antenny. – 2019. – ¹. 8. – S. 12-19.
3. Litvinov O. S., Zabelin A. N. Sravnenie algoritmov podavleniya pomekh v adaptivnykh antennykh reshetkakh na neirosetevom, kalmanovskom i traditsionnom upravlenii //volnovaya ehlektronika i infokommunikatsionnye sistemy. – 2022. – S. 181-187
4. Zabelin A. N., Litvinov O. S. Osobennosti primeneniya rekurrentnoi neironnoi seti dlya podavleniya nestatsionarnykh pomekh v adaptivnoi antennoi reshetke //Volnovaya ehlektronika i infokommunikatsionnye sistemy. – 2023. – S. 158-167.
5. Litvinov O. S. i dr. Podavlenie shumovykh pomekh v adaptivnykh antennykh reshetkakh na osnove neirosetevykh algoritmov //Antenny. – 2018. – ¹. 2. – S. 40-44.
6. Litvinov O. S. i dr. Sravnenie ehffektivnosti podavleniya pomekh v adaptivnykh antennykh reshetkakh na neirosetevom upravlenii //Antenny i Rasprostranenie Radiovoln. – 2019. – S. 67-69.
7. Yang R. et al. An Ultrathin Filtering Transmitarray Antenna Based on Machine-Learning-Enabled Phase Synthesis //IEEE Antennas and Wireless Propagation Letters. – 2025
8. Cui C., Rocca P., Massa A. Shaped Beampattern Synthesis of Planar Arrays with Fully Convolutional Networks //IEEE Antennas and Wireless Propagation Letters. – 2025.
9. Wang C. et al. Design of Sparse Antenna Array Using Physics-Aware Generative Adversarial Network //IEEE Transactions on Antennas and Propagation. – 2025.
10. Lin Q. et al. A novel DMK fusion model for Full-Cycle Data and Its Application in Electronic Equipment //IEEE Sensors Journal. – 2025.
11. Schenone V., Fedeli A., Randazzo A. An Antenna Array Diagnosis Approach Based on CNN Inversion and CFAR Detection //IEEE Transactions on Antennas and Propagation. – 2025.
12. Yao H. M. et al. Antenna array diagnosis using a deep learning approach //IEEE Transactions on Antennas and Propagation. – 2024. – Ò. 72. – ¹. 6. – Ñ. 5396-5401.
13. Huang Z. et al. Spectral efficiency optimization for millimeter wave hybrid massive MIMO systems using unsupervised learning //Intelligent and Converged Networks. – 2025. – Ò. 6. – ¹. 3. – Ñ. 247-263.
14. Triwidyastuti Y. et al. Unsupervised Learning-Based Joint Beamforming and Phase-Shift Optimization for RIS-Assisted DeepMIMO With Large-Scale Arrays //IEEE Access. – 2025.
15. Aluvihare H. et al. A Low-complexity Structured Neural Network Approach to Intelligently Realize Wideband Multi-beam Beamformers //arXiv preprint arXiv:2503.20694. – 2025.
16. Ma J. et al. Efficient Multidimensional Parameter Estimation Using Machine Learning-Assisted SAGE Algorithm //IEEE Signal Processing Letters. – 2025.
17. Peng F., Chen X., Xue J. AI-Assisted antenna optimization: Integrating evolutionary and inverse cascade neural networks with differential evolution //IEEE Transactions on Antennas and Propagation. – 2025.
18. Adama V. R. et al. Introducing Machine Learning in Design and Analysis of Antenna for Wireless Systems //2025 International Conference on Next Generation Computing Systems (ICNGCS). – IEEE, 2025. – Ñ. 1-7.
19. Nielsen M. A. Neural Networks and Deep Learning. Determination Press, 2015 [Electronic resource]. — URL: http://neuralnetworksanddeeplearning.com/
20. MIT Introduction to Deep Learning 2025 [Electronic resource]. — URL: https://introtodeeplearning.com
For citation:
Komarov À.À., Pashaev S.Y., Permyakov V.À., Mikhailov À.Ì. Implementation of the calculation of the gain of an active phased array antenna in specified directions using neural networks. // Journal of Radio Electronics. – 2026. – ¹. 2. https://doi.org/10.30898/1684-1719.2026.2.8 (In Russian)