Convolutional Neural Networks Aided Reinforcement Learning for Accelerated Optimization of Antenna Topology

Authors

  • Jiangling Dou Yunnan Key Laboratory of Computer Technologies Application Kunming University of Science and Technology, Kunming 650500, China, School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China
  • Hao Gong School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China
  • Shuaibing Wei School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China
  • Haokang Chen School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China
  • Yujie Chen School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China
  • Tao Shen College of Mechanical and Electrical Engineering Yunnan Electromechanical Vocational and Technical College, Kunming 650500, China
  • Jian Song Yunnan Key Laboratory of Computer Technologies Application Kunming University of Science and Technology, Kunming 650500, China, School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China

DOI:

https://doi.org/10.13052/2025.ACES.J.400105

Keywords:

Convolutional neural network (CNN), machine learning (ML), microstrip antenna, reinforcement learning (RL), surrogate model (SM), topology optimization

Abstract

A machine learning (ML) framework is proposed to achieve the automatic and rapid optimization of antenna topologies. A convolutional neural network (CNN) is utilized as a surrogate model (SM) and is combined with reinforcement learning (RL) algorithms. Specifically, the RL agent interacts with simulation software to learn. Data accumulated from electromagnetic (EM) simulations are used to train the SM. The CNN-based SM predicts antenna performance based on the topology of the antenna. Subsequently, the SM replaces EM simulations within the RL training environment. The RL agent interacts with the CNN-based SM to search for the optimal topology. This approach significantly reduces dependence on time-consuming EM simulations. To validate the effectiveness of the optimization method, a center-fed microstrip patch antenna is optimized. Simulation results demonstrate that, compared to other optimization methods, impedance bandwidth is improved, while the number of simulation samples and optimization time are significantly reduced.

Downloads

Download data is not yet available.

Author Biographies

Jiangling Dou, Yunnan Key Laboratory of Computer Technologies Application Kunming University of Science and Technology, Kunming 650500, China, School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China

Jiangling Dou was born in 1985 in Jiangsu Province, China. She received her Ph.D. degree in Electromagnetic Fields and Microwave Technology from Southeast University in 2018. Her research interests include electromagnetic field theory and applications.

Hao Gong, School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China

Hao Gong is currently pursuing a graduate degree at the School of Information Engineering and Automation, Kunming University of Science and Technology. His research interests include machine learning-assisted antenna optimization.

Shuaibing Wei, School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China

Shuaibing Wei is currently pursuing a graduate degree at the School of Information Engineering and Automation, Kunming University of Science and Technology. His research interests include machine learning-assisted antenna optimization.

Haokang Chen, School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China

Haokang Chen is currently pursuing a graduate degree at the School of Information Engineering and Automation, Kunming University of Science and Technology. His primary research focuses on millimeter-wave devices and systems.

Yujie Chen, School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China

Yujie Chen is currently pursuing a graduate degree at the School of Information Engineering and Automation, Kunming University of Science and Technology.

Tao Shen, College of Mechanical and Electrical Engineering Yunnan Electromechanical Vocational and Technical College, Kunming 650500, China

Tao Shen a member of IEEE. He earned his Ph.D. from the Illinois Institute of Technology in Chicago, Illinois, USA, in 2013. Presently, he holds the position of President at Yunnan Vocational and Technical College of Mechanical and Electrical Engineering. Dr. Shen has contributed to over 20 publications in prestigious SCIE-indexed journals and leading international conferences within his research domains. His areas of expertise include image processing, artificial intelligence, and the Internet of Energy.

Jian Song, Yunnan Key Laboratory of Computer Technologies Application Kunming University of Science and Technology, Kunming 650500, China, School of Information Engineering and Automation Kunming University of Science and Technology, Kunming 650500, China

Jian Song a member of IEEE. He obtained his Bachelor of Engineering degree in Electronics Information Engineering from Jiangxi University of Science and Technology in Ganzhou, China. He later earned his Ph.D. in Electromagnetic Fields and Microwave Technology from the University of Electronic Science and Technology of China in Chengdu in 2015. In 2019, he became a faculty member at Kunming University of Science and Technology. His research focuses on microwave engineering and the processing of biomedical images.

References

J. P. Jacobs, “Efficient resonant frequency modeling for dual-band microstrip antennas by gaussian process regression,” IEEE Antennas Wireless Propag. Lett., vol. 14, pp. 337-341, Oct. 2015.

M. Tarkowski and L. Kulas, “RSS-based DoA estimation for ESPAR antennas using support vector machine,” IEEE Antennas Wireless Propag. Lett., vol. 18, no. 4, pp. 561-565, Apr. 2019.

J. P. Jacobs, “Accurate modeling by convolutional neural-network regression of resonant frequencies of dual-band pixelated microstrip antenna,” IEEE Antennas Wireless Propag. Lett., vol. 20, no. 12, pp. 2417-2421, Dec. 2021.

C. Ferchichi, D. Omri, and T. Aguili, “Utilizing 1D convolutional neural networks for enhanced design and optimization of rectangular patch antenna parameters,” IEEE Int. Symp. Networks, Comput. Commun., pp. 1-6, Oct. 2023.

F. Peng and X. Chen, “A low-cost optimization method for 2-D antennas using a disassemblable convolutional neural network,” IEEE Trans. Antennas Propag., vol. 72, no. 9, pp. 7057-7067, Sep. 2024.

L.-Y. Xiao, W. Shao, F.-L. Jin, and B.-Z. Wang, “Multiparameter modeling with ANN for antenna design,” IEEE Trans. Antennas Propag., vol. 66, no. 7, pp. 3718-3723, July 2018.

J. M. Johnson and Y. Rahmat-Samii, “Genetic algorithms and method of moments (GA/MOM) for the design of integrated antennas,” IEEE Trans. Antennas Propag., vol. 47, no. 10, pp. 1606-1614, Oct. 1999.

N. Jin and Y. Rahmat-Samii, “Parallel particle swarm optimization and finite- difference time-domain (PSO/FDTD) algorithm for multiband and wide-band patch antenna designs,” IEEE Trans. Antennas Propag., vol. 53, no. 11, pp. 3459-3468, Nov. 2005.

J. Liu, Z.-X. Chen, W.-H. Dong, X. Wang, J. Shi, H.-L. Teng, X.-W. Dai, S. S.-T. Yau, C.-H. Liang, P.-F. Feng, “Microwave integrated circuits design with relational induction neural network,” arXiv:1901.02069, 2019.

B. Zhang, C. Jin, K. Cao, Q. Lv, and R. Mittra, “Cognitive conformal antenna array exploiting deep reinforcement learning method,” IEEE Trans. Antennas Propag., vol. 70, no. 7, pp. 5094-5104, July 2022.

S. Fujimoto, H. van Hoof, and D. Meger, “Addressing function approximation error in actor-critic methods,” in Proc. of the 35th Int. Conf. on Machine Learning, vol. 80, pp. 2673-2682, July 2018.

S. Singh, T. Jaakkola, M. L. Littman, and C. Szepesvári, “Convergence results for single-step on-policy reinforcement-learning algorithms,” Mach. Learn., vol. 38, no. 3, pp. 287-308, Mar. 2000.

F. J. Villegas, T. Cwik, Y. Rahmat-Samii, and M. Manteghi, “A parallel electromagnetic genetic-algorithm optimization (EGO) application for patch antenna design,” IEEE Trans. Antennas Propag., vol. 52, no. 9, pp. 2424-2435, Sep. 2004.

A. Aldhafeeri and Y. Rahmat-Samii, “Brain storm optimization for electromagnetic applications: Continuous and discrete,” IEEE Trans. Antennas Propag., vol. 67, no. 4, pp. 2710-2722, Apr. 2019.

Q. Wu, W. Chen, C. Yu, H. Wang, and W. Hong, “Machine-learning-assisted optimization for antenna geometry design,” IEEE Trans. Antennas Propag., vol. 72, no. 3, pp. 2083-2095, Mar.2024.

Downloads

Published

2025-01-30

How to Cite

[1]
J. . Dou, “Convolutional Neural Networks Aided Reinforcement Learning for Accelerated Optimization of Antenna Topology”, ACES Journal, vol. 40, no. 01, pp. 35–41, Jan. 2025.

Issue

2025: Vol 40 Iss 1

Section

General Submission

Categories