volume | PIER Journals

Abstract

A hybrid machine-learning-based optimization method is proposed for quick optimization of antenna shape design. The hybrid optimization method combines self-supervised learning and transfer learning. The application of self-supervised learning avoids the requirement to obtain labeled simulation data for electromagnetic samples, thereby reducing the difficulty of sample construction. The introduction of transfer learning further improves the sample utilization and optimizes efficiency in electromagnetic tasks. The proposed method enables rapid and high-degree-of-freedom optimization of antennas. To validate its effectiveness, a reflectarray antenna design incorporating distinct elements is employed as a case study. Simulation results indicate that the designed antenna exhibits a realized gain of 26.3 dBi and 46% aperture efficiency at the center frequency, and each element has a highly flexible independent structural design. During the optimization process, the proposed hybrid method demonstrates higher optimization efficiency than traditional methods, while significantly reducing sample construction time.

1. Rocca, P., R. J. Mailloux, and G. Toso, "GA-based optimization of irregular subarray layouts for wideband phased arrays design," IEEE Antennas and Wireless Propagation Letters, Vol. 14, 131-134, 2015.
doi:10.1109/lawp.2014.2356855

2. Xiong, Jie, Wen-Qin Wang, Huaizong Shao, and Hui Chen, "Frequency diverse array transmit beampattern optimization with genetic algorithm," IEEE Antennas and Wireless Propagation Letters, Vol. 16, 469-472, 2016.
doi:10.1109/lawp.2016.2584078

3. Robinson, J. and Y. Rahmat-Samii, "Particle swarm optimization in electromagnetics," IEEE Transactions on Antennas and Propagation, Vol. 52, No. 2, 397-407, 2004.
doi:10.1109/tap.2004.823969

4. Poli, Lorenzo, Paolo Rocca, Luca Manica, and Andrea Massa, "Handling sideband radiations in time-modulated arrays through particle swarm optimization," IEEE Transactions on Antennas and Propagation, Vol. 58, No. 4, 1408-1411, 2010.
doi:10.1109/tap.2010.2041165

5. Zhang, Si-Rui, Yi-Xuan Zhang, and Chao-Yi Cui, "Efficient multiobjective optimization of time-modulated array using a hybrid particle swarm algorithm with convex programming," IEEE Antennas and Wireless Propagation Letters, Vol. 19, No. 11, 1842-1846, 2020.
doi:10.1109/lawp.2020.3014366

6. Quevedo-Teruel, O. and E. Rajo-Iglesias, "Ant colony optimization in thinned array synthesis with minimum sidelobe level," IEEE Antennas and Wireless Propagation Letters, Vol. 5, 349-352, 2006.
doi:10.1109/lawp.2006.880693

7. Gregory, Micah D., Zikri Bayraktar, and Douglas H. Werner, "Fast optimization of electromagnetic design problems using the covariance matrix adaptation evolutionary strategy," IEEE Transactions on Antennas and Propagation, Vol. 59, No. 4, 1275-1285, 2011.
doi:10.1109/tap.2011.2109350

8. Wang, Jian, Xue-Song Yang, Xiao Ding, and Bing-Zhong Wang, "Antenna radiation characteristics optimization by a hybrid topological method," IEEE Transactions on Antennas and Propagation, Vol. 65, No. 6, 2843-2854, 2017.
doi:10.1109/tap.2017.2688918

9. Wang, Zhongbao, Shaojun Fang, Qiang Wang, and Hongmei Liu, "An ANN-based synthesis model for the single-feed circularly-polarized square microstrip antenna with truncated corners," IEEE Transactions on Antennas and Propagation, Vol. 60, No. 12, 5989-5992, 2012.
doi:10.1109/tap.2012.2214195

10. Xiao, Li-Ye, Wei Shao, Zhi-Xin Yao, and Shanshan Gao, "Data mining techniques in artificial neural network for UWB antenna design," Radioengineering, Vol. 27, No. 1, 70-78, 2018.
doi:10.13164/re.2018.0070

11. Xiao, Li-Ye, Wei Shao, Fu-Long Jin, and Bing-Zhong Wang, "Multiparameter modeling with ANN for antenna design," IEEE Transactions on Antennas and Propagation, Vol. 66, No. 7, 3718-3723, 2018.
doi:10.1109/tap.2018.2823775

12. Feng, Feng, Chao Zhang, Jianguo Ma, and Qi-Jun Zhang, "Parametric modeling of EM behavior of microwave components using combined neural networks and pole-residue-based transfer functions," IEEE Transactions on Microwave Theory and Techniques, Vol. 64, No. 1, 60-77, 2016.
doi:10.1109/tmtt.2015.2504099

13. Liu, Yan-Fang, Lin Peng, and Wei Shao, "An efficient knowledge-based artificial neural network for the design of circularly polarized 3-D-printed lens antenna," IEEE Transactions on Antennas and Propagation, Vol. 70, No. 7, 5007-5014, 2022.
doi:10.1109/tap.2022.3140313

14. You, Xi Chong and Feng Han Lin, "Inverse design of reflective metasurface antennas using deep learning from small-scale statistically random pico-cells," Microwave and Optical Technology Letters, Vol. 66, No. 2, e34068, 2024.
doi:10.1002/mop.34068

15. Zhu, Ruichao, Tianshuo Qiu, Jiafu Wang, Sai Sui, Chenglong Hao, Tonghao Liu, Yongfeng Li, Mingde Feng, Anxue Zhang, Cheng-Wei Qiu, and Shaobo Qu, "Phase-to-pattern inverse design paradigm for fast realization of functional metasurfaces via transfer learning," Nature Communications, Vol. 12, No. 1, 2974, 2021.
doi:10.1038/s41467-021-23087-y

16. Naseri, Parinaz and Sean V. Hum, "A generative machine learning-based approach for inverse design of multilayer metasurfaces," IEEE Transactions on Antennas and Propagation, Vol. 69, No. 9, 5725-5739, 2021.
doi:10.1109/tap.2021.3060142

17. Wei, Zhaohui, Zhao Zhou, Peng Wang, Jian Ren, Yingzeng Yin, Gert Frølund Pedersen, and Ming Shen, "Equivalent circuit theory-assisted deep learning for accelerated generative design of metasurfaces," IEEE Transactions on Antennas and Propagation, Vol. 70, No. 7, 5120-5129, 2022.
doi:10.1109/tap.2022.3152592

18. Lyu, Yanhe, Theng Huat Gan, and Zhi Ning Chen, "TE-TM balanced wide-angle metacells for low scan-loss metalens antenna using prior knowledge-guided generative deep learning-enabled method," IEEE Transactions on Antennas and Propagation, Vol. 73, No. 5, 2940-2949, 2025.
doi:10.1109/tap.2025.3552837

Dr. Hao Huang

Dr. Xue-Song Yang