Feature selection for determining input parameters in antenna modeling

doi:10.23919/JSEE.2023.000135

Journal of Systems Engineering and Electronics ›› 2025, Vol. 36 ›› Issue (1): 15-23.doi: 10.23919/JSEE.2023.000135

• ELECTRONICS TECHNOLOGY • Previous Articles

Feature selection for determining input parameters in antenna modeling

Zhixian LIU¹(), Wei SHAO¹^,*(), Xi CHENG²(), Haiyan OU¹(), Xiao DING¹()

¹ School of Physics, University of Electronic Science and Technology of China, Chengdu 611731, China
² School of Computer and Information Engineering, Xinjiang Agricultural University, Urumqi 830052, China

Received:2023-03-27 Accepted:2023-10-09 Online:2025-02-18 Published:2025-03-18
Contact: Wei SHAO E-mail:zxliu@std.uestc.edu.cn;weishao@uestc.edu.cn;chengxi@xjau.edu.cn;ouhaiyan@uestc.edu.cn;xding@uestc.edu.cn
About author:
LIU Zhixian was born in 1997. She received her B.S. degree in electronic information science and technology from China West Normal University, Nanchong, China, in 2019. She is pursuing her Ph.D. degree in radio physics from University of Electronic Science and Technology of China, Chengdu, China. Her current research interests include electromagnetic wave scattering and artificial intelligence. E-mail: zxliu@std.uestc.edu.cn

SHAO Wei was born in 1975. He received his B.E. degree in electrical engineering from University of Electronic Science and Technology of China (UESTC) in 1998, and M.S. and Ph.D. degrees in radio physics from UESTC in 2004 and 2006, respectively. He joined UESTC in 2007 and is now a professor there. From 2010 to 2011, he was a visiting scholar in the Electromagnetic Communication Laboratory, Pennsylvania State University, State College. From 2012 to 2013, he was a visiting scholar in the Department of Electrical and Electronic Engineering, University of Hong Kong. His research interests include computational electromagnetics and antenna design. E-mail: weishao@uestc.edu.cn

CHENG Xi was born in 1986. She received her B.E. degree in electrical engineering and M.S. degrees from Xidian University, Xi’an, China, in 2009, and 2012, respectively. She received her Ph.D. degree in physics from Université Paris-Saclay, Paris, France, in 2016. In 2016, she joined Xinjiang Agricultural University, where she is currently a lecturer. From 2017 to 2018, she was a postdoctor in University of Electronic Science and Technology of China, Chengdu, China. From 2018 to 2021, she was a postdoctor in Télécom Paris, Paris, France. Her current research interests include computational electromagnetics and artificial neural networks. E-mail: chengxi@xjau.edu.cn

OU Haiyan was born in 1982. She received her B.E. degree in electrical engineering from University of Electronic Science and Technology of China (UESTC) in 2000, and Ph.D. degree in optical engineering from Zhejiang University in 2009. She joined UESTC in 2009 and is now an associate professor. From 2010 to 2011, she was a visiting scholar in the Department of Engineering, Cambridge University, UK. From 2012 to 2013, she was a postdocter in the Department of Electrical and Electronic Engineering, University of Hong Kong. Her research interests include computational electromagnetics, microwave photonics and digital holography. E-mail: ouhaiyan@uestc.edu.cn

DING Xiao was born in 1982. He received his Ph.D. degree in radio physics from University of Electronic Science and Technology of China (UESTC), Chengdu, China, in 2013. In 2013, he was a research assistant with the Department of Electrical and Computer Engineering, South Dakota School of Mines and Technology, SD, USA. From June 2016 to June 2017, he was a visiting scholar with the Applied Electromagnetics Laboratory, University of Houston, TX, USA. He is a full professor at the Institute of Applied Physics, UESTC, Chengdu, China. He is a senior member of the Chinese Institute of Electronics. He has served on the review boards of various technical journals and many international conferences as the TPC member, a session organizer, and the session chair. His research interests include phased array antenna, frequency selective surface (FSS) radomes, and neural networks for electromagnetic device design. E-mail: xding@uestc.edu.cn
Supported by:
This work was supported by the National Natural Science Foundation of China (62161048), and Sichuan Science and Technology Program (2022NSFSC0547; 2022ZYD0109).

Abstract

Abstract:

In this paper, a feature selection method for determining input parameters in antenna modeling is proposed. In antenna modeling, the input feature of artificial neural network (ANN) is geometric parameters. The selection criteria contain correlation and sensitivity between the geometric parameter and the electromagnetic (EM) response. Maximal information coefficient (MIC), an exploratory data mining tool, is introduced to evaluate both linear and nonlinear correlations. The EM response range is utilized to evaluate the sensitivity. The wide response range corresponding to varying values of a parameter implies the parameter is highly sensitive and the narrow response range suggests the parameter is insensitive. Only the parameter which is highly correlative and sensitive is selected as the input of ANN, and the sampling space of the model is highly reduced. The modeling of a wideband and circularly polarized antenna is studied as an example to verify the effectiveness of the proposed method. The number of input parameters decreases from 8 to 4. The testing errors of |S₁₁| and axis ratio are reduced by 8.74% and 8.95%, respectively, compared with the ANN with no feature selection.

Key words: antenna modeling, artificial neural network (ANN), feature selection, maximal information coefficient (MIC)

Zhixian LIU, Wei SHAO, Xi CHENG, Haiyan OU, Xiao DING. Feature selection for determining input parameters in antenna modeling[J]. Journal of Systems Engineering and Electronics, 2025, 36(1): 15-23.

Figures/Tables 13

Fig 1

Fig 2

Fig 3

Table 1

Table 2

Table 3

Table 4

Fig 4

Table 5

Fig 5

Fig 6

Fig 7

Fig 8

References 32

1	CUI J Y, FENG F, NA W C, et al. Bayesian-based automated model generation method for neural network modeling of microwave components. IEEE Microwave Wireless Components Letter, 2021, 31(11): 1179−1182.
2	ZHANG J, CHEN J, GUO Q, et al. Parameterized modeling incorporating MOR-based rational transfer functions with neural networks for microwave components. IEEE Microwave Wireless Components Letters, 2022, 32(5): 379−382.
3	LIU Y S, WR-REHMAN M. An efficient method for antenna design based on a self-adaptive Bayesian neural network-assisted global optimization technique. IEEE Trans. on Antennas Propagation, 2022, 70(12): 11375−11388.
4	KOUHALVANDI L, MATEKOVITS L. Hyperparameter optimization of long short-term memory-based forecasting DNN for antenna modeling through stochastic methods. IEEE Antennas Wireless Propagation Letters, 2022, 21(4): 725−729.
5	HONG Y, SHAO W, LYU Y H, et al. Knowledge-based neural network for thinned array modeling with active element patterns. IEEE Trans. on Antennas Propagation, 2022, 70(11): 11229−11234.
6	IYE T, WYK P V, METSUMOTO T, et al. Neural network-based phase estimation for antenna array using radiation power pattern. IEEE Antennas Wireless Propagation Letters, 2022, 21(7): 1348−1352.
7	JIN J, SU Q, XU Y, et al. Efficient radiation pattern prediction of array antennas based on complex-valued graph neural networks. IEEE Antennas Wireless Propagation Letters, 2022, 21(12): 2467−2471.
8	GONG Y, XIAO S Q, WANG B Z. An ANN-based synthesis method for nonuniform linear arrays including mutual coupling effects. IEEE Access, 2020, 8: 144015−144026.
9	LUO H Y, SHAO W, DING X, et al. Shape modeling of microstrip filters based on convolutional neural network. IEEE Microwave Wireless Components Letters, 2022, 32(9): 1019−1022.
10	YANG Y, ZHANG Z, CHENG Q S S, et al. State-of-the-art: AI-assisted surrogate modeling and optimization for microwave filters. IEEE Trans. on Microwave Theory Techniques, 2022, 70(11): 4635−4651.
11	ZHAO P, WU K. Homotopy optimization of microwave and millimeter-wave filters based on neural network model. IEEE Trans. on Microwave Theory Techniques, 2020, 68(4): 1390−1400.
12	CONG R C, LIU N, LI X, et al. Design of wideband frequency selective surface based on the combination of the equivalent circuit model and deep learning. IEEE Antennas Wireless Propagation Letters, 2023, 22(9): 2110−2114.
13	YUAN L, WANG L, YANG X S, et al. An efficient artificial neural network model for inverse design of metasurfaces. IEEE Antennas and Wireless Propagation Letters, 2021, 20(6): 1013−1017.
14	ZHANG T, KEE C Y, ANG Y S, et al. Symmetry enhanced network architecture search for complex metasurface design. IEEE Access, 2022, 10: 73533−73547.
15	YE Z, SHAO W, DING X, et al. Knowledge-based neural network for multiphysical field modeling. IEEE Trans. on Microwave Theory Techniques, 2023, 71(5): 1967−1976.
16	ZHANG W, FENG F, YAN S, et al. EM-centric multiphysics optimization of microwave components using parallel computational approach. IEEE Trans. on Microwave Theory Techniques, 2020, 68(2): 479−489.
17	ZHAO Y H, HE D P, GUAN K, et al. The equivalence and realization of neural network and finite differences time domain. Proc. of the 15th European Conference on Antennas and Propagation, 2021. DOI: 10.23919/EuCAP51087.2021.9411021.
18	GUO L S, LI M K, XU S H, et al. Electromagnetic modeling using an FDTD-equivalent recurrent convolution neural network: accurate computing on a deep learning framework. IEEE Antennas Propagation Magazine, 2023, 65(1): 93−102.
19	QI S T, SARRIS C D. Numerical dispersion compensation for FDTD via deep learning. Proc. of the IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, 2022: 671−672.
20	WU F B, FAN M N, LIU W Y, et al. An efficient time-domain electromagnetic algorithm based on LSTM neural network. IEEE Antennas Wireless Propagation Letters, 2021, 20(7): 1322−1326.
21	WEI G F, ZHAO J, FENG Y L, et al. A novel hybrid feature selection method based on dynamic feature importance. Applied Soft Computing, 2020, 93: 106337.
22	XIAO L Y, SHAO W, LIANG T L, et al. Artificial neural network with data mining techniques for antenna design. Proc. of the IEEE International Symposium on Antennas and Propagation & USNC/URSI National Radio Science Meeting, 2017: 159−160.
23	WANG Y, CAO H B, XIONG M M, et al. Efficient test for nonlinear dependence of two continuous variables. BMC Bioinformatics, 2015, 16: 260.
24	RESHEF D. Detecting novel associations in large data sets. Science, 2011, 334(6062): 1518−1524.
25	SONG L, LANGFELDER P, HORVATH S. Comparison of co-expression measures: mutual information, correlation, and model based indices. BMC Bioinformatics, 2012, 13: 328.
26	GE R Q, ZHOU M L. McTwo: a two-step feature selection algorithm based on maximal information coefficient. BMC Bioinformatics, 2016, 17(1): 142.
27	WEI G F, ZHAO J. A novel hybrid feature selection method based on dynamic feature importance. Applied Soft Computing, 2020, 93: 106337.
28	XUAT S, PARK L. Low-profile broadband circularly polarized patch antenna using metasurface. IEEE Trans. on Antennas Propagation, 2015, 63(12): 5929−5934.
29	COHEN J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale: Lawrence Erlbaum Associates, 1988.
30	SCHMIDT S R, LAUNSBY R G. Understanding industrial designed experiments. Colorado Springs: Air Force Academy, 1992.
31	FENG F, ZHANG C, MA J G, et al. Parametric modeling of EM behavior of microwave components using combined neural networks and pole-residue-based transfer functions. IEEE Trans. on Microwave Theory Technique, 2016, 64(1): 60−77.
32	XIAO L Y, SHAO W, JIN F L, et al. Multiparameter modeling with ANN for antenna design. IEEE Trans. on Antennas Propagation, 2018, 66(7): 3718−3723.

Geometric variables	Minimum	Maximum	Step	Original value
g	0.4	0.6	0.036	0.5
l₁	6.8	10.2	0.567	8.5
l₂	5.6	8.4	0.467	7
l₃	4	6	0.3	5
p	6.4	9.6	0.533	8
w₁	2.4	3.6	0.2	3
w₂	2.4	3.6	0.2	3
w₃	10.4	15.6	0.867	13

Geometric variable	M_ave		K_ave
Geometric variable	S_M_ave	AR_M_ave	S_K_ave	AR_K_ave
g	0.93	0.93	0.04	0.12
l₁	0.98	0.73	0.93	0.06
l₂	0.77	0.84	0.07	0.39
l₃	0.98	0.98	0.67	0.26
p	0.60	0.71	0.25	0.51
w₁	0.54	0.50	0.02	0.05
w₂	0.98	0.34	0.36	0.02
w₃	0.69	0.58	0.42	0.50

Geometric variable	Training data			Testing data
Geometric variable	Minimum	Maximum	Step	Minimum	Maximum	Step
g	0.4	0.6	0.025	0.46	0.56	0.025
l₁	6.8	10.2	0.425	7.5	9.5	0.5
l₂	5.6	8.4	0.35	6	7.5	0.375
l₃	4	6	0.25	4.5	5.3	0.2
p	6.4	9.6	0.4	7	8.5	0.375
w₁	2.4	3.6	0.15	2.55	3.45	0.225
w₂	2.4	3.6	0.15	2.6	3.2	0.15
w₃	10.4	15.6	0.65	11	14.6	0.9

Model	Training error		Testing error
Model	\|S₁₁\|	AR	\|S₁₁\|	AR
Model 1 (${\boldsymbol{x}}'_{ {\text{all} } }$)	7.80	7.35	10.72	10.69
Model 2 (${\boldsymbol{x}}'_{ {\text{MIC} } }$)	4.67	3.95	5.24	4.43
Model 3 (${\boldsymbol{x}}'_K$)	1.79	1.58	1.98	1.74

Selected parameter	Training error		Testing error
Selected parameter	\|S₁₁\|	AR	\|S₁₁\|	AR
l₁, l₃, p, w₃ (Model 3)	1.79	1.58	1.98	1.74
g, l₂, w₁, w₂	3.45	3.76	3.36	3.62

Feature selection for determining input parameters in antenna modeling

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

Share this article

Figures/Tables 13

References 32

Related Articles 7

Recommended Articles

Metrics

Comments

[1]	Mingyu LI, Lu GAO, Hongwei XU, Kai LI, Yisong HUANG. Equipment damage measurement method of wartime based on FCE-PCA-RF [J]. Journal of Systems Engineering and Electronics, 2024, 35(3): 707-719.
[2]	Xiaolong XU, Wen CHEN, Xinheng WANG. RFC: a feature selection algorithm for software defect prediction [J]. Journal of Systems Engineering and Electronics, 2021, 32(2): 389-398.
[3]	Qiang QIN, Yunwen FENG, Feng LI. Structural reliability analysis using enhanced cuckoo search algorithm and artificial neural network [J]. Journal of Systems Engineering and Electronics, 2018, 29(6): 1317-1326.
[4]	Yintong Wang, Jiandong Wang, Hao Liao, and Haiyan Chen. Unsupervised feature selection based on Markov blanket and particle swarm optimization [J]. Systems Engineering and Electronics, 2017, 28(1): 151-.
[5]	Weiping Ding, Jiandong Wang, Zhijin Guan, and Quan Shi. Ensemble feature selection integrating elitist roles and quantum game model [J]. Systems Engineering and Electronics, 2015, 26(3): 584-594.
[6]	Yan Peng, Zheng Xuefeng, Zhu Jianyong & Xiao Yunhong. Lazy learner text categorization algorithm based on embedded feature selection [J]. Journal of Systems Engineering and Electronics, 2009, 20(3): 651-659.
[7]	Zheng Chunhong, Jiao Licheng & Li Yongzhao. Support vector classifier based on principal component analysis [J]. Journal of Systems Engineering and Electronics, 2008, 19(1): 184-190.