Physics-Informed Artificial Neural Network-Based Estimation of Dielectric Properties for Resonant Method
Keywords:
Artificial Neural Network, Microwave resonator, Dielectric characterization, Quality factor, PermittivityAbstract
Accurate determination of dielectric properties, particularly relative permittivity (εr) and loss tangent (tan δ), is essential in microwave material characterization. This paper proposes an artificial neural network (ANN)-based approach that leverages resonant features of a rectangular cavity resonator to estimate dielectric properties. Full-wave electromagnetic simulations in Computer Simulation Technology Studio Suite were performed over the X-band frequency range (8–12 GHz), covering a wide set of low-loss dielectric samples with permittivity between 1 and 10 and loss tangent values from 0.001 to 0.2. Transmission responses were generated, from which resonant frequency and quality factor (Q) were extracted as input features for a two-layer ANN implemented in MATLAB. The model achieved excellent predictive accuracy, with coefficient of determination values exceeding 0.999 and mean percentage errors below 0.2% for both εr and tan δ. Results demonstrate that resonant frequency is highly sensitive to permittivity variations, while Q-factor effectively captures dielectric losses. The proposed framework was further validated using a Teflon sample in the X-band, where the ANN predicted a relative permittivity of 2.01 and a loss tangent of 0.0014, in close agreement with literature values. These results highlight the potential of the proposed ANN framework for efficient and non-destructive dielectric property estimation.References
[1] C. Li, Z. Li, C. Zhang, N. Cheng, L. Guan and D. Wang, Achieving reliable and intelligent control-assisted transmission for satellite communications, IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), Vancouver, Canada, 2024, 1–6.
[2] Y. Rahayu, B. Kurniawan and Y. P. Saputera, Narrow wall beamwidth slotted waveguide antenna for portable coastal radar communication, Journal of Engineering Science and Technology, 18, 2023, 60–70.
[3] M. M. Alam, N. A. T. Yusof, Y. A. Wahab, M. S. Abdul Karim and M. S. Rahman, Design and optimization of π-shaped slotted dual-band SIW antenna for 5G applications, Applications of Modelling and Simulation, 9, 2025, 92–106.
[4] J. Krupka, Microwave measurements of electromagnetic properties of materials, Materials, 14, 2021, 5097.
[5] J. Sun, W. Wang and Q. Yue, Review on microwave-matter interaction fundamentals and efficient microwave-associated heating strategies, Materials, 9, 2016, 231.
[6] N. S. I. Didik Aprianto, N. Hasan, N. I. I. Mohd Nadzri, N. Mustafa, L. F. Qi, A. A. Mohd Faudzi, N. A. Talip Yusof and M. S. Abdul Karim, Factorial analysis on the preparation of barium titanate-epoxy resin composite for antenna substrate, Progress in Electromagnetics Research C, 147, 2024, 127–134.
[7] L. Liu, M. Ye and W. Xue, Silicon-on-insulator-based narrowband microwave photonic filter with widely tunable bandwidth, Journal of Lightwave Technology, 41, 2023, 6341–6347.
[8] S. Stefanovski Pajovic, M. Potrebic and D. Tosic, Advanced filtering waveguide components for microwave systems, Microwave Systems and Applications, IntechOpen, 2017, 51–74.
[9] N. S. Mat Hussain, A. N. Azman, N. A. Talip Yusof, N. A. A. Hj Mohtadzar and M. S. Abdul Karim, Design of resonator cavity for liquid material characterization, TELKOMNIKA Telecommunication Computing Electronics and Control, 20, 2022, 447–454.
[10] M. Mrvic, S. Stefanovski Pajovic, M. Potrebic and D. Tosic, Design of microwave waveguide filters with effects of fabrication imperfections, Facta Universitatis Series Electronics and Energetics, 30, 2017, 431–458.
[11] S. Sahin, N. K. Nahar and K. Sertel, A simplified Nicolson–Ross–Weir method for material characterization using single-port measurements, IEEE Transactions on Terahertz Science and Technology, 10, 2020, 404–410.
[12] E. J. Rothwell, J. L. Frasch, S. M. Ellison, P. Chahal and R. O. Ouedraogo, Analysis of the Nicolson-Ross-Weir method for characterizing the electromagnetic properties of engineered materials, Progress in Electromagnetics Research, 157, 2016, 31–47.
[13] J. Bonello, A. Demarco, I. Farhat, L. Farrugia and C. V. Sammut, Application of artificial neural networks for accurate determination of the complex permittivity of biological tissue, Sensors, 20, 2020, 4460.
[14] Q. Tan, H. Zhu, W. Ma, Y. Yang and K. Huang, High temperature dielectric properties measurement system at 915 MHz based on deep learning, International Journal of RF and Microwave Computer-Aided Engineering, 29, 2019, e21948.
[15] M. S. Sim, K. Y. You, R. Dewan, F. Esa, R. A. Rashid and K. Y. Lee, Advances in artificial neural network assisted microwave sensors for material characterization, IEEE International Conference on Advanced Telecommunication and Networking Technologies, 2024, 1–4.
[16] G. C. Ram, S. Hanumantha Rao, G. R. L. V. N. Srinivasa Raju, D. Ramesh Varma, M. Venkata Subbarao, D. Girish Kumar, S. Tripura Siva Satya Tejaswini and P. Sai Naga Sreya, Design and development of a novel microwave sensor for biochemical detection, Proceedings on Engineering Sciences, 7, 2025, 1973–1980.
[17] C. Gocen and M. Palandoken, UHF RFID microwave sensor tag design for an RSSI-based machine learning assisted binary ethanol–water mixture characterization, IEEE Sensors Journal, 24, 2024, 262–269.
[18] M. S. Hossain, S. M. Iqbal and Y. Zhou, Microwave glucose concentration classification by machine learning, IEEE Texas Symposium on Wireless and Microwave Circuits and Systems, Waco, USA, 2020, 1–4.
[19] M. Islam, C. L. Min, N. J. Shoumy, M. S. Ali, S. Khatun, M. S. Abdul Karim and B. S. Bari, Software module development for non-invasive blood glucose measurement using ultra-wideband and machine learning, Journal of Physics Conference Series, 1529, 2020,052066.
[20] H. Guo, L. Yao and F. Huang, A cylindrical cavity sensor for liquid water content measurement, Sensors and Actuators A Physical, 238, 2016, 133–139.
[21] H. Kawabata, Y. Kobayashi and S. Kaneko, Analysis of cylindrical cavities to measure accurate relative permittivity and permeability of rod samples, Asia-Pacific Microwave Conference, Yokohama, Japan, 2010, 1459–1462.
[22] S. Zinal and G. Boeck, Complex permittivity measurements using TE11p modes in circular cylindrical cavities, IEEE Transactions on Microwave Theory and Techniques, 53, 2005, 1870–1874.
[23] M. S. Bin Abdul Karim, N. Binti Abu Talip Yusof and T. Kitazawa, Scattering analysis of rectangular cavity with input and output waveguides and its application to material characterization, IEEE Asia Pacific Microwave Conference, Kuala Lumpur, Malaysia, 2017, 588–591.
[24] W. Bacsa, R. Bacsa and T. Myers, Introduction: wave optics near surfaces, in Wave Optics Near Surfaces, 2020, 1–7.
[25] T. Ozturk, M. Hudlička and İ. Uluer, Development of measurement and extraction technique of complex permittivity using transmission parameter S21 for millimeter wave frequencies, Journal of Infrared Millimeter and Terahertz Waves, 38, 2017, 1510–1520.
[26] T. Han, Y. Seo and E. Choi, Significant reduction of resonant frequency by multi-layered dielectric material-loaded coaxial cavity for microwave heating, Journal of Electromagnetic Engineering and Science, 22, 2022, 1–7.
[27] J. Baker-Jarvis, Dielectric and Conductor-Loss Characterization and Measurements on Electronic Packaging Materials, NIST Technical Note, Gaithersburg, USA, 2001.
[28] A. Yakubu, Z. Abbas, N. Ibrahim and A. Fahad, The effect of ZnO nanoparticles filler on complex permittivity of ZnO–PCL nanocomposite at microwave frequency, Physical Science International Journal, 6, 2015, 196–202.
[29] H. Ohsato, J. Varghese and H. Jantunen, Dielectric losses of microwave ceramics based on crystal structure, Electromagnetic Materials and Devices, IntechOpen, 2018.
[30] I. R. Ibrahim, M. M. Rashid, M. R. Abdullah and M. A. Hassan, A study on microwave absorption properties of carbon black and Ni0.6Zn0.4Fe2O4 nanocomposites by tuning the matching-absorbing layer structures, Scientific Reports, 10, 2020, 3135.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Nur Sofia Idayu Didik Aprianto, Syamimi Mardiah Shaharum, Nurfarhana Mustafa, Ahmad Afif Mohd Faudzi, Ahmad Syahiman Mohd Shah, Toshihide Kitazawa, Mohamad Shaiful Abdul Karim
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:Authors hold and retain copyright, and grant the journal right of first publication, with the work after publication simultaneously licensed under a Creative Commons Attribution 4.0 License CC BY that permits any use, reproduction and distribution of the work and article without further permission provided that the original work is properly cited.
Authors are permitted and encouraged to post their work online in institutional repositories, website and other social media before and after publication, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
