A Machine Learning Based Design of Frequency Reconfigurable Compact Microstrip Patch Antenna

Journal article

Mohd. Yusuf, Nilesh Dipak Bhandare, Sahaya Anselin Nisha A, Sourajeet Roy
2023 IEEE Microwaves, Antennas, and Propagation Conference (MAPCON), 2023

Semantic Scholar DOI

Cite

APA Click to copy
Yusuf, M., Bhandare, N. D., A, S. A. N., & Roy, S. (2023). A Machine Learning Based Design of Frequency Reconfigurable Compact Microstrip Patch Antenna. 2023 IEEE Microwaves, Antennas, and Propagation Conference (MAPCON).

Chicago/Turabian Click to copy
Yusuf, Mohd., Nilesh Dipak Bhandare, Sahaya Anselin Nisha A, and Sourajeet Roy. “A Machine Learning Based Design of Frequency Reconfigurable Compact Microstrip Patch Antenna.” 2023 IEEE Microwaves, Antennas, and Propagation Conference (MAPCON) (2023).

MLA Click to copy
Yusuf, Mohd., et al. “A Machine Learning Based Design of Frequency Reconfigurable Compact Microstrip Patch Antenna.” 2023 IEEE Microwaves, Antennas, and Propagation Conference (MAPCON), 2023.

BibTeX Click to copy

@article{mohd2023a,
  title = {A Machine Learning Based Design of Frequency Reconfigurable Compact Microstrip Patch Antenna},
  year = {2023},
  journal = {2023 IEEE Microwaves, Antennas, and Propagation Conference (MAPCON)},
  author = {Yusuf, Mohd. and Bhandare, Nilesh Dipak and A, Sahaya Anselin Nisha and Roy, Sourajeet}
}

Abstract

This paper proposes a machine learning (ML) based compact microstrip patch frequency reconfigurable antenna for multiple frequency band application. The antenna consists of four metalized patches symmetrically arranged and coupled via slots and metallic bars, respectively. Variations in the material, geometry, and bias parameters of the microstrip patch antenna, along with the R, L, and C values of the equivalent circuits of the varactor diode, add frequency reconfigurability. Frequency reconfigurability is predicted using the artificial neural network (ANN) surrogate model for multiple frequency bands of 0.5 GHz – 7.5 GHz with very high accuracy and time efficient manner as compared to full-wave electromagnetic simulations solvers such as high-frequency structure simulator (HFSS) or CST microwave studio (MWS).