Skip to main navigation Skip to search Skip to main content

Resonant Mode Recognition with Convolutional Neural Network

Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review

Abstract

This research proposes a simple approach to recognize resonate modes in rectangular dielectric resonator antennas (DRAs) with machine learning (ML) techniques. The proposed method utilizes a convolutional neural network (CNN) with a focal loss function to classify various modes present in the antenna. DRAs with different dimensions are simulated with ANSYS HFSS to obtain their electric fields at various resonant frequencies. The resulting dataset comprises 700 fields, manually labeled according to their corresponding resonant modes. After training the CNN model using this dataset, it is capable of accurately classifying 12 low-order resonant modes in rectangular DRAs. The performance of the model is evaluated using a test set, where it achieves a recognition accuracy of 97.8%. © 2023 IEEE.
Original languageEnglish
Title of host publication2023 International Conference on Microwave and Millimeter Wave Technology (ICMMT 2023) - Proceedings
PublisherIEEE
Number of pages3
ISBN (Electronic)9798350338874
ISBN (Print)9798350338881
DOIs
Publication statusPublished - 2023
Event15th International Conference on Microwave and Millimeter Wave Technology (ICMMT 2023) - Qingdao, China
Duration: 14 May 202317 May 2023

Publication series

NameInternational Conference on Microwave and Millimeter Wave Technology, ICMMT - Proceedings

Conference

Conference15th International Conference on Microwave and Millimeter Wave Technology (ICMMT 2023)
PlaceChina
CityQingdao
Period14/05/2317/05/23

Research Keywords

  • Artificial intelligence
  • convolutional neural network
  • dielectric resonator antenna
  • imbalanced dataset
  • machine learning
  • resonant mode

RGC Funding Information

  • RGC-funded

Fingerprint

Dive into the research topics of 'Resonant Mode Recognition with Convolutional Neural Network'. Together they form a unique fingerprint.

Cite this