Physics-Embedded Machine Learning for Efficient Modeling of High-Frequency Circuits

Yanan Liu, Hongliang Li, Jian Ming Jin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We propose the integration of analytical extension of eigenvalues (AEE) into neural network (NN) for efficient modeling of microwave circuits. By embedding the physics knowledge of lumped equivalent circuits, the proposed machine learning method allows us to build more generalizable relations between the input parameters of a circuit and its electromagnetic properties, showing a 10-fold improvement for input parameters outside the range of training set. The proposed model is also more data efficient in that it can achieve the same level of accuracy with fewer data compared with traditional NN-based circuit modeling.

Original languageEnglish (US)
Title of host publication2022 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, AP-S/URSI 2022 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages834-835
Number of pages2
ISBN (Electronic)9781665496582
DOIs
StatePublished - 2022
Externally publishedYes
Event2022 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, AP-S/URSI 2022 - Denver, United States
Duration: Jul 10 2022Jul 15 2022

Publication series

Name2022 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, AP-S/URSI 2022 - Proceedings

Conference

Conference2022 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, AP-S/URSI 2022
Country/TerritoryUnited States
CityDenver
Period7/10/227/15/22

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Signal Processing
  • Instrumentation

Fingerprint

Dive into the research topics of 'Physics-Embedded Machine Learning for Efficient Modeling of High-Frequency Circuits'. Together they form a unique fingerprint.

Cite this