"BSRF_dataset_S-parameters"

Read full text Summarise with VIRAT
Authors: Gayatri Routhu
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

"To predict reflection coefficient (|S11| < -10 dB) and transmission coefficient (|S21| < -50 dB) parameters based on the trained data. The trained model can predict the performance of the BSRF filter based on their design specifications. Three machine learning models are developed and compared, including Support Vector Machine (SVM), Random Forest (RF), and Artificial Neural Networks (ANN). Based on four performance evaluation metrics: coefficient of determination (R\\textsuperscript{2}), mean square error (MSE), root mean square error (RMSE), and mean absolute error (MAE), the developed models are compared. The BSRF filter is designed, measured, predicted, and validated with respect to the developed models. The developed models achieve good prediction accuracy with an R2 range from 0.980 to 0.999. The ANN achieves the highest accuracy in the prediction of S-parameters."

Fields of Study & Subjects

Microwave Engineering and Waveguides Machine Learning in Materials Science Microwave and Dielectric Measurement Techniques

Publication details

Journal / source
IEEE DataPort
Publication date
26 March 2026
Language
EN
Document type
Dataset