Monopole Antenna Parameters Prediction using Machine Learning for IoT Systems | ||
| SVU-International Journal of Engineering Sciences and Applications | ||
| Volume 6, Issue 2, December 2025, Pages 130-137 PDF (636.45 K) | ||
| Document Type: Original research articles | ||
| DOI: 10.21608/svusrc.2025.356607.1267 | ||
| Authors | ||
| Ahmed F Abdullah* 1; Gamal A El sheikh1; Ahmed Ibrahim2 | ||
| 1Electronics and Communication Engineering Department, PHI, 6 October, Giza, Egypt | ||
| 2Electrical Engineering Department, Faculty of Engineering, Minia University, Minia, Egypt | ||
| Abstract | ||
| Antenna design necessitates heavy simulation processes that require effort and time. However, the great developments in artificial intelligence (AI) approaches and the availability of pertinent computational facilities encouraged researchers to overcome the design constraints. Thus, this paper is devoted to presenting an approach for designing an antenna with an elliptical radiator for Internet of Things (IoT) applications via a machine learning (ML) algorithm. Machine learning (ML) algorithms are employed to optimize antenna designs, aiming to minimize simulation time and accelerate the overall design process. The geometric parameters of the antenna serve as inputs to the ML model, with a dataset comprising 200,200 samples. The model focuses on two output parameters: bandwidth and the reflection coefficient (S11). Initially, efforts were directed toward predicting the decibel magnitude of the reflection coefficient using various ML algorithms, the outputs are compared with each other to justify its performance. In addition, predicted outcomes obtained from ML algorithms are compared with those obtained from the simulation results to justify the accuracy of these approaches. The antenna design is suitable for the sub-6 GHz frequency spectrum from 3.55 GHz to 6.9 GHz. The Random Forest Regressor algorithm delivered the most accurate results for predicting the reflection coefficient parameters, achieving 0.99927% R-squared, 0.0341% MSE, 0.0443% MAE, and 0.184% RMSE values | ||
| Keywords | ||
| Antenna Design; Artificial Intelligence; Internet of Things (IoT); Machine Learning | ||
|
Statistics Article View: 21 PDF Download: 11 |
||
