Design of Patch Antenna Utilizing Machine Learning Algorithms for Wireless Communication | ||
| SVU-International Journal of Engineering Sciences and Applications | ||
| Volume 6, Issue 2, December 2025, Pages 147-154 PDF (730.5 K) | ||
| Document Type: Original research articles | ||
| DOI: 10.21608/svusrc.2025.335633.1248 | ||
| Authors | ||
| Marwa Medhat Mohammed* 1; Ahmed Ibrahim2; Mohamed Hamada3 | ||
| 1Electronics and Communication Engineering Dep , Faculty of Engineering , Minia University Minia , Egypt | ||
| 2Electronics and Communication Engineering Dep, Faculty of Engineering ,Minia university Minia, Egypt | ||
| 3Electronics and Communication Engineering Dep, Faculty of Engineering , Minia University Minia , Egypt | ||
| Abstract | ||
| The efficient design of microstrip patch antennas is critical in modern wireless communication systems, where return loss (S11) significantly impacts system performance. However, conventional electromagnetic (EM) simulation methods are often computationally expensive and time-consuming, especially when optimizing multiple geometrical parameters. This creates a strong need for faster, more scalable prediction approaches. In this context, machine learning (ML) offers a promising solution by enabling accurate performance predictions without repeated simulations. This work aims to accelerate antenna design by evaluating and comparing the predictive capabilities of 18 different ML regression algorithms. A dataset of 20,021 entries was generated through EM simulations, with each entry reflecting varied geometrical parameters. The dataset was split into 80% for training and 20% for testing. While several models such as K-Nearest Neighbors (KNN), Random Forest, and Decision Tree performed well, the Extra Trees model achieved the best results, with an R-squared (R²) value of 0.999 and a Mean Squared Error (MSE) of just 0.0089. This study underscores the potential of ML to streamline antenna design and significantly enhance the overall development workflow. | ||
| Keywords | ||
| AI; machine learning; S11 prediction; microstrip antenna; optimization | ||
|
Statistics Article View: 6 PDF Download: 3 |
||
