Dual band antenna design for 4G/5G application and prediction of gain using machine learning approaches

Narinderjit Singh Sawaran Singh (INTI International University)
Md. Ashraful Haque (Daffodil International University)
Redwan A. Ananta (Daffodil International University)
Md. Sharif Ahammed (Daffodil International University)
Md. Abdul Kader Jilani (Friedrich Schiller University Jena)
Liton Chandra Paul (Pabna University of Science and Technology)
Rajermani Thinakaran (INTI International University)
Malathy Batumalay (INTI International University)
JosephNg Poh Soon (INTI International University)
Deshinta Arrova Dewi (INTI International University)

TELKOMNIKA (Telecommunication Computing Electronics and Control)
Vol 23, No 2: April 2025

In this research, we disclose our findings from exploring a machine learning (ML) approach to enhancing the antenna’s performance in Industrial and Innovation contexts, particularly for4G and 5G (n77, n78) contexts. Methods for evaluating antenna performance utilizing simulation, the resistor, inductor, and capacitor (RLC) equivalent circuit model, and ML are discussed. Gain is a maximum of 6.56 dB and efficiency is about 97% for this antenna. The predicted antenna gain is calculated using an alternative supervised regression ML technique. Multiple measures, including as the variance score, R-square (R2), mean square error (MSE), and mean absolute error (MAE), can be used to assess an ML model’s performance. The linear regression (LR) model predicts profit with the fewest errors and highest accuracy of the five ML models. Finally, computer simulation technology (CST) and advanced design system (ADS) modeling findings, along with ML results, show that the proposed antenna is a promising option for 4G and 5G applications.