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

Aug 5, 2026

Upload by Miles

Dual band microstrip patch antenna design targets industrial and innovation applications in 4G/5G (n77, n78) scenarios using a machine learning enhancement workflow. Antenna performance is evaluated through simulation and an RLC equivalent circuit model combined with ML-based supervised regression for gain prediction. The proposed antenna reaches a maximum gain of 6.56 dB with about 97% efficiency. Model quality is assessed with variance score, R-square, MSE, and MAE, where linear regression achieves the fewest errors and highest accuracy. CST and ADS modeling, together with ML results, indicate strong potential for 4G/5G use.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Article Info

Abstract

Introduction

Dual band antenna design for 4G/5G application and prediction of gain using machine learning approaches Narinderjit Singh Sawaran Singh1, Md. Ashraful Haque2, Redwan A. Ananta2, Md. Sharif Ahammed2, Md. Abdul Kader Jilani4, Liton Chandra Paul3, Rajermani Thinakaran1, Malathy Batumalay1, JosephNg Poh Soon1, Deshinta Arrova Dewi1 1Faculty of Data Science and Information Technology, INTI International University, Nilai, Malaysia 2Department of Electrical and Electronic Engineering, Daffodil International University, Dhaka, Bangladesh 3Department of Electrical, Electronic and Communication Engineering, Pabna University of Science and Technology, Pabna, Bangladesh 4Faculty of Physics and Astronomy, Friedrich Schiller University Jena, Jena, Germany | Article history:
Received Mar 26, 2024 Revised Dec 25, 2024 Accepted Jan 13, 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.
This is an open access article under the CC BY-SA license.
| | --- | --- | | Keywords:
4G/5G
Gain prediction Industrial and innovation Machine learning Microstrip patch antenna | | Corresponding Author: Narinderjit Singh Sawaran Singh Faculty of Data Science and Information Technology, INTI International University Persiaran Perdana BBN, Putra Nilai, Nilai 71800, Negeri Sembilan, Malaysia [Email: narinderjits.sawaran@newinti.edu.my](Email: narinderjits.sawaran@newinti.edu.my) Article Info ABSTRACT 1. INTRODUCTION The demand for antennas capable of operating across a wide range of frequencies has risen sharply as researchers explore methods to enhance the efficiency and versatility of wireless communication systems. The multiband microstrip patch antenna is an exceptional and advanced choice due to its capacity to accommodate multiple communication protocols and frequencies within a single, lightweight package. This makes it an ideal solution for modern communication needs, offering flexibility and high performance in a compact form [1] . Several methods are employed by the multiband microstrip patch antenna to accomplish this goal. These methods include the use of numerous resonators, parasitic elements, and tailored feeding networks. These advancements allow the antenna to resonate at different frequencies, making it suitable for use with a wide range of wireless communication protocols [2] . Microstrip patch antennas have advanced and found new uses in the internet of things (IoT), 5G connectivity, and other areas as wireless technology have expanded [3] . Cellular broadband communication standard, 4th generation long-term evolution (4G LTE) . It’sa major step forward in mobile network technology that should make it easier to stay online while on the go with gadgets like smartphones, tablets, and even some PCs. The use of the 2.4 GHz frequency range, a critical enabler for reliable IoT connection, is at the center of this revolution [4] . IoT devices can easily connect, communicate, and share data thanks to today’s 2.4 GHz wireless communic

FAQs

What antenna approach is presented for 4G/5G applications?

The document presents a dual band microstrip patch antenna design intended for 4G/5G contexts, specifically including n77 and n78 bands.

How is gain predicted using machine learning?

Predicted antenna gain is calculated using an alternative supervised regression ML technique, with model performance evaluated by metrics such as R2, MSE, and MAE.

What performance results does the proposed antenna achieve?

The maximum reported gain is 6.56 dB, with efficiency around 97%, supported by simulation and modeling results from CST and ADS.

Aug 5, 2026

Upload by Miles