High-isolation dual-band MIMO antenna for next-generation 5G wireless networks at 28/38 GHz with machine learning-based gain prediction.

Journal: Scientific reports
Published Date: Jul 1, 2025

Abstract

This research outlines the results on implementing a Machine Learning (ML) approach to improve the throughput of Multiple-Input Multiple-Output (MIMO) based 5G millimeter wave applications. The research will cover frequencies between 28 and 38 GHz, significantly affecting high-band 5G applications. We have chosen to employ a Rogers RT 5880 material with a low loss as the substrate layer to reduce the antenna size. In addition to being small, the recommended design has a maximum gain of 10.14 dB, better isolation than 29 dB, and wide bandwidth, ranging from 27.2 GHz to 32.2 GHz & 36.5 GHz to 40.7 GHz. Advanced design system (ADS) is used to make a circuit like the suggested microstrip patch antenna (MPA) to compare the reflection coefficient from CST. The approach of supervised regression machine learning is applied to accurately forecast the antenna's gain. Among the five different regression machine learning models considered, it was discovered that the Random Forest Regression (RFR) model performed the best in accuracy and achieved the lowest error when predicting gain. This article explores many approaches, including simulation, integration of an RLC-equivalent circuit model, and multiple regression models, to evaluate the suitability of an antenna for its 5G applications.

Authors

  • Md Ashraful Haque
    Department of Electrical and Electronic Engineering, Daffodil International University, Dhaka, 1207, Bangladesh. limon.ashraf@gmail.com.
  • Redwan A Ananta
    Department of Electrical and Electronic Engineering, Daffodil International University, Dhaka, 1207, Bangladesh.
  • Md Sharif Ahammed
    Department of Electrical and Electronic Engineering, Daffodil International University, Dhaka, 1207, Bangladesh.
  • Jamal Hossain Nirob
    Department of Electrical and Electronic Engineering, Daffodil International University, Dhaka, 1207, Bangladesh.
  • Narinderjit Singh Sawaran Singh
    Faculty of Data Science and Information Technology, INTI International University, 71800, Nilai, Malaysia.
  • Liton Chandra Paul
    Department of Electrical, Electronic and Communication Engineering, Pabna University of Science and Technology, Pabna, Bangladesh.
  • Reem Ibrahim Alkanhel
    Department of Information Technology, College of Computer and Information Sciences, Princess Nourah Bint Abdulrahman University, P.O. Box 84428, 11671, Riyadh, Saudi Arabia.
  • Ahmed A Abd El-Latif
    EIAS Data Science Lab, College of Computer and Information Sciences, and Center of Excellence in Quantum and Intelligent Computing, Prince Sultan University, 11586, Riyadh, Saudi Arabia.
  • May Almousa
    Department of Information Technology, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, P.O. Box 84428, Riyadh, 11671, Saudi Arabia. mmalmousa@pnu.edu.sa.
  • Abdelhamied A Ateya
    EIAS Data Science Lab, College of Computer and Information Sciences, and Center of Excellence in Quantum and Intelligent Computing, Prince Sultan University, 11586, Riyadh, Saudi Arabia. aateya@psu.edu.sa.

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals