Metamaterial based tri-band compact MIMO antenna system for 5G IoT applications with machine learning performance verification

Citation

Rahman, Md Afzalur and Al-Bawri, Samir Salem and Larguech, Samia and Alharbi, Sultan S. and Alsowail, Saeed and Jizat, Noorlindawaty Md. and Islam, Mohammad Tariqul (2025) Metamaterial based tri-band compact MIMO antenna system for 5G IoT applications with machine learning performance verification. Scientific Reports, 15 (1). ISSN 2045-2322

[img] Text
s41598-025-06391-1.pdf - Published Version
Restricted to Repository staff only

Download (5MB)

Abstract

This paper presents a novel tri-band Multiple Input Multiple Output (MIMO) antenna module designed for millimeter and microwave frequency bands, employing metamaterial (MTM) technology to enhance performance. The compact antenna module measures 36 × 36 × 1.6 mm2 and uses a Rogers RT-5880 substrate. Its structure includes a multi-stubbed radiating patch, a partial ground plane, and a 2 × 1 epsilon-negative MTM array positioned between antenna elements in an orthogonal layout. Operating at 3.5 GHz, 5.2 GHz, and 28 GHz, the integration of MTM significantly improves the antenna’s overall performance by influencing phase, amplitude, and electromagnetic field distribution. Bandwidth enhancements of 10.01% and 6.4% are achieved for the 3.5 GHz and 5.2 GHz microwave bands, respectively, and 4.43% for the 28 GHz millimeter-wave band. Isolation levels improved from 20 dB to 24 dB in microwave bands and from 26 dB to 32 dB in the millimeter-wave band, ensuring reduced interference. The realized gain also increased from 3.6 dBi, 4.2 dBi, and 7.4 dBi to 4.8 dBi, 5.3 dBi, and 9.3 dBi across the respective frequency bands. The proposed MIMO antenna showcases excellent diversity performance with an envelope correlation coefficient (ECC) of below 0.002/0.001/0.0003 across all bands and a diversity gain (DG) exceeding 9.98 dB. Machine learning-based performance verification analysis assessed bandwidth and efficiency, where the K-Nearest Neighbors (KNN) model achieved 97.8% accuracy. This MIMO antenna holds great potential for various Internet of Things (IoT) applications, including Vehicle-to-Network, Vehicle-to-Cloud communications, 5G cellular networks, Wi-Fi, WiMAX, and both sub-6 GHz and millimeter-wave 5G bands, reinforcing its suitability for 5G IoT sectors.

Item Type: Article
Uncontrolled Keywords: MIMO, Metamaterials, Multiband, IoT, Fifth generation, Machine learning
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television
Divisions: Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User: Ms Suzilawati Abu Samah
Date Deposited: 29 Jul 2025 03:56
Last Modified: 29 Jul 2025 03:56
URII: http://shdl.mmu.edu.my/id/eprint/14365

Downloads

Downloads per month over past year

View ItemEdit (login required)