Home > Published Issues > 2026 > Volume 21, No. 1, 2026 >
JCM 2026 Vol.21(1): 127-138
Doi: 10.12720/jcm.21.1.127-138

High-Performance Multiband THz MIMO Antenna for Future 6G Wireless Communications with Machine Learning Validation

Md. Ashraful Haque1, Md. Sharif Ahammed1, Jamal Hossain Nirob1, Jun-Jiat Tiangspan 2,*, and Narinderjit Singh Sawaran Singh 3
1Department of Electrical and Electronic Engineering, Daffodil International University, Dhaka, 1207 Bangladesh
2Centre for Wireless Technology, CoE for Intelligent Network, Faculty of Artificial Intelligence and Engineering, Multimedia University, Persiaran Multimedia, 63100 Cyberjaya, Selangor, Malaysia
3Faculty of Data Science and Information Technology, INTI International University, Persiaran Perdana BBN, Putra Nilai, Nilai 71800, Negeri Sembilan, Malaysia
Email: limon.ashraf@gmail.com (M.A.H.); sharif33-1152@diu.edu.bd (M.S.A.); jamal33-1243@diu.edu.bd (J.H.N.), jjtiang@mmu.edu.my (J.J.T.); narinderjits.sawaran@newinti.edu.my (N.S.S.S.)
*Corresponding author

Manuscript received June 19, 2025; revised August 19, 2025; accepted August 27, 2025; published February 12, 2026.

Abstract—This study presents a comprehensive industrial and innovation design along with an in-depth analysis of a THz Multiple Input Multiple Output (MIMO) antenna intended for future 6G communication systems. The antenna utilizes polyimide as the substrate and graphene as the patch material, with copper serving as the ground plane. This design enables the antenna to operate across six distinct frequency bands, making it a multiband antenna. The resonance frequencies of the antenna are 3.512 THz, 4.4448 THz, 6.704 THz, 7.672 THz, 8.664 THz, and 9.672 THz, with corresponding gains of 11.72 dB, 12.59 dB, 13.077 dB, 13.945 dB, 14.77 dB, and 16.028 dB, respectively. To further understand the electrical behavior of the antenna, an RLC equivalent circuit was developed using Advanced Design System (ADS) software. Subsequently, we employed supervised regression Machine Learning techniques following extensive data sampling using CST MWS (Microwave Studio) simulations. The results, highlighted by robust R-squared and variance scores, demonstrate that Extra Trees Regression delivers exceptional accuracy, approaching 98%. Additionally, it achieves the lowest error in predicting resonance frequency.


Keywords—industrial and innovation, THz antenna, Multiple-Input Multiple-Output (MIMO), Graphene, Sixth-Generation (6G), Envelop Correlation Coefficient (ECC), Radio Link Control (RLC)



Cite: Md. Ashraful Haque, Md. Sharif Ahammed, Jamal Hossain Nirob, Jun-Jiat Tiang*, and Narinderjit Singh Sawaran Singh, “High-Performance Multiband THz MIMO Antenna for Future 6G Wireless Communications with Machine Learning Validation," Journal of Communications, vol. 21, no. 1, pp. 127-138, 2026.



Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).

Article Metrics in Dimensions