Design and development of ultra-broadband THz metamaterial MIMO antenna with efficient diversity parameters optimized with machine learning for TWPAN applications

TL;DR

This paper presents a machine learning-optimized ultra-broadband THz MIMO antenna design suitable for high-speed TWPAN applications.

Contribution

The novel contribution is the integration of machine learning optimization to enhance THz MIMO antenna performance with minimal correlation and high gain.

Findings

The antenna achieves a high gain of 15.7 dBi and a 20 THz bandwidth.

MIMO diversity parameters show ECC near 0 and DG of 10 dB, indicating strong performance.

Machine learning optimization yields an R2 value of 0.99, minimizing prediction errors.

Abstract

The high-speed communication development is revolutionizing the way with interact with technology by enabling ultra-fast and intelligent connectivity. There is a need for antenna design that operates with ultrabroadband in the THz regime to be applicable for Terahertz Wireless Persona Area Network (TWPAN) applications. We have proposed an ultra-fast, broadband, and high-gain MIMO antenna design which not only smart but also small in size and low cost to be considered for high-speed communication applications. The designed antenna shows a high gain of 15.7 dBi. The ultrabroadband response gives a bandwidth of 20 THz. The MIMO diversity parameters show the ECC value near 0 and DG of 10 dB. The CCL values are also 0.0083 bits/Hz. Their values show that there is minimal correlation, which means better MIMO performance. The performance is also optimized using parametric optimization and machine learning optimization. The machine learning algorithm gives the highest R2 value of 0.99, which gives a minimum prediction error and higher antenna performance. The THz metamaterial design with optimum diversity parameters makes it a good candidate for TWPAN applications.

Figures