Computational Design and Analysis of a High-Isolation 5G MIMO Antenna Using a Binary GWO-Optimized Pixelated Metasurface

Compact 5G millimeter-wave (mm-Wave) multiple-input multiple-output (MIMO) systems face a serious challenge as high isolation is required for high spectral efficiency. This paper presents a novel computational design framework for enhancing the isolation of a two-port ultra-wideband (UWB) MIMO antenna, specifically targeting the 5G n257 band (26.5–29.5 GHz). A pixelated metasurface is presented and optimized with the help of a binary-coded Grey Wolf Optimizer (B-GWO) algorithm through a MATLAB-Computer Simulation Technology (CST) co-simulation interface, which is used in contrast to some conventional decoupling structures. A Geometric Mirror Symmetry method is used to accelerate the optimization process, which halves the number of optimization variables and significantly reduces the computational load. Crucially, this symmetry is also a fundamental requirement to ensure that the reflection coefficients (S11, S22) of the antennas remain identical. The proposed design achieves isolation levels better than 20 dB across the entire target band, reaching a peak isolation of 32.58 dB at 28.67 GHz, while maintaining reflection coefficients (S11, S22) below −10 dB. The MIMO diversity performance is comprehensively validated with an Envelope Correlation Coefficient (ECC) <0.005, a Diversity Gain (DG) of 9.99 dB, and a Total Active Reflection Coefficient (TARC) <−10 dB. Moreover, the suppression of surface waves enhances the realized gain to 4.51 dBi, providing a 0.57 dB improvement over the reference antenna. In addition, an equivalent passive RLC circuit model is constructed to observe the physical process of the pixelated surface, which shows the optimized structure as a band stop filter at the coupling frequency. The high correlation of the Equivalent Circuit Model and full-wave simulation outcomes confirms that the suggested design procedure is a strong verification alternative to physical fabrication.