Metasurface absorber for millimeter waves: a deep learning-optimized approach for enhancing the isolation of wideband dual-port MIMO antennas.

In this communication, the concept of metasurface absorber is utilized to enhance the isolation in the dual port multiple-input multiple-output (MIMO) antenna specially designed for a wideband millimeter wave operation. The frequency of operation of the designed module is 32.5-42.5 GHz with sufficient gain attributes. The designed metasurface array consists of two circular rings on two different layers. The concept of deep learning is utilized to optimize the dimensional configuration of the metasurface to achieve the maximum absorption of electromagnetic waves in the band of interest. The suppression of mutual coupling by double-ring metasurface is analyzed with the help of the wave theory concept. In contrast to previous decoupling methods using metasurfaces, the suggested metasurface is intended to be in the same plane as the array. The findings demonstrate the capability of effectively separating antenna elements in wideband MIMO antenna without compromising the geometrical complexity. Diversity metrics such as envelope correlation coefficient (ECC), mean effective gain (MEG), channel capacity loss (CCL), and total active reflection coefficient (TARC) for the proposed frequency range are also used to evaluate the performance of the constructed MIMO antenna. The wideband characteristics with a compact configuration make the design MIMO module a suitable candidate for mm-wave applications. The congruence between simulation and measurement confirms the validity of the suggested design.