In this paper, a radiating element consisting of a modified circular patch is proposed for MIMO arrays for 5G millimeter-wave applications. The radiating elements in the proposed 2 × 2 MIMO antenna array are orthogonally configured relative to each other to mitigate mutual coupling that would otherwise degrade the performance of the MIMO system. The MIMO array was fabricated on Rogers RT/Duroid high-frequency substrate with a dielectric constant of 2.2, a thickness of 0.8 mm, and a loss tangent of 0.0009. The individual antenna in the array has a measured impedance bandwidth of 1.6 GHz from 27.25 to 28.85 GHz for S₁₁ ≤ −10 dB, and the MIMO array has a gain of 7.2 dBi at 28 GHz with inter radiator isolation greater than 26 dB. The gain of the MIMO array was increased by introducing frequency-selective surface (FSS) consisting of 7 × 7 array of unit cells comprising rectangular C-shaped resonators, with one embedded inside the other with a central crisscross slotted patch. With the FSS, the gain of the MIMO array increased to 8.6 dBi at 28 GHz. The radiation from the array is directional and perpendicular to the plain of the MIMO array. Owing to the low coupling between the radiating elements in the MIMO array, its Envelope Correlation Coefficient (ECC) is less than 0.002, and its diversity gain (DG) is better than 9.99 dB in the 5G operating band centered at 28 GHz between 26.5 GHz and 29.5 GHz.

This paper presents a novel approach to reducing undesirable coupling in antenna arrays using custom-designed resonators and inverse surrogate modeling. To illustrate the concept, two standard patch antenna cells with 0.07λ edge-to-edge distance were designed and fabricated to operate at 2.45 GHz. A stepped-impedance resonator was applied between the antennas to suppress their mutual coupling. For the first time, the optimum values of the resonator geometry parameters were obtained using the proposed inverse artificial neural network (ANN) model, constructed from the sampled EM-simulation data of the system, and trained using the particle swarm optimization (PSO) algorithm. The inverse ANN surrogate directly yields the optimum resonator dimensions based on the target values of its S-parameters being the input parameters of the model. The involvement of surrogate modeling also contributes to the acceleration of the design process, as the array does not need to undergo direct EM-driven optimization. The obtained results indicate a remarkable cancellation of the surface currents between two antennas at their operating frequency, which translates into isolation as high as −46.2 dB at 2.45 GHz, corresponding to over 37 dB improvement as compared to the conventional setup.