Quad-port 5G antenna system optimized with ANN–MFO hybrid approach

This paper focuses on the design and optimization of a 4-Port Multiple Input Multiple Output (MIMO) antenna suitable for 5G smartphones. The antenna is designed to operate in the frequency band of 3.3-3.8 GHz and fabricated on FR4 substrate. Leveraging Defected Ground Structure (DGS) and slot structures, the return loss performance of the radiating element is enhanced. A single radiating element is optimized through parametric analysis and modelled using an equivalent circuit. As the relative positioning of the radiating elements would greatly alter the impedance matching and coupling characteristics, the space between elements needs to be optimized. A hybrid Artificial Neural network (ANN)-Moth Flame Optimization (MFO) framework is used to determine the optimum positioning of the antenna elements. Results from simulations, modelling and experimental measurements are found to be in good agreement. Furthermore, SAR analysis revealed that the radiation levels are within the prescribed safety limits.