Machine Learning Techniques for Multi-Objective Antenna Optimization (Invited)

Different types of microstrip antennas with non-uniform metasurfaces are designed and optimized using statistical methods with various formulations. The design objectives include impedance matching bandwidth, realized gain, and axial ratio bandwidth for linearly polarized antennas and circularly polarized antenna, respectively. The designed antennas are to operate in the sub-6 GHz frequency band of the 5G New Radio system.