In the current world of highly integrated communications, reliable and robust systems will be required to develop the 6G networks. The millimeter-wave band (30 GHz–100 GHz) and the sub-terahertz band (100 GHz–300 GHz) have promising possibilities in radar and communication systems, such as broad bandwidth, device miniaturization, and high integration with electronic technology. As 6G communications will be the dominant technology in the coming years, highly-accurate antenna design is becoming essential to building systems that meet the expected performance standards. Despite the wide availability of antenna models working at frequencies below 10 GHz, they need to be in-depth reviewed and reformulated, especially in the sub-terahertz band. Thus, the work developed in this doctoral dissertation provides a framework of analytical methods for electromagnetic antenna modeling, enabling the design of microstrip patch antennae up to 300 GHz. This work covers unprecedentedly diverse models in frequency ranges from radio frequency to the sub-terahertz band. The proposed model formulations consider the geometrical and electrical imperfections of materials used for antenna design. They show high accuracy in the modeled frequency response for measured antennas and transmission lines up to 110 GHz; and for simulated microstrip patch antennas up to 300 GHz, with thickness up to 5 % of the free-space wavelength, copper layers up to 35 μm thick, and with surface roughness up to 1 μm.