Work performed under this cooperative agreement between AFRL Space Vehicles Directorate, and the University of New Mexico focused on fundamental research to establish the communications-theoretic foundations for future Extremely High Frequency (EHF) satellite and terrestrial communications. This project started the investigation of two of the promising new concepts in wireless communications, namely the Intelligent Reflecting Surfaces (IRS) and 3D beamforming, in the context of millimeter wavelength satellite communications. Simulations were conducted to assess the performance of multiple user networks at higher frequencies. Two algorithms, a reward based greedy algorithm and a deep reinforcement learning algorithm implemented with a Deep Q-Network, were proposed for determining an optimal trajectory for a UAV base station to maximize the coverage of distinct ground users. Finally, simulations were conducted to assess performance impacts in an adversarial environment.