With the ubiquitous use of wireless technology today, we are experiencing a severe shortage of spectrum. The terahertz frequency band (100 GHz to 10 THz) is a largely unused part of the spectrum that can potentially be used for high rate short-range wireless links thus easing the spectrum pressure to some degree. However, this frequency band suffers from large signal attenuation with distance, resulting in a need for highly directional transmissions (or pencil beams). This research identifies fundamental challenges in implementing such directional transmissions for providing coverage in rooms and other short-range application domains. Additionally, the project examines the problem of allocating spectrum resources to mobile users given the highly directional nature of coverage, which causes a spatial dependence on frequency availability. The impact of this work is broad: it addresses the fundamental problem of spectrum scarcity by examining a hitherto unexplored part of the spectrum and it will impact the emerging 5G wireless standard. The PI will also teach a graduate seminar class on terahertz communications, thus impacting education. The project describes the frequency and angular dependence of using dense antenna arrays to provide highly directional coverage across the terahertz spectrum. A consequence of this behavior is that as a user moves even a few feet, the data rate can drop by orders of magnitude, unless coverage is planned carefully. This project presents a systematic approach to studying the coverage problem for terahertz networks using clusters of dense antenna arrays. The project also examines a novel approach of using lenses to provide coverage. The first problem will be studied by building a detailed terahertz propagation simulator, which includes models for dense antenna arrays and models for mutual coupling effects (which are significant at this frequency band). The latter approach will be studied using a large set of measurements. Given these studies, the project will then examine the problem of providing coverage to mobile users in rooms. Since spectrum resources show a spatial dependence, the problem of ensuring users continue to receive their required quality of service will combine resource allocation approaches with geometric constraints. The outcome of this research will be a detailed understanding of the terahertz communication channel as well as a set of tools and measurements that can be employed by other researchers in the field.