Wireless communication is sensitive to ambient noise as well as interference due to the use of a shared medium, unlike wired communication. The range of frequencies used for wireless communication is limited, which prompts the reuse of frequencies causing increased interference. Wireless devices are often portable or mobile units with limited battery capacity, which makes effective low-power communication in harsh noise environments a high priority. In addition, wireless communication is significantly affected by the surrounding terrain like buildings, hills, foliage, etc. Many of these problems can be addressed through careful antenna design. Typically, omni-directional antennas are used in wireless communication due to their relative simplicity. These antennas radiate signal equally in all directions resulting in a waste of transmitter power and a source of interference to other nodes. More sophisticated techniques are available, but these can be challenging as they require complex software, which in turn demands powerful and expensive hardware. Use of adaptive directional antenna arrays can minimize interference while also being more energy efficient. Such “smart antennas” use digital beamforming based on signal processing algorithms to compute the appropriate weights to form effective antenna patterns. Smart antennas require the knowledge of the signal received at each antenna in the antenna array, thereby increasing the complexity of hardware and cost. Analog beamforming is a low hardware complexity alternative to smart antennas, which relies on a single transceiver and power splitter/combiner. Computer controlled analog amplifiers and phase shifters are used to form the desired beam patterns. The disadvantage of analog beamforming is that the computer only has access to the combined signals in the antenna array. We use analog beamforming in our implementation along with Noise-to-Signal ratio of nearby communicating nodes to mitigate this disadvantage. Conventional smart antennas optimize results for each individual node, while it is preferable to have a global optimal solution. It is also possible to compute the antenna array weights to adapt to the radio propagation characteristics of the surrounding terrain. This requires the knowledge of path loss at various distances and directions from the transmitter calculated using one of several available path loss models to compute a radio propagation map. This research has the following objectives: The first objective is to develop a new analog beamforming technique using Noise-to-Signal ratio as the performance measure. An iterative algorithm is developed that uses distributed node information along with analog beamforming, instead of individual antenna information. The antenna weights are sequentially adjusted across all nodes in the route to achieve optimization across the network. This provides a network optimized, low cost and hardware complexity solution for adaptive beamforming. The second objective is to explore another novel method called Virtual Terrain Leveling (VTL) where the effects of the terrain are virtually nullified using antenna arrays up to a specified distance from the transmitter. This work investigates whether VTL can mitigate the difficulties that the terrain features pose for implementation of wireless networks by attempting to provide uniform propagation in the presence of terrain with the use of directional antennas. The final objective is to use high performance computing (HPC) to develop more efficient storage and processing of radio propagation maps. Processing and storage of radio propagation maps can be computationally intensive based on the area and the resolution of the map. The technique of adaptive region construction is developed to efficiently store radio propagation maps while preserving the shape of the external contour for a given set of points. A GPU implementation of the algorithm with optimizations tailored to the hardware shows possible near - real time performance.