Date of Award

Spring 1-1-2018

Document Type


Degree Name

Master of Science (MS)

First Advisor

Eric W. Frew

Second Advisor

Nisar Ahmed

Third Advisor

Tam Vu


Small Unmanned Aircraft Systems (sUAS) are attracting significant attention for their use in a wide range applications. These applications can be categorized into two types based on their communication objective; communication-focused or communication-enabling. In both types it is beneficial to completing its mission if the sUAS is communicational-aware or it has information to assess the performance of a given communication channel. In the literature there is a lack of adaptable, robust, and online methods to provide the necessary information to be communication-aware. This thesis expands on a few authors' work to formally define and assess an online hybrid architecture that is also adaptable and robust. The architecture has a Bayesian approach combining an a-priori RF propagation model with a machine learning correction tool to provide an initial estimate and learn the deviation of that a-priori model to provide a combined prediction at any given point. The hybrid architecture is implemented and a series of assessments using simulation and flight data are performed on its capabilities and performance. The results from these assessments are that the online hybrid architecture provides a benefit over learning the field directly and when applied to a wireless airborne relaying application can perform as well as naive approaches.