Date of Award


Document Type


Degree Name

Master of Science (MS)


Electrical Engineering


Due to the unprecedented spatial and temporal resolution they offer, wireless sensor networks are considered an enabling technology for the distributed monitoring of industrial, military, and natural environments. As these systems migrate into vastly different and novel applications, new constraints are discovered that affect network reliability and utility. For example, wireless sensors are typically statically deployed and, unlike mobile systems, cannot move to a new location for better radio reception. As a result, the signal fades caused by non-optimal environmental conditions can increase the outage probability of the system, potentially rendering the network unreliable and ineffectual. Stochastic models that quantify link reliability and the effectiveness of diversity methods are often employed to understand the impact of such fading. However, the performance of these models applied to wireless sensor networks is entirely dependent on the appropriateness of the model with respect to the environment. This work first presents an empirical study of the propagation environment for a wingless, rotary aircraft, showing that the wireless environment within exhibits frequency-selective fading much more severe than predicted by current worst-case models (i.e., Rayleigh). An analysis is then given of the effectiveness of several diversity methods operating within such environments (referred to as hyper-Rayleigh). These fade mitigation techniques are simple enough to be employed for use with low-complexity wireless sensor hardware, and include spatial diversity, polar diversity, two-element passive combining, and two-element phased combining. Two-element phased combining is further developed by examining the effect that smaller element spacing has on diversity gain. A demonstration of a wireless