University of Houston

We provide a thorough study on the impact of uncertainties in mobile and sensor networks where we investigate both unexpected, and desired uncertainties. Our work concern both theoretical analysis and applications of uncertainties. On the theory side, we propose high order maximum Voronoi diagram of disks to model sensor location suncertainties. On the application side, we focus on a number of problems in sensor networks and spatial queries in mobile networks. Particularly, we analyze network tasks including network coverage, and target localization and tracking when sensor location uncertainties uniformly distributed in disks of different sizes. Specifically, we propose an algorithm to compute the minimum sensing range for guaranteed k-coverage. In addition, we devise localization and tracking models for fast target localizationand tracking in sensor networks that cope with sensor location uncertainty. We further investigate uncertainties in spatial queries in mobile networks, where user location uncertainties are used as a mean to protect user privacy. We devise a framework for user identity privacy in k-nearest queries, in which we propose a k-anonymous cloaking method and a algorithm for k nearest neighbor search with cloaks. We further extend uncertainties into spatial skyline queries to cope with different kinds uncertainty including user location and fuzzy domination. The work opens up new directions in spatial skyline queries with presence of uncertainties.