Danqing Xiao

Date of Award


Level of Access

Campus-Only Thesis

Degree Name

Master of Science (MS)


Spatial Information Science and Engineering


Silvia Nittel

Second Committee Member

Michael F. Worboys

Third Committee Member

Kate Beard-Tisdale


Wireless sensor networks become increasingly important in environmental monitoring applications to monitor spatial but non-topological spatial changes. This thesis focuses on modeling metric/Euclidean spatial changes of 2-dimensional continuous phenomena represented as 2D areal objects in observation fields. Metric changes represent the changes of quantitative spatial properties of spatial objects over time, while Euclidean changes additionally capture the directional change. By surveying previous research w.r.t. change detection in environmental monitoring applications, we propose four non-topological changes: expansion, contraction, change in shape and change in location. These four major non-topological changes are then integrated with existing mathematical models for non-topological spatial changes; the result is captured in an ontology written in OWL (Web Ontology Language). Selecting one of the non-topological changes, i.e. ?change in size?, we propose a novel in-network algorithm that efficiently detects the area change of a 2D-phenomenon with wireless sensor network applications. This algorithm calculates area through local aggregation of hierarchical sensors without connecting all boundary nodes and maintaining the boundary loops. This in-network area detection algorithm through local aggregation has O(n) complexity. As shown in the simulation analysis, this area change detection algorithm computes area of dynamic phenomena modeled as areal objects accurately and energy-efficient. The communication cost is significantly reduced while the accuracy is similar to traditional centralized approach. Moreover, this in-network area computation algorithm is independent of the topological properties of the areal objects.