Wireless sensor networks (WSNs) can provide real-time information about geospatial environments, and so have the potential to play an important role in the monitoring of geographic phenomena. The research reported in this paper uses WSNs to provide salient information about spatially distributed dynamic fields, such as regional variations in temperature or concentration of a toxic gas. The focus is on topological changes to areas of high-activity that occur during the evolution of the field. Topological changes investigated include region merging and splitting, and hole formation or elimination. Such changes are formally characterized, and an algorithm is developed that detects such changes by means purely of in-network processing. The efficiency of this algorithm is investigated both theoretically and using simulation experiments.
Jixiang Jiang, Michael F. Worboys