Information fusion for self-organizing sensor networks

Sensor networks have been widely deployed for environmental monitoring which involves collecting readings over time for contaminant propagation, building comfort, intrusion detection and so on. In such emerging applications, data is collected in the form of continuous data streams, as opposed to finite stored databases. Due to the dynamic nature of the underlying data, several data compression and transformation techniques are well utilized in related studies. For answering user queries, both approximation and adaptability are identified to be important issues for effective synopsis maintenance. In this paper, the concept of self-organizing networks is extended through novel data management techniques. Specifically, information fusion with data localization and resource awareness for energy conservation are addressed.