ABSTRACT

Future sensor networks will be more versatile in data recording. The sensors will not only collect information in numerical form, they will also collect and store textual data. The data will be accumulated in large quantities over a long period of time, and, many times, the data have to be stored in the network. This project aims to build an advanced sensor network data management system that extracts information from this widely distributed, massively accumulated, mixed and somehow unstructured data in an efficient and secure fashion.

The scope of this project goes beyond the prior research in sensor network data management by considering more sophisticated schemes of data storage, information retrieval, data mining, security and privacy support. It encompasses algorithm design, system design, real implementation and experiments to test the new ideas emerging in this future computing platform. Expected results include an information retrieval system for sensor networks, data mining algorithms for various problems (e.g., outlier detection), public key implementation and protocols for security and privacy support.

This project has a high impact in enriching the sensor network applications by supporting physical world search and more advanced queries for small devices. The research conducted in this project is expected to not only empower the future sensor networks, but also, in a more general sense, contribute to the research for the pervasively deployed small devices around people. Results from this project will be disseminated via conferences, journals, and the Internet.

Please report errors in award information by writing to: awardsearch@nsf.gov.