Services in PLACE are characterized by the continuous and concurrent evaluation of multiple queries whose responses depend on progressively accumulated data. We refer to such queries as active queries. Our work is driven by (i) solutions that maintain scalability as the number of users and services increase and (ii) achieving real-timeliness in an environment requiring the continuous and concurrent evaluation of active queries. The project includes the development of a flexible and adaptive framework for query execution. Active queries are either executed at the servers or at the objects issuing the queries. For query processing at the servers, we have developed a novel indexing techniques to match data to queries. For query processing at the objects, we employ different data dissemination solutions based on bandwidth, communication costs, and computational capabilities of the objects.
The project includes the design of a small-scale,
GPS-based location-aware system which makes
use of the campus's wireless service areas.
A server tracks, monitors, and responds to queries of objects
moving on the Purdue campus.
The combination of off-the-shelve GPS devices (from
Garmin, Magellan, and Pharos), mobile handheld devices (including
HP-Jornada's and Compaq-iPAQ's) and web-based map servers has resulted in a
novel use of existing technologies.
The prototype sysyem allows us to
realistically observe and evaluate relevant parameters and allows
us to evaluate
active query processing, server- versus object-processing approaches,
and data dissemination strategies.
PLACE Publications
Related projects
GPS related links