Sensors 201212(12), 17074-17093; doi:10.3390/s121217074
Article

Geosensor Data Representation Using Layered Slope Grids

Yongmi Lee 1 emailYoung Jin Jung 2,emailKwang Woo Nam 3 emailSilvia Nittel 4 emailKate Beard 4 email and Keun Ho Ryu 1 email
1 Database/Bioinformatics Lab, Chungbuk National University, Cheongju 361-763, Korea2 Korea Institute of Science Technology and Information, 245 Daehangno, Yuseong, Daejeon 305-806, Korea3 Department of Computer and Information Engineering, Kunsan National University, Kunsan 573-701, Korea4 School of Computing and Information Science, University of Maine, Orono, 5711 Boardman Hall, Rm. 344, Orono, ME 04467, USA
* Author to whom correspondence should be addressed.
Received: 17 October 2012; in revised form: 4 December 2012 / Accepted: 6 December 2012 / Published: 12 December 2012
(This article belongs to the Section Sensor Networks)
PDF Full-text Download PDF Full-Text [3257 KB, uploaded 12 December 2012 11:03 CET]
Abstract: Environmental monitoring applications are designed for supplying derived and often integrated information by tracking and analyzing phenomena. To determine the condition of a target place, they employ a geosensor network to get the heterogeneous sensor data. To effectively handle a large volume of sensor data, applications need a data abstraction model, which supports the summarized data representation by encapsulating raw data. For faster data processing to answer a user’s queries with representative attributes of an abstracted model, we propose such a data abstraction model, the Layered Slopes in Grid for Sensor Data Abstraction (LSGSA), which is based on the SGSA. In a single grid-based layer for each sensor type, collected data is represented by slope directional vectors in two layered slopes, such as height and surface. To answer a user query in a central monitoring server, LSGSA is used to reduce the time needed to extract event features from raw sensor data as a preprocessing step for interpreting the observed data. The extracted features are used to understand the current data trends and the progress of a detected phenomenon without accessing raw sensor data.
Keywords: sensor data abstraction; sensor data representation; geosensor network; slope grid; GIS; surface model

 

Advertisements