Experiments in Data Management for Wireless Sensor Networks

Author

Davcev, Danco ; Kulakov, Andrea ; Gancev, Stojanco

Author_Institution

Fac. of Electr. Eng. & Inf. Technol., Univ. of Ss. Cyril & Methodius, Skopje

fYear

2008

fDate

25-31 Aug. 2008

Firstpage

191

Lastpage

195

Abstract

In this paper we present our experimental results in data management for wireless sensor networks when an adapted Fuzzy ART model of neural networks is applied. Our system provides high dimension reduction and transfer savings, sending only classified identification number instead of whole sample. The system implementation is based on MicaZ sensor motes and adapted Fuzzy ART model. It was applied in real-time environments and presents an efficient tool for monitoring and visualization of sensor data. Testing, debugging, data robustness and other implementation issues are also discussed.

Keywords

data visualisation; fuzzy neural nets; telecommunication computing; telecommunication network management; wireless sensor networks; MicaZ sensor; data management; fuzzy ART model; high dimension reduction; neural networks; wireless sensor networks; Data visualization; Debugging; Fuzzy neural networks; Fuzzy systems; Monitoring; Neural networks; Sensor systems; Subspace constraints; Testing; Wireless sensor networks; data robustness; experimental results; neural-networks; real time visualization; unsupervised pattern learning; wireless sensor networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Sensor Technologies and Applications, 2008. SENSORCOMM '08. Second International Conference on

Conference_Location

Cap Esterel

Print_ISBN

978-0-7695-3330-8

Electronic_ISBN

978-0-7695-3330-8

Type

conf

DOI

10.1109/SENSORCOMM.2008.18

Filename

4622661