Event detection and reservoir sampling in wireless sensor networks

Graduate Thesis uoadl:1324088 410 Read counter

Unit:
Department of Informatics and Telecommunications
Library of the School of Science
Deposit date:
2015-10-08
Year:
2015
Author:
Μπερσίμης Μάρκος
Supervisors info:
Χατζηευθυμιάδης Ευστάθιος
Original Title:
Ανίχνευση συμβάντων και δειγματοληψία δεξαμενής σε ασύρματα δίκτυα αισθητήρων
Languages:
Greek
Translated title:
Event detection and reservoir sampling in wireless sensor networks
Summary:
The main aim of this thesis is to analyze the temperature values sent over a
wireless sensor network using two different implementations. The first one
implements Shewhart algorithm processing every value sent in real time, whereas
the second one, accumulates the values and exports a sample using a reservoir
algorithm. From that sample, called reservoir, the max or the average value is
extracted and then forwarded as input in the Shewhart algorithm. Finally, the
efficiency of the two outputs is compared by means of Hamming distance.
Keywords:
sensors, Shewhart algorithm, event detection, reservoir algorithm, Hamming distance
Index:
Yes
Number of index pages:
8-10
Contains images:
Yes
Number of references:
32
Number of pages:
59
document.pdf (2 MB) Open in new window