Unit:
Κατεύθυνση Ηλεκτρονικός Αυτοματισμός (Η/Α, με πρόσθετη εξειδίκευση στην Πληροφορική και στα πληροφοριακά συστήματα)Library of the School of Science
Author:
Παπαδόπουλος Μιχαήλ
Supervisors info:
Χατζηευθυμιάδης Ευστάθιος Επίκ. Καθηγητής
Original Title:
Μηχανική εκμάθηση σε ασύρματα δίκτυα αισθητήρων
Translated title:
Μachine learning on wireless sensor networks
Summary:
Well-timed fire detection in forests can protect human lives and valuable
forest areas, necessary for the maintenance of environmental equilibrium and
for the living of animals residing there. In the present thesis we examine the
implementation of a wireless sensor network which gathers temperature and
humidity data from the surrounding environment. The data then are fed to a
neural network which is implemented on one of the network’s nodes in order to
obtain a valid conclusion as to whether there is fire activity or not. The main
goal of this thesis is to investigate whether a neural network, supported by
Oracle’s SUNSpot platform development programming language (j2me), can be
implemented in order to provide valid and time effective information as to
whethere a forest fire has bursted out or not.
Keywords:
Data Fusion, Neural Networks, Sensor Networks, Moving Average