|
Keynote Presentation
|
"Early Warning Systems for Natural Disasters: Orphan
nodes and Hopping Sensors"
By Prof. Saman K. Halgamuge, University of Melbourne,
Australia |
 |
Bio
Dr. Halgamuge is a Professor in the Melbourne School of
Engineering of The University of Melbourne and a member of
the school wide initiative of Biomedical Engineering. He
received Dipl.‐Ing (1990) and Dr.‐Ing (1995) degrees in
Electrical Engineering from Technical University of Darmstadt
and B.Sc. Engineering degree from University of
Moratuwa (1984).
He leads a group of postdoctoral fellows and PhD students
working on Pattern Recognition and Optimization.
Dr. Halgamuge is the co‐author of over 60 journal papers, 150
conference papers (h‐index 21), 8 books and 15
book chapters.
Abstract
Propagation of a natural disaster across a landscape is
stochastic in nature. However, the pattern may have an
underlying structure in a given geographical primitive
(alongside the river banks in the case of a flood, driven by
the
wind patterns in the case of a forest fire, etc). We discuss
about models that combines prior knowledge with
in-coming stochastic data to predict expected times of failure
at various locations to optimally prepare sensor nodes
to capture vital environmental events with limited on-board
power, a hopping mechanism to enable sensor nodes to be
mobile to achieve maximum resolution of sensing, and a new
method of orphan node management enabling the sensor
network to be robust in the presence of random malfunctioning
of sensor nodes. We will also present some of the
recent sensor network deployments.
Back
|
