The Colloquium Series of the
            Department of Computer Science,
		  University of Wyoming

			presents

                      Jim Partan
           University of Massachusetts-Amherst
                         and
           Woods Hole Oceanographic Institution

        "Underwater Sensor Networks and Whales"

                Thursday, January 25, 2007
                        ENG 4066
                     4:00 - 5:00 p.m.
   
Abstract:

I.) Underwater Sensor Networks

There is increasing interest in underwater sensor networks, in a range of
application areas: military (mine and submarine detection), commercial (seismic
monitoring of offshore oilfields), and environmental/scientific (long-term
ecological monitoring, pollution plume localization, seismic monitoring and
tsunami warning systems).

The underwater communication channel is challenging. Radio waves do not
propagate, leading to use of acoustics for wireless communication. Due to
frequency-dependent attenuation, there is a very limited bandwidth (typically
5kbit/sec with a 25kHz carrier at 2km in shallow water), significant
computational complexity to correct for the time-varying multipath channel, and
long propagation delays due to the speed of sound in water (1500m/s).

Terrestrial, radio-based sensor networks are often built from large numbers
of cheap, stationary sensor nodes, where battery energy is the primary
optimization metric. While some underwater sensor networks will be similar
to this, we believe that many underwater networks will be quite different,
built from a small number of relatively expensive, mobile sensor nodes, 
forming a mobile and often disconnected network. Rather than energy 
optimization, issues in this sort of mobile underwater network will be 
reliability, disruption-tolerant routing, fairness in channel access, 
routing over mixed media (e.g. underwater acoustic wireless and above-water
RF wireless links), and network-mobility interactions, such as collaborative
surveys of networked vehicles.

II.) Whale Tags and Behavioral Studies of Marine Mammals

Most whales spend the majority of their time submerged and unobservable.
Several years ago, I worked with engineers and biologists to help develop
and deploy a whale tag which would stick to a whale's back with suction cups,
while simultaneously recording sound and motion (pitch,roll,heading,depth) at
high resolutions. Light does not propagate well at depth underwater, whereas
sound propagates very well. Whales therefore use sound socially and for
foraging, and are impacted by human sounds such as ship noise, sonars, and
seismic oil surveys. This tag has allowed behavioral studies of many species
of whales, showing detailed use sound in foraging, and potential impacts of
human sounds on whale behavior.