Speaker: Radhika Vind
Day:  Wednesday, 10/03/2007
Room: ITEB 336
Time: 2:00-3:00pm


 
Title: Model-Driven Data Acquisition in Sensor Networks
 
Abstract: Wireless sensor networks consist of spatially distributed
autonomous devices using sensor to cooperatively monitor and report
physical or environmental conditions such as temperatures, humidity or
light. Each node in a sensor network is typically equipped with a radio
transceiver or other wireless communications device, a small
microcontroller, and an energy source, usually a battery. Since the size of
the cell is limited, so is amount of available energy. Hence most
applications, architectures and deployment strategies are developed with
low energy consumption in consideration. A typical application of sensor
network is habitat monitoring and results from queries are stored in a
database and computations are performed on retrieved data. When no data is
available, the nodes are queried to gather more information and this
increases energy consumption of the device. In this talk, I will present an
existing  statistical model that uses probabilistic models making use of
the correlation between nodes temporally and spatially is introduced. The
developed model also makes use of inter attribute correlation, leading to
high performance gains and much lower energy consumption.

Reference:
"Model Driven Data Acquisition in Sensor Networks" A.Desphande, C Guestrin,
S Madden, J Hellerstein, W. Hong