|
Activity Number:
|
132
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #305063 |
|
Title:
|
Parameter Updating Suppression Scheme for Wireless Sensor Networks
|
|
Author(s):
|
Kristian Lum*+
|
|
Companies:
|
Duke University
|
|
Address:
|
Department of Statistical Science, Durham, NC, 27708,
|
|
Keywords:
|
missing data ; sensor networks ; environmental statistics
|
|
Abstract:
|
One of the major challenges in wireless sensor networks is designing a transmission rule to maximize the lifetime of the network and the precision of inference. I will present a suppression scheme that uses very little memory and computing power (thus allowing implementation on very simple sensor nodes) to dynamically update process parameters to decide which readings to transmit and bound those that are not. I will demonstrate this method for a hierarchical model that describes how light from the Duke forest filters through the canopy and on a stochastic differential equation model with simulated data.
|
