STRATEGIES FOR SENSOR SELECTION IN NETWORKS MONITORING CHEMICAL DIFFUSION SCENARIOS

Nikhil Padhye, Chilukuri K. Mohan, Kishan G. Mehrotra, Pramod K. Varshney

Research output: Chapter in Book/Entry/PoemConference contribution

Abstract

This paper addresses the task of tracking the diffusion of a toxic chemical using distributed sensor networks. Assuming a Gaussian-puff model for chemical release and diffusion, the tracking problem is formulated as a parameter estimation task, carried out by minimizing mean squared error (M.S.E), computed from training samples collected by the sensors and predicted concentrations based on estimated parameters. A hybrid evolutionary algorithm utilizing a Particle Swarm Optimizer (PSO) coupled with Sequential Quadratic Programming (SQP), is proposed to minimize MSE. We also propose and evaluate several sensor selection strategies to query informative sensors. Finally, we address parameter estimation under time varying wind components.

Original languageEnglish (US)
Title of host publication1st International Conference on Sensor Networks and Applications 2009, SNA 2009
EditorsGordon K. Lee
PublisherInternational Society for Computers and Their Applications (ISCA)
Pages7-12
Number of pages6
ISBN (Electronic)9781615676675
StatePublished - 2009
Event1st International Conference on Sensor Networks and Applications, SNA 2009 - San Francisco, United States
Duration: Nov 4 2009Nov 6 2009

Publication series

Name1st International Conference on Sensor Networks and Applications 2009, SNA 2009

Conference

Conference1st International Conference on Sensor Networks and Applications, SNA 2009
Country/TerritoryUnited States
CitySan Francisco
Period11/4/0911/6/09

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Hardware and Architecture
  • Software

Fingerprint

Dive into the research topics of 'STRATEGIES FOR SENSOR SELECTION IN NETWORKS MONITORING CHEMICAL DIFFUSION SCENARIOS'. Together they form a unique fingerprint.

Cite this