Technical Program

Paper Detail

Session:Signal Processing for Wireless Sensor Networks II
Time:Wednesday, May 19, 14:42 - 14:59
Presentation: Special Session Lecture
Topic: Special Sessions: Signal Processing for Wireless Sensor Networks
Authors: Ke Liu; University of Wisconsin-Madison 
 Akbar M. Sayeed; University of Wisconsin-Madison 
Abstract: We propose a novel decentralized detection schemebased on the method of types and type detectors. Contrary toconventional methods, the proposed decentralized type detection(DTD) scheme asymptotically achieves the performance of centralizeddetection, provided there exists an arbitrarily small (butnon-zero) communication capacity between the sensing nodes andthe decision center. The sensing nodes extract and transmit thetype information of the observed data sequence, which is thenrecovered and used by a type detector at the decision center.The key to DTD is that the required network information flow(type information) is of asymptotically vanishing entropy ratebut still sufficient for the hypothesis testing problem. The simplestructure of type sensors greatly reduces system cost in termsof computational and communication capabilities for sensornodes. In particular, the sensor nodes need not know the signalstatistics under different hypothesis. Theoretical analysis andnumerical simulations demonstrate the excellent performance ofthe proposed DTD scheme.

Home -||- Organizing Committee -||- Technical Committee -||- Technical Program -||- Plenaries
Paper Submission -||- Special Sessions -||- ITT -||- Paper Review -||- Exhibits -||- Tutorials
Information -||- Registration -||- Travel Insurance -||- Housing -||- Workshops

©2015 Conference Management Services, Inc. -||- email: -||- Last updated Wednesday, April 07, 2004