High Accuracy Distributed Target Detection and Classification in Sensor Networks Based on ...

206 Pages·2016·1.92 MB·English

by  Xiaoling Wang

UUnniivveerrssiittyy ooff TTeennnneesssseeee,, KKnnooxxvviillllee
TTRRAACCEE:: TTeennnneesssseeee RReesseeaarrcchh aanndd CCrreeaattiivvee
EExxcchhaannggee
Doctoral Dissertations Graduate School
12-2004
HHiigghh AAccccuurraaccyy DDiissttrriibbuutteedd TTaarrggeett DDeetteeccttiioonn aanndd CCllaassssiifificcaattiioonn iinn
SSeennssoorr NNeettwwoorrkkss BBaasseedd oonn MMoobbiillee AAggeenntt FFrraammeewwoorrkk
Xiaoling Wang
University of Tennessee - Knoxville
Follow this and additional works at: https://trace.tennessee.edu/utk_graddiss
Part of the Electrical and Computer Engineering Commons
RReeccoommmmeennddeedd CCiittaattiioonn
Wang, Xiaoling, "High Accuracy Distributed Target Detection and Classification in Sensor Networks Based
on Mobile Agent Framework. " PhD diss., University of Tennessee, 2004.
https://trace.tennessee.edu/utk_graddiss/2253
This Dissertation is brought to you for free and open access by the Graduate School at TRACE: Tennessee
Research and Creative Exchange. It has been accepted for inclusion in Doctoral Dissertations by an authorized
administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact
[email protected].
To the Graduate Council:
I am submitting herewith a dissertation written by Xiaoling Wang entitled "High Accuracy
Distributed Target Detection and Classification in Sensor Networks Based on Mobile Agent
Framework." I have examined the final electronic copy of this dissertation for form and content
and recommend that it be accepted in partial fulfillment of the requirements for the degree of
Doctor of Philosophy, with a major in Electrical Engineering.
Hairong Qi, Major Professor
We have read this dissertation and recommend its acceptance:
Mongi A. Abidi, Michael J. Roberts. Daniel B. Koch, Hamparsum Bozdogan
Accepted for the Council:
Carolyn R. Hodges
Vice Provost and Dean of the Graduate School
(Original signatures are on file with official student records.)
TotheGraduateCouncil:
I am submitting herewith a dissertation written by Xiaoling Wang entitled “High Accuracy
Distributed Target Detection and Classification in Sensor Networks Based on Mobile Agent
Framework.” Ihaveexaminedthefinalelectroniccopyofthisdissertationforformandcontent
and recommend that it be accepted in partial fulfillment of the requirements for the degree of
DoctorofPhilosophy, withamajorinElectricalEngineering.
HairongQi
MajorProfessor
Wehavereadthisdissertation
andrecommenditsacceptance:
MongiA.Abidi
MichaelJ.Roberts
DanielB.Koch
HamparsumBozdogan
AcceptedfortheCouncil:
AnneMayhew
ViceChancellorandDeanof
GraduateStudies
(Originalsignaturesareonfilewithofficialstudentrecords.)
High Accuracy Distributed Target Detection
and Classification in Sensor Networks Based
on Mobile Agent Framework
A Dissertation
Presented for the
Doctor of Philosophy
Degree
The University of Tennessee, Knoxville
Xiaoling Wang
December, 2004
Copyright (cid:0)c2004byXiaolingWang
Allrightsreserved.
ii
Acknowledgments
Iwouldliketoextendmysinceregratitudeandappreciationtoalltheindividualswhomade
thisdissertationpossible.
Firstandforemost,I amgratefulto myadviserDr. HairongQi. Herguidanceandencour-
agementthrough mystudieshas allowed meto develop myskills as a researcher. Without her
support,thisworkwouldnothavebeenpossible. AlsoIamhighlyindebtedtoDr. Hamparsum
Bozdogan,Dr. MongiAbidi,Dr. MichaelRoberts,andDr. DanielKochforservingasmembers
ofmycommitteeandprovidingalltheenlighteningsuggestions.
Special acknowledgment is given to Mr. Steve Beck, Mr. Joe Reynolds, and Ms. Carol
BreweratBAESystems,Inc. fortheirelaborateworkhelpingussetupthefielddemoatAustin,
TXandcollectthevaluableexperimentaldata. IalsoappreciatetheworkbyProfessorYuHen
HuandhisstudentsatUniversityofWisconsin-Madisonforgeneratingthecross-validationdata
setbasedontheSITEX02fielddemo.
Last but not the least, I sincerely thank my husband Hongtao, my parents, my parents-in-
law,mybrotherandsisterfortheirunconditionalsupport,love,andencouragementthroughmy
life.
iii
Publication History
Thisdissertationappearsinpartinthefollowingacademicjournalsandconferences.
X.Wang,H.Qi,S.Beck(2004). “Distributedmulti-targetdetectioninsensornetworks”,
(cid:0)
FrontiersinDistributedSensorNetworks. Editors: R.Brooks,S.S.Iyengar, CRCPress.
X.Wang,H.Qi,S.Beck,H.Du(2004). “Aprogressive approachtodistributed multiple
(cid:0)
targetdetectioninsensornetworks”,SensorNetworkOperations. Editor: S.Phoha,IEEE
Press.
X.Wang,H.Qi,Y.Tian,andS.S.Iyengar(2004). “Collaborativesignalandinformation
(cid:0)
processinghierarchyindistributed sensornetworks”,SubmittedtoInternationalJournal
onDistributedSensorNetwork.
H.Qi, Y. Xu,andX.Wang(2003). “Mobile-agent-basedcollaborative signalandinfor-
(cid:0)
mationprocessinginsensornetworks”,ProceedingsofIEEE,91(8): 1172-1183,August.
H. Qi, X. Wang, S. S. Iyengar, and K. Chakrabarty (2002). “High performance sensor
(cid:0)
integrationindistributed sensornetworksusingmobileagents,” InternationalJournalof
HighPerformanceComputingApplications, 16(3): 325-336,August.
X.Wang,H.Qi(2004). “Collaborative unknowntargetrecognitioninsensornetworks”,
(cid:0)
TheNineteenthNationalConference onArtificialIntelligence, WorkshoponSensorNet-
works,SanJose,CA,July.
X. Wang, H. Qi (2004). “Mobile agent based progressive multiple target detection in
(cid:0)
sensornetworks”,InternationalConferenceonAcoustics,Speech,andSignalProcessing,
Montreal,Canada,May.
X. Wang, H. Qi, H. Du (2003). “Distributed source number estimation for multiple
(cid:0)
target detection in sensor networks”, IEEE Workshop on Statistical Signal Processing,
iv
St. Louis,MO,September.
X.Wang,H.Qi,andS.S.Iyengar(2002). “Collaborative multi-modalitytargetclassifi-
(cid:0)
cationin distributed sensornetworks,” InternationalConference onInformationFusion,
pp.285-290,Annapolis,MA,July.
X. Wang, and H. Qi (2002). “Acoustic target classification using distributed sensor ar-
(cid:0)
rays,”, International Conference on Acoustics, Speech and Signal Processing, Orlando,
FL,May.
Y. Tian, H. Qi, and X. Wang (2002). “Target detection and classification using seismic
(cid:0)
signal processing in unattended ground sensor systems,” International Conference on
Acoustics,SpeechandSignalProcessing, Orlando,FL,May.
H. Qi, X. Wang, S. S. Iyengar, and K. Chakrabarty (2001). “Multisensor data fusion
(cid:0)
in distributed sensor networks using mobile agents,” Information Fusion, TuC2-11-16,
Canada,August.
v
Abstract
High-accuracy distributed information exploitation plays an important role in sensor net-
works. This dissertation describes a mobile-agent-based framework for target detection and
classificationinsensornetworks. Specifically, wetacklethechallengingproblemsofmultiple-
targetdetection,high-fidelitytargetclassification,andunknown-target identification.
Inthisdissertation,wepresentaprogressivemultiple-target detectionapproachtoestimate
thenumberoftargetssequentiallyandimplementitusingamobile-agentframework. Tofurther
improvetheperformance,wepresentacluster-based distributed approachwheretheestimated
resultsfromdifferentclustersarefused. Experimentalresultsshow thatthedistributed scheme
withtheBayesianfusionmethodhavebetterperformanceinthesensethattheyhavethehighest
detectionprobabilityandthemoststableperformance. Inaddition,theprogressiveintra-cluster
(cid:0)(cid:2)(cid:1)(cid:4)(cid:3)(cid:6)(cid:5)(cid:2)(cid:5)(cid:8)(cid:7) (cid:0)(cid:10)(cid:9)(cid:11)(cid:3)(cid:6)(cid:12)(cid:14)(cid:13)(cid:15)(cid:7)
estimationcanreducedatatransmissionby andconserve energy by compared
tothecentralizedscheme.
Forcollaborative target classification,wedevelop ageneralpurposemulti-modality, multi-
sensor fusion hierarchyfor informationintegration in sensor networks. The hierarchyis com-
posed of four levels of enabling algorithms: local signal processing, temporal fusion, multi-
modality fusion, and multi-sensor fusion using a mobile-agent-based framework. The fusion
hierarchy ensures fault tolerance and thus generates robust results. In the meanwhile, it also
takesintoaccountenergy efficiency. Experimentalresultsbasedontwofielddemosshowcon-
stantimprovementofclassificationaccuracy overdifferentlevelsofthehierarchy.
Unknowntargetidentificationinsensornetworkscorrespondstothecapabilityofdetecting
targets without any a priori information, and of modifying the knowledge base dynamically.
In this dissertation, we present a collaborative method to solve this problem among multiple
(cid:0)(cid:11)(cid:16)(cid:17)(cid:7)
sensors. When applied to the military vehicles data set collected in a field demo, about
unknown target samples can be recognized correctly, while the known target classification ac-
(cid:7)
curacystaysabove (cid:18)(cid:2)(cid:19) .
vi
Contents
1 Introduction 1
1.1 SensorNetworks-StateoftheArt . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 PotentialsofSensorNetworks . . . . . . . . . . . . . . . . . . . . . . 3
1.1.2 ChallengesofSensorNetworks . . . . . . . . . . . . . . . . . . . . . 4
1.1.3 ApplicationsofSensorNetworks . . . . . . . . . . . . . . . . . . . . 6
1.1.4 SensorNetworksinAction . . . . . . . . . . . . . . . . . . . . . . . . 9
1.2 SensorNodeArchitecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.3 ProtocolStackofSensorNetworks . . . . . . . . . . . . . . . . . . . . . . . . 15
1.3.1 PhysicalLayer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.3.2 DataLinkLayer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.3.3 NetworkLayer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.3.4 TransportLayer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.3.5 ApplicationLayer . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.3.6 PowerManagement,MobilityManagement,andTaskManagement . . 23
1.4 ComputingParadigmsofSensorNetworks . . . . . . . . . . . . . . . . . . . . 24
1.4.1 CentralizedClient/ServerModel . . . . . . . . . . . . . . . . . . . . . 24
1.4.2 DecentralizedPeer-to-PeerModel . . . . . . . . . . . . . . . . . . . . 25
1.4.3 Mobile-Agent-BasedComputingParadigm . . . . . . . . . . . . . . . 27
vii