SmartShadow: Models and Methods for Pervasive Computing

139 Pages·2013·4.578 MB·English

by  Zhaohui Wu, Gang Pan (auth.)

Advanced Topics in Science and Technology in China
Zhaohui Wu
Gang Pan
SmartShadow:
Models and Methods
for Pervasive
Computing
123
ADVANCED TOPICS
IN SCIENCE AND TECHNOLOGY IN CHINA
ADVANCED TOPICS
IN SCIENCE AND TECHNOLOGY IN CHINA
ZhejiangUniversityisoneoftheleadinguniversitiesinChina.InAdvancedTopics
inScienceandTechnologyinChina,ZhejiangUniversityPressandSpringerjointly
publishmonographsbyChinesescholarsandprofessors,aswellasinvitedauthors
and editors from abroad who are outstanding experts and scholars in their fields.
Thisserieswillbeofinteresttoresearchers,lecturers,andgraduatestudentsalike.
AdvancedTopicsinScienceandTechnologyinChinaaimstopresentthelatestand
mostcutting-edgetheories,techniques,andmethodologiesinvariousresearchareas
in China. It covers all disciplines in the fields of natural science and technology,
including but not limited to, computer science, materials science, life sciences,
engineering,environmentalsciences,mathematics,andphysics.
Forfurthervolumes:
http://www.springer.com/series/7887
Zhaohui Wu • Gang Pan
SmartShadow: Models
and Methods for Pervasive
Computing
123
ZhaohuiWu
GangPan
DepartmentofComputerScience
ZhejiangUniversity
Hangzhou,China,People’sRepublic
ISSN1995-6819 ISSN1995-6827(electronic)
ISBN978-3-642-36381-8 ISBN978-3-642-36382-5(eBook)
DOI10.1007/978-3-642-36382-5
SpringerHeidelbergNewYorkDordrechtLondon
JointlypublishedwithZhejiangUniversityPress,Hangzhou
ISBN:978-7-308-09799-4ZhejiangUniversityPress,Hangzhou
LibraryofCongressControlNumber:2013939580
©ZhejiangUniversityPress,HangzhouandSpringer-VerlagBerlinHeidelberg2013
Thisworkissubjecttocopyright.AllrightsarereservedbythePublishers,whetherthewholeorpartof
thematerialisconcerned,specificallytherightsoftranslation,reprinting,reuseofillustrations,recitation,
broadcasting,reproductiononmicrofilmsorinanyotherphysicalway,andtransmissionorinformation
storageandretrieval,electronicadaptation,computersoftware,orbysimilarordissimilarmethodology
nowknownorhereafterdeveloped.Exemptedfromthislegalreservationarebriefexcerptsinconnection
with reviews or scholarly analysis or material supplied specifically for the purpose of being entered
and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of
this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publisher’s locations, in its current version, and permission for use must always be obtained from
Springer.PermissionsforusemaybeobtainedthroughRightsLinkattheCopyrightClearanceCenter.
ViolationsareliabletoprosecutionundertherespectiveCopyrightLaw.
Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublication
doesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevant
protectivelawsandregulationsandthereforefreeforgeneraluse.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors northe editors northe publishers can accept anylegal responsibility
foranyerrorsoromissionsthatmaybemade.Thepublishersmakenowarranty,expressorimplied,with
respecttothematerialcontainedherein.
Printedonacid-freepaper
SpringerispartofSpringerScience+BusinessMedia(www.springer.com)
Preface
Pervasive computing is becoming an emerging paradigm of next-generation
computing. It seeks to transform computer-centric computing to human-centered
computing.Itsgoalistobuildapersonallivingenvironmentwheredigitalservices
can be provided seamlessly at the right time at the right place in the right way.
Pervasive computing is an innovation of information and computing technology,
which leads to revolutionin the relationshipbetween the physicalenvironmentof
humanlifeandthecyberspace.Itwillgreatlychangeourfuturewayoflife.
ThisbookcompilessomerecentworkfromthePervasiveComputingGroupof
the CCNT (advanCedComputingaNd sysTem) laboratoryat ZhejiangUniversity,
China.Pervasivecomputing,asanewcomputingmodel,spansanumberofresearch
areas with unprecedented complexity and diversity. This book describes a new
model for pervasive computing, called SmartShadow, and its methods. A good
modelisnotonlyessentialtoabstractlymodel“user-oriented”environmentswhich
is the core philosophy of pervasive computing, but which is also important in
design,analysis,implementation,deployment,assessment ofpervasivecomputing
systems to support such systems with a wide range of theoretical guidance for
thescalability,maintainability,adaptability,easeofuse, andstandardizationatthe
model-level.Theorganizationofthebookisasfollows.Chapter1givesanoverview
of the SmartShadow. It consists of a user model which represents information
aboutofauser’spersonality(calledPersonalityShadow)andaspacemodelwhich
representsavirtualsmartspaceofservicesfollowingauser(calledServiceShadow).
Chapter2describeshowausertaskintheSmartShadowcanmigratecontinuously
and seamlessly between different physical spaces when users move from one to
another. Context-awarenessis strongly related to how smart a system is, which is
presentedinChap.3.Weproposeathree-layercontextmodelintheSmartShadow
tohandlelarge-scalecontextualdata.InChap.4,wedescribeanewfilemanagement
scheme to link files with contextual data in SmartShadow, so that we can have a
human-orientedview of file dynamics. Chapter 5 builds a software infrastructure
withsemanticsandhighadaptationtosupporttheSmartShadow.Chapter6presents
asmartcarprototypetodemonstrateanapplicationoftheSmartShadow.
v
vi Preface
Thisbookwouldnothavebeenpossiblewithoutmanycontributorswhosenames
did not make it to the cover. We would like to give our special thanks to Mr. Li
Zhang,Mr.YuqiongXu,Dr.QingWu,Dr.Jie Sun,Dr. YanfeiLiu,Dr.ShijianLi,
Mrs.TongLi,andMr.QunjieQiu,whohavebeenaffiliatedtotheCCNT(advanCed
ComputingaNdsysTem)laboratoryofZhejiangUniversity.Foralongtimealready,
ithasbeenourpleasuretodoresearchwiththeminpervasivecomputing.Theyhave
devotedtheirenergyandenthusiasmtothisareaandrelevantresearchprojects.
The work in this book was mainly supported by the National Science Fund
for Distinguished Young Scholars (No. 60525202), the Natural Science Founda-
tion of China (No. 61070067, No. 60533040), the National Key Basic Research
Program of China (No. 2013CB329504), the National 863 High-Tech Program
(No.2009AA011900,No.2008AA01Z132,No.2006AA01Z198),theProgramfor
NewCenturyExcellentTalentsinUniversity(NCET-04-0545),ZhejiangProvincial
NaturalScienceFoundation(No.RC01058,No.Y1090690),andQianjiangTalent
ProgramofZhejiang(2011R10078).
Hangzhou,China ZhaohuiWu
October2012 GangPan
Contents
1 SmartShadowModel ........................................................ 1
1.1 Introduction ............................................................. 1
1.2 SmartShadow:AnOverview ........................................... 3
1.2.1 PersonalityShadow:ModelofPeople’sLife .................. 3
1.2.2 ServiceShadow:User-CenteredVirtualSpace ................ 4
1.3 SemanticsinSmartShadow............................................. 5
1.3.1 BasicConcepts.................................................. 5
1.3.2 ModelsandSystemsinPervasiveComputing................. 7
1.3.3 SemanticIssuesinSmartShadow.............................. 8
1.4 ModelingHumanPersonalities......................................... 10
1.4.1 Low-LevelPersonalInformation............................... 11
1.4.2 Mid-LevelPersonalInformation............................... 13
1.4.3 High-LevelPersonalInformation:BDPModel............... 15
1.5 ModelingComputingEnvironments................................... 17
1.6 PlanningforaVirtualServiceSpace:ServiceShadow................ 20
1.7 Conclusion .............................................................. 22
References..................................................................... 22
2 TaskMigrationinSmartShadow........................................... 25
2.1 Introduction ............................................................. 25
2.2 RelatedWork............................................................ 27
2.3 TaskMigrationChallenges............................................. 28
2.3.1 Scenario:WatchingMovie ..................................... 28
2.4 TaskModelinSmartShadow........................................... 29
2.4.1 ModelingTaskProcess......................................... 29
2.4.2 DescribingaPrimitiveTask.................................... 31
2.5 Context-AwareTask-to-ServiceMapping.............................. 32
2.5.1 SemanticServiceFiltering ..................................... 33
2.5.2 SatisfactionMeasurementforTask-OrientedMatching ...... 35
2.5.3 Task-to-ServiceMappingAlgorithm .......................... 37
2.6 TaskShadow:AnOSGi-BasedImplementation ....................... 37
vii
viii Contents
2.7 Evaluation............................................................... 39
2.7.1 PerformanceofTask-to-ServiceMapping..................... 40
2.7.2 TimeCostforTaskMigrationof“WatchMovie”............. 40
2.8 Conclusion .............................................................. 41
References..................................................................... 41
3 Context-AwarenessinSmartShadow...................................... 43
3.1 Introduction ............................................................. 43
3.2 HierarchicalContextModel............................................ 45
3.2.1 ContextAtomLayer............................................ 46
3.2.2 ContextSituationLayer ........................................ 46
3.3 ScudContext:Large-ScaleEnvironmentalContextServices.......... 51
3.3.1 ContextManagement........................................... 51
3.3.2 TheScudContextInfrastructure................................ 55
3.4 ScudCORE:Context-DrivenReasoningEngine....................... 59
3.4.1 Context-DrivenReasoninginPervasiveComputing.......... 59
3.4.2 ScudCOREArchitecture ....................................... 60
3.4.3 Context-DrivenMechanism.................................... 62
3.5 Conclusion .............................................................. 65
References..................................................................... 66
4 FileModelinginSmartShadow............................................. 67
4.1 Introduction ............................................................. 67
4.2 RelatedWork............................................................ 68
4.3 ModelingFileswithContextStreams.................................. 70
4.3.1 ContextStream:DynamicsofContexts ....................... 70
4.3.2 ModelingFileswithContextStream .......................... 71
4.4 ScudFile:AMetaFileSystemwithContextStreams................. 72
4.4.1 ArchitectureoftheScudFile ................................... 72
4.4.2 AnMS-Windows-BasedImplementation ..................... 74
4.5 Applications............................................................. 76
4.5.1 FileBrowsingwithContextStream............................ 76
4.5.2 FileSearchingwithContextStream........................... 77
4.5.3 FileRecommendingwithContextStream..................... 78
4.6 Evaluation............................................................... 79
4.6.1 TimeCost ....................................................... 79
4.6.2 UsabilityTest ................................................... 80
4.7 Conclusion .............................................................. 81
References..................................................................... 81
5 ScudWare:SoftwareInfrastructureforSmartShadow.................. 83
5.1 Introduction ............................................................. 83
5.2 ScudWareArchitecture ................................................. 85
5.3 SemanticVirtualAgents................................................ 86
5.3.1 SVADefinition.................................................. 86
5.3.2 SIPDefinition................................................... 88
Contents ix
5.4 SemanticContextManagementService ............................... 90
5.4.1 SemanticViewforContext..................................... 90
5.4.2 OntologyofSmartShadowSpace.............................. 91
5.4.3 SemanticContextServiceFunctionalModules .............. 93
5.5 AdaptiveComponentManagementService ........................... 95
5.5.1 ResourceAbstract .............................................. 95
5.5.2 ACMSFunctionality............................................ 96
5.5.3 ComponentInterdependenceGraph ........................... 97
5.5.4 Run-TimeComponentHookProxy............................ 98
5.6 Conclusion .............................................................. 99
References..................................................................... 99
6 SmartCarSpace:AnApplication ......................................... 101
6.1 Introduction ............................................................. 101
6.2 SmartCarSpace:AFramework........................................ 102
6.3 ContextModelinginSmartCarSpace................................. 104
6.3.1 ContextAtoms ................................................. 104
6.3.2 ComplexContext ............................................... 106
6.4 DriverBehaviorModelinginSmartCarSpace........................ 108
6.4.1 ComfortableandSafeDriverBehaviorModel
ofCar-Following................................................ 108
6.4.2 MultitaskingDriverBehaviorModelforSmart
CarSpace........................................................ 112
6.5 Prototype ................................................................ 119
6.5.1 NetworkInfrastructure ......................................... 119
6.5.2 SoftwareInfrastructure......................................... 120
6.5.3 In-CarSpace .................................................... 120
6.5.4 ServicesofSmartCarSpace................................... 123
6.6 Conclusion .............................................................. 125
References..................................................................... 125
Index............................................................................... 129