面向城市交通網絡的時空索引研究.pdf

1、中國科學技術大學碩士學位論文面向城市交通網絡的時空索引研究姓名：張旭申請學位級別：碩士專業(yè)：計算機應用技術指導教師：金培權20090501ABSTRACTABSTRACTRecentlyconstraintnetworkbasedspatiotemporaldatamanagementhasbeenanewresearchtopicinspatio—temporaldatamanagement。Thespatio—temporalind

2、exfortheconstraintnetworkisoneofthekeyissuesinthisfieldFormostapplicationsinvolvingahugeamountofmovingobjects，suchasurbantrafficmanagement，buildingefficientspatio—temporalindexisthemostimportantwaytosuittheefficiencyrequ

3、irementsoftheapplicationsInthispaperwetakeurbantrafficnetworkasthebackground，summarizetheexistingnetworkdatamodel，providetherelevantdatastructuresandproposetheimproveddatamodelofurbantrafficnetworkAtthesametime，wepresent

4、anovelspatiotemporalindexforconstraintnetwork，whichiScalledNBR—tree(Network—BasedRtree)，andwefocusonthebackgroundoftheapplicationsinallurbantrafficnetworkTheNBRtreeisanimprovementonthepreviousindexnamedMON—tree，withanana

5、lysisonthespecificpropertiesofthemovingobjectsinanurbantrafficnetworkFinallywecomparetheperformanceoftheNBR—treeandtheMON—treeindexthroughtheexperimentThemaincontributionofthispapercanbesummedupasfollows：(1)Wesummarizeth

6、eexistingnetworkdatamodel，improveandconsummatetheGcitingroadnetworkdatamodel，providetheurbantrafficnetworkandthecompletedatastructuresofobject(includingmovingobjectandstaticobject)；(2)Wepresentanovelspatiotemporalindexfo

7、rconstraintnetwork：NBRtreeTheNBR—treeisanimprovementonthepreviousindexnamedMONtree，withananalysisonthespecificpropertiesofthemovingobjectsinanurbantrafficnetworkWediscusstheindexstructure，operatingalgorithmsaswellasquery

8、algorithmsoftheNBRtreeindeiail；(3)ThecomparedexperimentsarecarriedonthedatasetfromBrinkhoffsmovingobjectgeneratorTheexperimentalresultsshowthatourproposedindexisabletosupportNNqueries，windowqueriesandqueriesgivenstartpos