基于流量矩陣的多源數(shù)據(jù)融合分析的研究.pdf

1、哈爾濱工程大學碩士學位論文基于流量矩陣的多源數(shù)據(jù)融合分析的研究姓名：高悅申請學位級別：碩士專業(yè)：計算機軟件與理論指導教師：王慧強201203哈爾濱工程大學碩士學位論文AbstractNetworksituationreferstothenetworkstatusandcurrenttrendsoftheentirenetworkwhichiscomposedofthestatusofmultiplenetworkdevices，netw

2、orkbehaviors，anduserbehaviorsandSOonNetworkSecuritySituationalAwarenesstechnologyisamethodtomonitorthenetworkinrealtimeandmakeearlywarningsandevaluatesthesecuritystatustreadsofthenetworkinfutureNetworksysteminChinaisinth

3、ehistoricalstageoftransformationToimproveouremergencyresponsecapacity，reducetheharmcausedbynetworkattacksandimprovethesystem’SabilitytofightbackIthasimportantsignificancetodoresearchonNetworkSecuritySituationalAwarenesst

4、echnologyTherefore，howtoimprovenetworksituationalawareness，makedatamoreintuitive，moreobjfectiveofthecurrentnetworkstatusbecomesahottopicinresearchersNetworkdatafusionanalysistechniquesbornoutinsuchdemandsInthispaper，wema

5、keresearchontherelatedtechnologiesinvolvedinSNMPdataandNetFlowdataintegrationFirstly，byanalyzingthetwodifferentdata，thetheoryoftrafficmatrixdatafusionisproposedTheadvantagesandthereasonswhythetheorybeenchosenaredescribed

6、indetailstooSecondly，takingintoaccountoftheimpactofanumberoffactorstotrafficmatrixestimation，includingthecoverageofNetFlowdatacollectingdevicesandthechangesofrouting，animprovedtrafficmatrixdatafusionmethodisproposedThird

7、ly，multisourcedatafusionanalysissystemprototypeisproposedAndeverymoduleofthesystemisanalyzedindetailsFinally，theimprovedtrafficmatrixdatafusionmethodisverifiedfeasibilityandeffectivenessinourexperimentsExperimentsshowtha

8、ttheimprovedalgorithmbasedontrafficmatrix，canlocatenetworkanomaliesandnetworkcongestion，andfindthekeylinksofthenetworkquicklyandeffectivelyAdditionally，basedonthepreviousstudy，technologiessuchassocketprogrammingareusedin

9、implementingdatafusionBymonitoringdataflowintheviewoftrafficmatrix，wecouldgetcomprehensiveperspectiveofthenetworkstatus，whichisveryusefulfordataanalysisKeywords：Networksituationawareness；Datafusion；Flowmatrix；Anomalydete