基于NetFlow流特征證據(jù)推理的主干網(wǎng)異常檢測方法研究.pdf

1、東南大學(xué)碩士學(xué)位論文基于Flow流特征證據(jù)推理的主干網(wǎng)異常檢測方法研究姓名：邢蘇霄申請學(xué)位級別：碩士專業(yè)：計算機科學(xué)與技術(shù)；計算機系統(tǒng)結(jié)構(gòu)指導(dǎo)教師：龔儉20110701AbstractResea比hofNetworkAnomolyDetectiononBackboneNetworkBasedonNetFlowFeatureAnalysisandD—SEvidenceTheoI‘yXINGSuxia0，GoNGJi鋤SchoolofCom

2、puterSience鋤dEngine硎ng，Southe鵲tUnivers時Inrecentyears，networksecuri夠eVentSoccurf把quently鋤dthethreatscopesbecomela唱er卸dla曙erw池ther印iddeVelopmentoftlleIntemetM卸yla唱ea仳acl(Ssuch笛DDoS，sc柚鋤d、vornlhaVese—ouslya僳：ctedpeopIe’snom

3、alworl(，鯽udyandresearchButifwemonitorthenet、Ⅳorkt州麗c∞thehighspe酣backboneneMorkinrealtime，wec鋤notonlydetectthe鋤omalytra肺c鋤djudgci乜夠pe，butalsotal∞p嘲a肚io璐intimet0stopla彈甜ack’sm脅erpropaga：cion鋤dbmadcastNowadaysthereexistsome

4、issuesinthefieIdofne陽，ork鋤omalydetection，forex踟ple’highfalseal姍rale，limitcdtypesofatt∞kstheappr0∞hesc鋤detect鋤dlowdetectionefficiency∞highsp∞d鵬t、vorkin砌time，∞theya他notappliedwidelyinp嘲iceThispaper駝lectsNetFlow弱thedatasour

5、∞ofnetwork觚0m“yd酏ectionaimingattheb∞kbonenetworkenVir0姍entwithhighspeed鋤dl鶘eVol哪etIa幣c，鋤dthenpropo駝sanoVel柚omalydetectionapproachonb∞kbonene“v0一(Thisnewde坂炯rc鋤eaectiVelym∞multiplelightcomputationNetFlowfea：tu他sofnetworkt

6、黼icinordert0achieVelowfaI∞al鋤r曩【e鋤dmissing眥，卸datthes鋤etime，satis匆thedetectiondem鋤dofhighspeedbackb∞enetworkInadditio兒weincorporateh豁hdatastmctu代and∞10adaptalionmechanismt0yieldhighaccur∞y0fdetectionindyn鋤icn咖0rkAtfir鞏thi

7、sp印erela：bomtesthe∞一ousconditionofnetworksecurityatfirst，pointSouttheu唱ency鋤dneceSsityof鋤omalydetect伽陀se甜chonb∞kbonen娟vo心Furthemo佗，Iightcomputationfea：tu他sofnet、Ⅳorktm歷careselectedbycl雒si毋ing卸d鋤alyzingthef鋤iIi盯繃omalynet、

8、Ⅳork仃affic帥es徹b∞kbonenet、ⅣorkSe∞ndlyitin刪ucesDSeViden∞n喀iontheo∥鋤dappli囂itintotIledesignof鋤omalydetectionmodelB部edonthemodel，theanomalydete“0nsystemis他ali∞d傭NBOS(Net、vorkBehaViorObservationSystem)This鯽omalydetectionsyste

9、mh笛thr∞modulesina11Theya托flowrecordp陀仃e柵entmodule，floweviden∞generalionmodule鋤dD—SevidencefusionmoduIeT11esy啦m’sinterfkesandd砒出笛ea佗als0in觚’ducedAtl笛t，脅ctionverification鋤dperf0冊撲cetcsta佗presented鋤dtheresuItSindicalethatth