外文翻译adhoc网络的一个简要概述挑战与研究方向教学内容.docx

外文翻译adhoc网络的一个简要概述挑战与研究方向教学内容.docx

《外文翻译adhoc网络的一个简要概述挑战与研究方向教学内容.docx》由会员分享，可在线阅读，更多相关《外文翻译adhoc网络的一个简要概述挑战与研究方向教学内容.docx（12页珍藏版）》请在冰豆网上搜索。

外文翻译adhoc网络的一个简要概述挑战与研究方向教学内容

附录

一、英文原文

ABriefOverviewofadhocNetworks:

ChallengesandDirections

Oneofthemostvibrantandactive“new”fieldstodayisthatofadhocnetworks.Significantresearchinthisareahasbeenongoingfornearly30years,alsounderthenamespacketradioormulti-hopnetworks.

adhocnetworkisa（possiblymobile）collectionofcommunicationsdevices（nodes）thatwishtocommunicate,buthavenofixedinfrastructureavailable,andhavenopre-determinedorganizationofavailablelinks.Individualnodesareresponsiblefordynamicallydiscoveringwhichothernodestheycandirectlycommunicatewith.

Adhocnetworkingisamulti-layerproblem.Thephysicallayermustadapttorapidchangesinlinkcharacteristics.Themultipleaccesscontrol（MAC）layerneedstominimizecollisions,allowfairaccess,andsemi-reliablytransportdataoverthesharedwirelesslinksinthepresenceofrapidchangesandhiddenorexposedterminals.Thenetworklayerneedstodetermineanddistributeinformationusedtocalculatepathsinawaythatmaintainsefficiencywhenlinkschangeoftenandbandwidthisatapremium.It’salsoneedstointegratesmoothlywithtraditional,nonadhoc-awareinternetworksandperformfunctionssuchasauto-configurationinthischangingenvironment.Thetransportlayermustbeabletohandledelayandpacketlossstatisticsthatareverydifferentthanwirednetworks.Finally,applicationsneedtobedesignedtohandlefrequentdisconnectionandreconnectionwithpeerapplicationsaswellaswidelyvaryingdelayandpacketlosscharacteristics.

Adhocnetworksaresuitedforuseinsituationswhereinfrastructureiseithernotavailable,nottrusted,orshouldnotbereliedonintimesofemergency.Afewexamplesinclude:

militarysoldersinthefield;sensorsscatteredthroughoutacityforbiologicaldetection;aninfrastructurelessnetworkofnotebookcomputersinaconferenceorcampussetting;theforestryorlumberindustry;rareanimaltracking;spaceexploration;underseaoperations;andtemporaryofficessuchascampaignheadquarters.

History

Thehistoryofadhocnetworkscanbetracedbackto1972andtheDoD-sponsoredPacketRadioNetwork（PRNET）,whichevolvedintotheSurvivableAdaptiveRadioNetworks（SURAN）programintheearly1980s[l].Thegoaloftheseprogramswastoprovidepacketswitchednetworkingtomobilebattlefieldelementsinaninfrastructureless,hostileenvironment（soldiers,tanks,aircraft,etc.,formingthenodesinthenetwork）.

Intheearly1990saspateofnewdevelopmentssignaledanewphaseinadhocnetworking.Notebookcomputersbecamepopular,asdidopen-sourcesoftware,andviablecommunicationsequipmentbasedonRFandinfrared.Theideaofaninfrstructurelesscollectionofmobilehostswasproposedintwoconferencepapers[2,3],andtheIEEE802.11subcommitteeadoptedtheterm“adhocnetworks.”Theconceptofcommercial（non-military）adhocnetworkinghadarrived.Othernovelnon-militarypossibilitiesweresuggested（asmentionedintheintroduction）,andinterestgrew.

Ataroundthesametime,theDoDcontinuedfromwhereitleftoff,fundingprogramssuchastheGlobalMobileInformationSystems（GloMo）,andtheNear-termDigitalRadio（NTDR）.ThegoalofGloMowastoprovideoffice-environmentEthernet-typemultimediaconnectivityanytime,anywhere,inhandhelddevices.ChannelaccessapproacheswerenowintheCSMA/CAandTDMAmolds,andseveralnovelroutingandtopologycontrolschemesweredeveloped.TheNTDRusedclusteringandlinkstaterouting,andself-organizedintoatwo-tieradhocnetwork.NowusedbytheUSArmy,NTDRistheonly“real”（non-prototypical）adhocnetworkinusetoday.

Spurredbythegrowinginterestinadhocnetworking,anumberofstandardsactivitiesandcommercialstandardsevolvedinthemidtolate’90s.WithintheIETF,theMobileAdhocNetworking（MANET）workinggroupwashorn,andsoughttostandardizeroutingprotocolsforadhocnetworks.ThedevelopmentofroutingwithintheMANETworkinggroupandthelargercommunityforkedintoreactive（routesondemand）andproactive（routesready-to-use）routingprotocols[4].The802.11subcommitteestandardizedamediumaccessprotocolthatwasbasedoncollisionavoidanceandtoleratedhiddenterminals,makingitusable,ifnotoptimal,forbuildingmobileadhocnetworkprototypesoutofnotebooksand802.11PCMCIAcards.HIPERLANandBluetoothweresomeotherstandardsthataddressedandbenefitedadhocnetworking.

OpenProblems

Despitethelonghistoryofadhocnetworking,therearestillquiteanumberofproblemsthatareopen.Sinceadhocnetworksdonotassumetheavailabilityofafixedinfrastructure,itfollowsthatindividualnodesmayhavetorelyonportable,limitedpowersources.Theideaofenergy-efficiencythereforebecomesanimportantprobleminadhocnetworks.Surprisingly,therehasbeenlittlepublishedworkintheareaofenergy-efficiencyofadhocnetworksuntilfairlyrecently.Mostexistingsolutionsforsavingenergyinadhocnetworksrevolvearoundthereductionofpowerusedbytheradiotransceiver.AttheMAClevelandabove,thisisoftendonebyselectivelysendingthereceiverintoasleepmode,orbyusingatransmitterwithvariableoutputpower（andproportionateinputpowerdraw）andselectingroutesthatrequiremanyshorthops,insteadofafewlongerhops[8].

Theabilityoffixed,wirelessnetworkstosatisfyqualityofservice（QoS）requirementsisanotheropenproblem.AdhocnetworksfurthercomplicatetheknownQoSchallengesinwirelinenetworkswithRFchannelcharacteristicsthatoftenchangeunpredictably,alongwiththedifficultyofsharingthechannelmediumwithmanyneighbors,eachwithitsownsetofpotentiallychangingQoSrequirements.Reflectingthemulti-layernatureofadhocnetworks,therearenumerousattemptstoimprovetheQoSproblemsfromtheservicecontracts[9]totheMAClayer.ApromisingmethodforsatisfyingQoSrequirementsisamoreunifiedapproachofcross-layerorvertical-layerintegration.Theideaistoviolatemanyofthetraditionallayeringstylestoallowdifferentpartsofthestacktoadapttotheenvironmentinawaythattakesintoaccounttheadaptationandavailableinformationatotherlayers.

Asimilarmulti-layerissueisthatofsecurityinadhocnetworks[10].Sincenodesusethesharedradiomediuminapotentiallyinsecureenvironment,theyaresusceptibletodenialofservice（DoS）attacksthatarehardertotrackdownthaninwirednetworks.Also,sincealargeportionofthenetworknodeswillbedynamicallyreorganizingandforwardingpacketsonbehalfofothers,adhocnetworksareparticularlysusceptibletotheinjectionofbogusnetworkcontroltraffic.Finally,adhocnetworkscanhevictimsofspecializedkindsofsecurityattackssuchasDoSattacksthatcauseanodetouseitstransceiversomuchthatitdepletesitsbattery.

Robotsandsensorsalsoprovidenewhardwarecapabilitiesripefornewmethodsofenhancingadhocefficiency.Robots,forexample,haveatightintegrationbetweentheprocessesofmovement,decision-making,andnetworkingthatallowthemtomodifytheiractionswhiletakingintoaccounttheeffectsonmanydifferentsystemaspects[12].Similarly,sensorsareoftendeployedinawaythatmakestheirrolesandcapabilitiesredundant,suggestingnewwaysofcombiningapplicationknowledgeofdeliveredinformationwiththeroutinglayer[13].

Finally,aproblemthatoverarchesalltheseothersisthelackofwelldefinedandwidelyacceptedmodelsforRFpathattenuation,mobility,andtraffic.Thesetightlyinterrelatedmodelsareneededforquantifyingandcomparingadhocsystemperformancetoacommonbaseline.Thephysicalprocessesofrefraction,reflection,andscatteringofRFradiationismoderatelywellunderstoodbutdifficulttoquantifyindetailwhenincludingalargenumberofcomplexobjectssuchasfoliage,cars,orbuildings.Incontrast,thepatternofmovementofthenodesandtheflowoftrafficcancertainlybeeasilydescribedindetail,butthedependencyonthetargetapplication,thelackofexistingsystemsavailableforstudy,andthelikelyinteractionsbetweenconnectivity,movement,anduserapplications,causesthesemodelstobeill-defined.

TheFuture

Imaginethefollowingscenarios:

awirelessmeshofrooftop-mountedadhocrouters;anadhocnetworkofcarsforinstanttrafficandotherinformation;sensorsandrobotsformingamultimedianetworkthatallowsremotevisualizationandcontrol;multipleairbornerouters（fromtinyrobotstoblimps）automaticallyprovidingconnectivityandcapacitywhereneeded（e.g.,atafootballgame）;anadhocnetworkofspacecraftaroundandintransitbetweentheEarthandMars.

Thesemayseemlikesciencefiction,butareinfactideaspursuedseriouslybytheadhocresearchcommunity.Whileonlytimecantellwhichoftheseimaginedscenarioswillbecomereal,theaboveoffersaglimpseintoboththetechnologicalpotentialandtheevolvingstateoftheart.Wediscussinthissectiontheforcesatplaythatarelikelytoshapethefutureofadhocnetworking,anddiscussthedirectionsinwhichitmayevolve.

Toappreciatetheroleadhocnetworksarelikelytoplayinthefuture,considerthis:

bandwidth-hungryapplicationsandthelawsofphysicsdrivewirelessarchitecturesawayfromcellulartowardadhoc.Thisisbecausemorecapacityimpliestheneedforahighercommunicationsbandwidthandbetterspatialspectralreuse.Higherbandwidthisfoundathigherfrequencies,wherethepropagationisdismal.Further,mobiledeviceshavetobepower-thrifty.Propagation,spectralreuse,andenergyissuessupportashiftawayfromasinglelongwirelesslink（asincellular）toameshofshort