原文5000论文翻译.docx
- 文档编号:29507011
- 上传时间:2023-07-24
- 格式:DOCX
- 页数:56
- 大小:192.59KB
原文5000论文翻译.docx
《原文5000论文翻译.docx》由会员分享,可在线阅读,更多相关《原文5000论文翻译.docx(56页珍藏版)》请在冰豆网上搜索。
原文5000论文翻译
MobileNetwAppl(2012)17:
298–311
DOI10.1007/s11036-011-0339-x
EnergySavingsinWirelessMeshNetworks
inaTime-VariableContext
AntonioCapone·FilippoMalandra·BrunildeSansò
Publishedonline:
26July2011
©SpringerScience+BusinessMedia,LLC2011
AbstractEnergyconsumptionofcommunicationsys-
temsisbecomingafundamentalissueand,amongall
thesectors,wirelessaccessnetworksarelargelyrespon-
siblefortheincreaseinconsumption.Inadditiontothe
accesssegment,wirelesstechnologiesarealsogaining
popularityforthebackhaulinfrastructureofcellular
systemsmainlyduetotheircostandeasydeployment.
Inthiscontext,WirelessMeshNetworks(WMN)are
commonlyconsideredthemostsuitablearchitecture
becauseoftheirversatilitythatallowsflexiblecon-
figurations.Inthispaperwecombinetheflexibility
ofWMNwiththeneedforenergyconsumptionre-
ductionbypresentinganoptimizationframeworkfor
networkmanagementthattakesintoaccountthetrade
offbetweenthenetworkenergyneedsandthedaily
variationsofthedemand.Aresolutionapproachand
athoroughdiscussiononthedetailsrelatedtoWMN
energymanagementarealsopresented.
Keywordsgreennetworking·
wirelessmeshnetworks·optimization·
networkmanagement·energyconsumption·
mathematicalprogramming
1Introduction
GreenNetworkingconsistsofarethinkingoftheway
networksarebuiltandoperatedsothatnotonlycosts
andperformancearetakenintoaccountbutalsotheir
energyconsumptionandcarbonfootprint.Itisquickly
becomingoneofthemajorprinciplesintheworld
ofnetworking,giventheexponentialgrowofInternet
trafficthatispushinghugeinvestmentsaroundthe
worldforincreasingcommunicationinfrastructuresin
thecomingyears.Infact,theInformationandCommu-
nicationTechnology(ICT)sectorissaidtoberespon-
siblefor2–2.5%oftheGHGannualemission[1–3]as
itgeneratesaround0.53Gt(billiontonnes)ofcarbon
dioxideequivalent(CO2e).Thisamountisexpectedto
increaseto1.43GtCO2ein2020(datafrom[4]).
AmongInternetrelatednetworkingequipment,itis
theaccesstheonewiththemajorimpactinenergyex-
penditures.Ithasbeenestimatedthataccessnetworks
consumearound70%ofoveralltelecommunications
networkenergyexpendituresandthispercentageis
expectedtogrowinthenextdecade[5,6].Animpor-
tantpartoftheenergyconsumptionisgivenbythe
wirelesspartoftheaccessandithasbeenestimatedthat
thebasestationsrepresent80%ofthetotalwireless
consumption[7].Itfollowsthatbeingabletomini-
mizebasestationconsumptionrepresentsanimportant
greennetworkingobjective.
Anincreasinglypopulartypeofwirelessaccessare
theso-calledWirelessMeshNetworks(WMNs)[8]that
providewirelessconnectivitythroughmuchcheaper
andmoreflexiblebackhaulinfrastructurecompared
withwiredsolutions.Thenodesofthesedynamically
self-organizedandself-configurednetworkscreatea
changingtopologyandkeepameshconnectivityto
A.Capone(B)·F.Malandra
PolitecnicodiMilano,
PiazzaLeonardodaVinci,32,
Milano,Italy
e-mail:
capone@elet.polimi.it
B.Sansò
EcolePolytechniquedeMontreal,
BoulEdouard-Montpetit,2900,
Montreal,Canada
MobileNetwAppl(2012)17:
298–311
299
offerInternetaccesstotheusers.Obviously,theuseof
wirelesstechnologiesalsoforbackhaulingcanpoten-
tiallymaketheissueofenergyperformanceevenmore
severeifappropriateenergysavingstrategiesarenot
adopted.
Asamatteroffact,theresourcesofWirelessAccess
Networksare,forlongperiodsoftime,underemployed,
sinceonlyafewpercentageoftheinstalledcapacity
oftheBaseStations(BS)iseffectivelyusedandthis
resultsinhighenergywaste[9,10].InWMNsalso,
networkdevicesareactivebothinbusyhoursandin
idleperiods.Thismeansthattheenergeticconsumption
doesnotdecreasewhenthetrafficislowandthatit
wouldbepossibletosavelargeamountsofenergyjust
byswitchingoffunnecessarynetworkelements.
Thefocusofourworkistocombinetheversatilityof
WirelessMeshNetworkswiththeneedofoptimizing
energyconsumptionbygettingadvantageofthelow
demandperiodsandthedynamicreconfigurationsthat
arepossibleinWMNs.Weproposetominimizetheen-
ergyinatimevaryingcontextbyselectingdynamically
asubsetofmeshBSstoswitchonconsideringcoverage
issuesoftheservicearea,trafficrouting,aswellas
capacitylimitationsbothontheaccesssegmentandthe
wirelessbackhaullinks.Toreachourobjective,wepro-
videanoptimizationframeworkbasedonmathematical
programmingthatconsiderstrafficdemandsforasetof
timeintervalsandmanagestheenergyconsumptionof
thenetworkwiththegoalofmakingitproportionalto
theload.
Energymanagementinwirelessaccessnetworks
havebeenconsideredveryrecentlyinafewprevious
articles[1,2,9,11–14](seeSection2foradetailed
reviewofthestateoftheart).Inthispaper,wepresent
anovelapproachforthedynamicenergymanagement
ofWMNsthatprovidesseveralnovelcontributions:
–
–
Weconsidernotonlytheaccesssegmentbutalso
thewirelessbackhaulofwirelessaccessnetworks;
Wecombinetogethertheissueofwirelesscover-
age,fortheaccesssegment,andtherouting,forthe
backhaulnetwork,andoptimizethemjointly;
Weexplicitlyincludetrafficvariationsoverasetof
timeintervalsandshowhowitispossibletohave
energyconsumptionfollowingthesevariations;
Weprovidearigorousmathematicalmodelingof
theenergyminimizationproblembasedonMixed
IntegerLinearProgramming(MILP),andsolveit
totheoptimum.
inSection3.Theoptimizedmodelingapproachfor
systemmanagementisintroducedinSection4.The
resolutionapproachandathoroughresultsanalysisare
presentedinSection5.Section6concludesthepaper.
2Relatedwork
Theproblemofenergyconsumptionofcommunication
networksandthemaintechnicalchallengestoreduceit
havebeenpresentedintheseminalworkbyGuptaand
Singh[15].Severalproposalstoreducenetworksfoot
printaswellasenergyconsumptionhaveappearedin
thelastfewyears,consideringbothwirelessandwired
networks[3,16–22].
Goodoverviewsoftheresearchongreennetworking
andmethodologicalclassificationsaregivenin[23,24]
wheredifferentmethodsadoptedintheliteraturefor
bothwiredandwirelessnetworksaresurveyed.
Inwhatfollows,wefocusonwirelessnetworksonly.
Theliteratureinwirelessdeviceenergyoptimization
isquitelarge,giventhelimitationofthebatteryand
thenaturalrestrictionsofthewirelessmedium.Infact,
energyconsumptionhasalwaysbeenaconcernfor
wirelessengineeringgiventhemobilityofusersthatre-
quireportability,whichmakescoverageandbatterylife
issuesatruechallenge.Thereis,indeed,alargebodyof
workonenergy-efficiencyfordevicesandprotocolsfor
cellular,WLANandcellularsystems(see[25],foran
excellentsurvey).However,theinterestforenergyop-
timizationofthewirelessinfrastructurehasonlypicked
upinrecentyearsgiventheexplosioninInternetwire-
lessapplications.
Therehasbeensomeworktocomparewirelessand
wirelineinfrastructureconsumption.Forinstance,let
usmention[15]wheretheenergycost(Wh/Byte)for
atransmissionovertheInternetwascomparedtothe
costofthesametransmissioninawirelesscontext
(forinstanceWi-Fi802.11b).Wirelessresultedmore
efficientbyasmallfactorwithomnidirectionalantenna
anditwasfoundthatthefactorcouldbeimproved
usingdirectiveantennas.
Ourmainconcern,however,iswirelessnetwork
managementforwhichwehavefoundarticlesthatdeal
eitherwithWirelessLocalAreaNetworks(WLANs)
orwithtraditionalcellularaccessnetworks.
InWLANs,wementiontheworkof[10]thatpre-
sentedstrategiesbasedontheresourceon-demand
(RoD)concept.Marsanetal.[9]proposedananalytical
modeltoassesstheeffectivenessofRoDstrategiesand
[11]showsmanagementstrategiesforenergysavingsin
solarpowered802.11wirelessMESHnetworks.
–
–
Thepaperhasbeenstructuredasfollows.Aftera
briefsurveyoftheliteratureconcerninggeneraland
wirelessGreenNetworkinginICTinSection2we
presentthesystemmodelandpreliminarydescriptions
300
MobileNetwAppl(2012)17:
298–311
Concerningcellularaccessnetworks,[2]considered
thepossibilityofswitchingoffsomenodesbutwithout
consideringtrafficvariations,whichcanproducesub-
stantialsavingsgiventhatcellularsystemsaregener-
allydimensionedforpeaktrafficconditions.Marsan
etal.[13],ontheotherhand,studieddeterministic
trafficvariationstocharacterizeenergysavingsand
showedthattheycanbearound25–30%fordifferent
typesofregularcelltopologies.Anotherenergyman-
agementstudyisprovidedby[12]whereitisshownthat
theon-offstrategyforUMTSBSisfeasibleinurban
areas.Gongetal.[14]consideredarandomtraffic
distributionanddynamicallyminimizedthenumberof
activeBSstomeetthetrafficvariationsinbothspace
andtimeand[1]presentedanoptimizationapproach
fordynamicallymanagingtheenergyconsumption.
Thedifferencesofourworkwiththepapersmen-
tionedaboveisthatthelaterdealexclusivelywith
accessnetworkswhileourgoalistomanagetheenergy
consumptionofWMNsthatusethewirelessmedium
notonlyfortheaccesssegmentbutalsofortheback-
bone.Thepresenceofthewirelessbackboneforcesus
toconsidertheroutingoftrafficfrombasestations(or
meshaccesspoints)tothemeshgateways(intercon-
nectingtheWM
- 配套讲稿:
如PPT文件的首页显示word图标,表示该PPT已包含配套word讲稿。双击word图标可打开word文档。
- 特殊限制:
部分文档作品中含有的国旗、国徽等图片,仅作为作品整体效果示例展示,禁止商用。设计者仅对作品中独创性部分享有著作权。
- 关 键 词:
- 原文 5000 论文 翻译