USF Libraries
USF Digital Collections

Comparison and improvement of different access methods in femtocell networks

MISSING IMAGE

Material Information

Title:
Comparison and improvement of different access methods in femtocell networks
Physical Description:
Book
Language:
English
Creator:
Demirdogen, Ibrahim
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Femtocell
Open Access
Closed Access
Closed Subscriber Group
Dissertations, Academic -- Electrical Engineering -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: A variety of wide band wireless systems have been pushed towards their limits in order to meet growing interest for high data rate in wireless communications.In particular, the limit due to the spectrum scarcity forces communication systems to utilize the spectrum resource at maximum efficiency level. One of the methods that allow effective spectrum employing is to cover multiple systems over same spectrum source by allowing bearable interference to occur between them. Femtocells have been recently introduced as a remedy to spectrum scarcity and coverage problems in current cellular structures. Femtocells are personal use base stations and they share the spectrum in a way that they can coexist with the macrocell. This thesis provides a critical reviews of different access methods in femtocell networks and further introduces improvements related to these access methods. Simulation results validate capacity improvement of proposed techniques compared to the existing access methods.
Thesis:
Thesis (M.S.E.E.)--University of South Florida, 2010.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Ibrahim Demirdogen.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains X pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E14-SFE0003467
usfldc handle - e14.3467
System ID:
SFS0027782:00001


This item is only available as the following downloads:


Full Text

PAGE 3

IamgratefultoDr.IsmailGuvencnotonlyforadvisingmethroughoutourcollaborationwithNTTDOCOMO,butalsoforhissincerefriendship.IalsowanttoacknowledgeDr.MustafaEminSahinfromDOCOMOUSALabsforhissupportandsincerefriendshipandsupportthroughoutmyexperienceinUSF. IowemuchtomyfriendsHasanBasriCelebi,SabihGuzelgoz,OzgurYurur,EvrenTerzi,Dr.HishamMahmoud,AliGorcin,IsmailButun,MuradKhalid,M.BahadrCelebi,AlphanSahin,Dr.BilalBabayigit,Dr.SerhanYarkan,M.CenkErturk,HazarAk,andSadiaAhmed.Wesharedsomanythingswiththemovertheyearsthatwespenttogether.Theyalsotaughtmesomanyvirtues.Sincerefriendshiptostartwith,unselshness,toler-ance,andhelpfulness.Iamgratefultothemformakingmeabetterperson. IalsowanttoexpressmygratitudetomyfriendsintheTurkishcommunityinTampa,FL,especiallytoIsmailUlukaya,OmerOzbek,SelmanTurk,FatihDemir,ErdemOnsal,SalihErdem,andErdo~ganBulutfortheirsupporttous,students,wheneverweneedit. MysincereappreciationgoestomydearbrothersOguz,andYavuzandmydearsisterSumeyyaforleadingmetotherightdirection,andalwaysencouragingmeforpursuinghigherdegrees.Itisnotpossibletothankthemenough,butIwantthemtoknowthatIwillbegratefultothemthroughoutmylife.

PAGE 4

LISTOFFIGURESiii ABSTRACTv CHAPTER1INTRODUCTION1 1.1BenetsofFemtocellNetworks1 1.2FemtocellChallenges3 1.3CurrentandFutureMarketStatus4 1.3.1IndustrialVendors5 1.3.2FutureExpectations7 CHAPTER2SIMULATIONENVIRONMENT8 2.1SimulationEnvironmentDescription10 2.1.1SimulationParametersandPathLossModels11 CHAPTER3CLOSEDACCESSFEMTOCELLS20 3.1CapacitiesofMacrocellandFemtocellUsers21 3.2ClosedAccesswithDynamicSpectrumReuse24 3.2.1MaximumSumCapacity25 3.2.2MinimumMacrocellLoss30 3.2.3MinimumEectiveInterference33 3.3SimulationResults35 CHAPTER4OPENACCESSFEMTOCELLS42 4.1CapacityAnalysisofOpenAccessSchemeinTwoTieredNetworks44 4.1.1MacrocellCapacitywithoutFemtocellDeployment44 4.1.2CapacityImprovementwithFemtocellDeployment45 4.1.2.1DedicatedSpectrumResourceAllocation45 4.1.2.2Co-channelSpectrumResourceAllocation46 4.2OpenAccesswithLoadBalancing(OA-LB)48 4.3SimulationResults50 4.3.1ComparisonofDedicatedChannelvs.Co-channelModes50 4.3.2OpenAccesswithLoadBalancing(OA-LB)53 CHAPTER5CONCLUSIONANDFUTUREWORK60 REFERENCES62i

PAGE 5

Table1.2Compilationofpubliclydeclaredfemtocellworldwidetrialsbytheendoftherstquarterof2010.6 Table2.1Parametersandassumptionsformacrocellsystem.9 Table2.2Parametersandassumptionsforfemtocellsystem.10 Table2.3Simulationparameters.17 Table3.1Comparisonofmediancapacities(inmbps)w.r.t.dierentmetrics.37 Table4.1Comparisonofmediancapacities(inMbps)withandwithoutfemtocells.53ii

PAGE 6

Figure2.1Simulatormainmenu.11 Figure2.2MNBandMUEsettings.12 Figure2.3MacrocellbasestationRSScoverage15 Figure2.4MacrocellbasestationRSScoverage16 Figure2.5RSSofmacrocell(top)andfemtocell(bottom)basestations.18 Figure2.6SINRofmacrocell(top)andfemtocell(bottom)basestations..19 Figure3.1Systemmodelandinterferencescenariosforcoexistingfemto-cellandmacrocellnetworks.22 Figure3.2Interferenceanalysisofmacrocelluserssubjectedtoclosedac-cesspolicy.25 Figure3.3ULsum-capacityandfemtocellcapacitywithDSR(Pf;j=10dBm,Pm;i=20dBm,dmBS=500m).29 Figure3.4ULsum-capacityandfemtocellcapacitywithDSR(Pf;j=10dBm,Pm;i=20dBm,dmBS=1000m).30 Figure3.5DLsum-capacityandfemtocellcapacitywithDSR(Pf;j=10dBm,Pm;i=40dBm,dmBS=500m).31 Figure3.6DLsum-capacityandfemtocellcapacitywithDSR(Pf;j=10dBm,Pm;i=40dBm,dmBS=1000m).32 Figure3.7Femtocellcapacitylossratio(f)vs.overlappedband(OB)ra-tio(Pm;i=141:5dBmandIf;j=f111:5;116:5;:::;156:5gdBm).34 Figure3.8ComparisonofmacrocellusercapacityCDFs(indensemMSenvironment).38 Figure3.9ComparisonoffemtocellusercapacityCDFswithf;jvalue(indensemMSenvironment).39iii

PAGE 7

Figure3.11ComparisonoffemtocellusercapacityCDFswithvariousf;jvalues(insparsemMSenvironment).41 Figure4.1(a)Dedicatedspectrumallocationvs.(b)Co-channelspec-trumallocationoffemtocellandmacrocellnetworks.45 Figure4.2Openaccessandclosedaccessmodesoffemtocellnetworks.48 Figure4.3CapacityCDFsfornofemtocells,dedicatedchannelfemtocells,andco-channelfemtocells(indoorusers).51 Figure4.4CapacityCDFsfornofemtocells,dedicatedchannelfemtocells,andco-channelfemtocells(outdoorusers).52 Figure4.5Scenariofortheopen-accesssimulations.54 Figure4.6CapacityandRSSofmMSforassociationswithdierentcells(dmBS=800m).55 Figure4.7CapacityofmMSfordierenthand-oapproaches(dmBS=300m.).56 Figure4.8CapacityofmMSfordierenthand-oapproaches(dmBS=500m.).57 Figure4.9CapacityofmMSfordierenthand-oapproaches(dmBS=800m.).58 Figure4.10MeancapacityofmMSoverthetrajectoryinFig.4.5.59iv

PAGE 10

Figure1.1Flatarchitectureoffemtocellnetworks.2

PAGE 11

Architectureoffemtocellnetworks,atnetworkarchitectureinsteadofhierarchicaltype,reducesthenumberofcomponentstosetupanetwork.Flatnetworkalsosimpliesthedeploymentprocessbyeliminatingthebasestation,radionetworkcontroller,andcomplexhierarchicalbasestationrelationshipthatcharacterizetraditionalmacroaccessnetworks[4].Fig.1.1illustratestheathierarchicalstructureofasimplefemtocellnetwork.HomeNode-Bs(HNBs)aredirectlyconnectedtothemobileoperatorcorenetwork(MOCN)viainternetbackhaul(e.g.cable(ber),xDSL)andtheotherlegoftheMOCNisconnectedtomacrocellnetwork.Thus,itconvertspacketdatanodesofmacro-cellularradionetworkintoasmallfemtocellaccesspointsandenablesaneasyplug-and-playprocess. Itcanbesaidthatbasedontheadvantagesandopportunitieslistedabove,femtocellshavebeenoneoftheopenresearchanddevelopmentareasinwirelessnetworksforseveralyears.1.2FemtocellChallenges

PAGE 12

Femtocelldeployments,unlikepicocelldeploymentconsideredforbusinessenvironment,havebeenmuchmorefocusedonend-userdeployment.However,serviceprovidersstillcontinuetocategorizeanumberofmajorusersectorsforinterestingservicescenariosforfemtocellssuchaspublicaccess,subwayandruralenvironment.Althoughthecoverageoffemtocellservicedeploymentisspreadingoutyearbyyearallaroundtheworld,noneofthecurrentserviceprovidershaslaunchedtheforthgeneration(4G)(WiMAXorLongTermEvaluation(LTE))femtocellservice[5].However,rsttiercellularcompanieshavebeendeclaredthatitishighlyprobablethatLTEandfollowinghighcapacityairinterfaceswillbedeployedviafemtocells.

PAGE 13

Operator(Country) Price Capabilities Optimus(Portugal) 99.99euroupfront,7.8euromonthly Upto43G-users SFR(France) 99.99euroupfront Upto43G-users NTTDOCOMO(Japan) $10permonth Upto43G-users ChinaUnicom(China) FAP:CYN1.200,CYN10permonth Upto43G-users AT&T(USA) $159 Upto43G-users Vodafone(UK) Variousoptions Upto43G-users Verizone(USA) $249.99 upto32G1xRTT-users StarHub(Singapore) $32.1permonth Upto43G-users Sprint(USA) $4.99permonth Upto32G-users ($10unlimitedcall,$20familyplan) Thegrowthofthefemtocellmarketcanalsobeunderstoodthroughthetrialsthathavebeenexperiencedatallregions.Althoughthenumberofpublicannouncementofthosetrialsisquitelimitedandsomeofthosetrialsareleftasclosedboxes,thenumberofthetrialdeclarationshasbeenincreasedsofar.SomeofpubliclydeclaredglobalfemtocelltrialsaresummarizedinTable1.22.1.3.1IndustrialVendors

PAGE 14

CellularCompany Country ChinaMobile China Cellcom USA ChunghwaTelecom Taiwan Comcast USA FgupZnisTechnopark Russia Maxxis Malaysia Mobilkom Austria PortugalTelecom Portugal T-Mobile Germany,Poland TDC Denmark TelefonicaO2 Europe TIM Italy Vodafone Spain videdforinitialmarketphase.Currentlynineworldwideend-to-endsolutionprovidersandfemtocellsystemsintegratorsarepresent.ThepotentialofthemarketisconrmedthroughtheattentionofsomepioneercompaniesinthissectorsuchasAlcatel-Lucent,HuaweiandCisco.

PAGE 15

Thesevelegsofthemarketconstituteahealthyfemtocellecologyandenablerapidpene-trationoffemtocellconcepttotheendusers.Theprogressinthefemtocellmarkethasbeenacceleratingyearbyyearandproductrangealsohasbeenexpanding.FlexiblereferenceplatformshavebeenintroducedinordertoproduceFAPstosupportbroadbandaccesstechnologies.1.3.2FutureExpectations

PAGE 16

BesidestheTable2.1andTable2.2,in[7]variousnumbersoffemtocells(N)areuni-formlydistributedwithinthemacrocellwhereNis1,2,4,or10permacrocellcoveragearea.Also,totalnumberofusers(Ntot)issetto60usersforeachentiremacroarea.Femtocellhasupto4users(Nf)andremainingusers(Ntot(NfN))areuniformlydistributedtothemacrocellarea.Coverageofafemtocellisassumedtobe40mandtheminimumdistancebetweennewnodeandregularnodesisassumedtobelargerthan35m. Althoughavarietyofdierentparametersisdeliveredbythemeetingdocuments,therearealsosomeparameterswhicharere-denedinalmosteverydocumentimplyingthatparametersarenotsettledyet.Oneoftheseparametersispathlossmodelemployedforuserequipments(UE)becausetherearevarietyofscenariosthatshouldbeconsideredsuchassuburbanandurban(dense)deploymentwithdierentfemtocellnodedeploymentssuch8

PAGE 17

Parameter Assumption CellularLayout Hexagonalgrid,3sectorspersite,reuse1 Inter-sitedistance 500mor1732m. Numbersites 19(=57cells)or7(=21cells) CarrierFrequency 2GHz Shadowingstandarddeviation 8dB PenetrationLoss(WallLoss) 10dBor20dB NumberofBSantennas 2Rx,2Tx NumberofUEantennas 2Rx,1Tx TotalBSTXpower(Ptot) 46dBm UEdistribution UEdistributionUEsdistributedwithuniformdensity withintheindoors/outdoorsmacrocoveragearea. UEspeedsofinterest 3km/h DLReceiverType Maximumratiocombining(MRC)forsinglestreamor Minimummeansquareestimation(MMSE)formultiple. ULReceiverType MRC asdual-stripe,and55gridmodel.Pathlossmodelgivenin[8]isemployedinthissimulationwhichexplainedindetailinSection2.1. 9

PAGE 18

Parameter Assumption FemtocellFrequencyChannel Co-Channel(Samefrequencyasmacrocell)or DedicatedChannel(Adjacentfrequencyasmacrocell) NumberRxantennasHeNB 2 Exteriorwallpenetrationloss 10or20dB Shadowingstandarddeviation 4dB Min/MaxHeNBTxpower 0/20dBm Fig.2.1showsthemainGUIwhichhelpstodenethescenariotobesimulated.Fem-tocellcoveragewith/withoutpotentialfemtocelllocationandmacrocellcoverageareseenthroughthewindowwithacolorbarindicatingpowerimage(dBm/Hz)oftheentirecell.Capacitycumulativedistributionfunction(CDF)ofeithermacrocellorfemtocellisalsogivenbythisinterfaceasanoutputofdenedscenario.EveryCDFhasthreelabelsin-dicating\outdooruser",\indooruser"and\allusers"alsomediancapacitiesofthoseusertypesaregivenatthebottomoftheinterface. Fig.2.2showsmacrocellbasestationsettings,femtocellbasestationsettings,userequip-mentsettingsandchannelsettings.Mainbasestationparameterssuchascarrierfrequency,10

PAGE 19

Figure2.1Simulatormainmenu.11

PAGE 20

whereFcisthecentraltransmissionfrequencyinMHz,Ndistancepowerlosscoef-cient,disthetransmitterreceiverseparationinmeter,LfistheoorpenetrationfactorindB.Notethatatleast1meterdistanceisnecessaryinordertocalculatethepathlossproperly.Onekeyadvantageofthemodelisthatitdoesnotneedanyspecicinformationdescribingtheindoorenvironmentsuchasnumberofwallsandwindowsbetweentwoends. Figure2.2MNBandMUEsettings.12

PAGE 21

dbpwhered6dbpLUdB=Lbl+20+40log10d dbpwhered>dbp;(2.2) andlowerboundgivenasLLdB=Lbl+20log10d dbpwhered6dbpLLdB=Lbl+40log10d dbpwhered>dbp;(2.3) wheredisthedistancebetweentwoendsinmeter,dbpisthedistancewherevaryingrateofchangeinthepathlossboostscalledbreakpointdistance.Breakpointdistancedbpisempiricallycalculatedasdbpu4HtxHrx whereHtxandHtxaretheantennaheightsoftransmitterandreceiverabovethestreetlevelrespectivelyandisthewavelengthofthesignalcalculatedby(c=Fc)wherecisthespeedoflight.Lblin(2.2)and(2.3)isthebasictransmissionlossat13

PAGE 22

Besidesthepathlossmodel,otherparametersthatusedthroughoutthesimulationsaregiveninTable2.3aslongastheyarenotindicatedasdierentvalue.Theparameterswhicharenotcoveredintheinitialphaseofthesimulatorareleftasfuturework. Thecoverageimagebasedonthereceivedsignalstrength(RSS)ofthemacrocellandfemtocellnetworksshowninFig.2.3andFig.2.4respectivelywiththecolorbarsindBm/Hzunit.ItcanbeseenfromFig.2.3thatindoorusersaresuerfromsevereRSSresultsinpoorcapacityforthatusers.Ontheotherhand,Fig.2.4showsthatfemtocellsusedinindoorcanbeusedasaremedyforthecoverageproblemandthecapacityproblemoftheindooruserscanbehandledbydeployingfemtocellswhichresultinbetterRSS.Pleasenotethatlog-normalshadowingeectisnotincludedneitherformacrocellnorfemtocellssothesignaldegradationseemsquitesmooth.Theseparametersareconsideredasfurtherversionofthesimulator. Amacrocell/femtocellscenarioasinFig.2.3andFig.2.4areconsideredwithanmBSlocatedinthecenterofthecell(shownwithatriangle).Thebuildings(squares)and150mobileusers(circles)arescatteredovera600600gridsuchawaythat80%oftheusersareindoorwhereas20%oftheusersareoutdoor[6],withinthe(hexagonal)bordersofthemacrocell,where34%oftheareaisinsidethebuildingsand66%oftheareaisoutdoors.CellselectionisbasedontheSINRmetric,whereMSsjointhefemtocell/macrocellthathasthebestSINR.Inallsimulations,idlefemtocells(withnousers)aredetectedanddisabled(i.e.,theirtransmitpoweraresettozero)tominimizeinterference. A3DplotoftheRSSreceivedfrommacrocellandfemtocellusersisillustratedinFig.2.51,whichshowsthat(duetoisolationprovidedbythewallpenetrationloss)

PAGE 23

SINRresultsinFig.2.62conrmthatthereisasharpdropintheSINRofthemacrocellinindoorlocations,wherefemtocellshaveagoodcoverage.15

PAGE 25

Parameter Value Centralfrequency 2:1GHz Bandwidth 5MHz Coverage(radius)(mBS,fBS) 0:5km,8m Max.TXpower(mBS,fBS) 41:75dBm,20dBm Thermalnoisedensity Wallpenetrationloss(WL) 10dB,20dB Antennagain(mBS,fBS) 17dBi,2dBi Feeder/cableloss(mBS,fBS) 3dB,1dB Antennaheights(mBS,fBS,MS) 15m,1:5m,1:5m Housesize 15m15m Streetwidth 5m Distancebetweengridpoints 5m Numberofuserspermacrocell 150 Indoorareavs.outdoorarea 34%vs.66% Schedulingstrategy Equaluserbandwidth Indoor/indoorPLmodel ITUP.1238 Indoor/outdoorPLmodel ITUP.1411+WL Outdoor/outdoorPLmodel ITUP.1411 Outdoor/indoorPLmodel ITUP.1411+WL 17

PAGE 28

Co-channeldeploymentoffemtocellnetworkswithamacrocellnetworkisapopularapproachinordertoecientlyutilizetheavailablespectrumresources.Ontheotherhand,suchco-channeldeploymentsalsoresultinco-channelinterference(CCI)problemsbetweenthefemtocell(s)andthemacrocell.In[8,10],adetaileddiscussionofsixdierentCCIscenariosbetweenfemtocellbasestations(fBSs),macrocellbasestation(mBS),andthemobilestations(MSs)ispresented.Itisarguedthattheaccesscontrolmethodusedbyafemtocellmaysignicantlyimpacttheinterferencescenariosthatmaybeobservedbythemacrocellandthefemtocellusers. Therearetwomajoraccesscontroloptionsforco-channelfemtocellnetworks.Foropenaccessfemtocells,anymacrocellMS(mMS)isallowedtojoinaparticularfemtocellnetwork.Thismayconsiderablyincreasethenumberofusersperfemtocell,thereforedecreasingtheaveragebandwidthavailableperfemtocelluser.Alternatively,forclosedaccesstypeof20

PAGE 29

Inthischapter,wedealwithclosedaccessfemtocellsnetworkwheremMSsintroducehighinterferencewhentheyareclosetoafemtocell.Intheliterature,[14]proposedamethodthatmitigatesdownlink(DL)interferenceatmMSviaspectralresourcepartitioningandpreventingtheusageoftheoverlappedresourceatfemtocell.In[15],ahybridspectrumsharingmethodisproposedwhichusesaninterferencethresholdforoverlappedspectrumavoidance.Cognitivelymeasuredinterferencesignatureofthenetworkisusedbyfemtocellstodeterminethereusepriorityofchannelsin[16].However,detailedevaluationsonwhentoavoidreusingtheoverlapspectrumatafemtocellisnotpresentedinanyofthesereferences.Weproposedierentmethodsforreuseoftheoverlapspectruminadynamicmanner.Inparticular,threedierentmetricsareconsidered:maximumsumcapacity,minimummMSloss,andminimumeectiveinterference. Thechapterisorganizedasfollows:Section3.1reviewstheuplink(UL)andDLcapaci-tiesoffemtocellandmacrocellusersconsideringtheinterferencefromeachother.Section3.2proposesandanalyzesthreedierentcriteriaforthereuseoftheoverlapband(OB)atthefemtocellsinordertoimprovethecapacityinclosedaccessmode.Simulationresultsillus-tratingthepotentialgainsusingtheproposedmethodsarepresentedinSection3.3andthelastsectionconcludesthechapter.3.1CapacitiesofMacrocellandFemtocellUsers

PAGE 31

whereiisthemMSindex,jisthefMSindex,Nm;iisthenumberofresourceunitsusedbythemMS-i,Nf;jisthenumberofresourceunitsusedbyfMS-j,Bisthebandwidthperresourceunit,Pm;iisthereceivedenergyforthedesiredsignalatmMS-i,Pf;jisthereceivedenergyforthedesiredsignalatfMS-j,Im;iisthereceivedenergyfortheinterferencesignalfrommMS-i,If;jisthereceivedenergyfortheinterferencesignalfromfMS-j(allofthedesired/interferencesignalenergiesareperresourceunitandinterferencetermsareassumedtohaveaGaussiandistribution),andN0isthenoisepower.Then,theuplinksum-capacityforthemMSandfMSuserscanbeexpressedas2CULtot(df;i)=Xi2SmCULm;i(df;i)+Xj2SfCULf;j(df;i);(3.3) where,SmisthesetofallmMSsofinterestandSfisthesetofallfMSsofinterest. Asanalternativetohavingaco-channeloperation,thefemtocellmayalsoavoidusingthespectrumresourcesofthemMSs,inordertopreventinterferenceproblems.Then,the

PAGE 32

where,theULsum-capacitycanbeexpressedas~CULtot(df;i)=Xi2Sm~CULm;i(df;i)+Xj2Sf~CULf;j(df;i):(3.6) Comparing(3.6)with(3.3),thefemtocellnolongerbenetsfromreleasingtheoverlappedbandwidth(OB)in(3.6).However,forsmallermMS-fBSdistancedf;i,interferenceintheOBwillbesignicantforboththefemtocellandthemacrocellasexpressedin(3.1)and(3.2).Therefore,avoidingthereuseoftheOBasin(3.4)and(3.5)isexpectedtoimprovethesum-capacityin(3.6)whenthereis(are)mMS(s)inthevicinityoffemtocell(s).Usingasimilarapproachastheoneinabove,DLcapacityequationswithandwithoutreuseoftheOBcanbeeasilyformulated.3.2ClosedAccesswithDynamicSpectrumReuse Weconsiderthatthefemtocelliscapableofimplementingperfectspectrumsensing(SS)andusesadynamicspectrumreuse(DSR)policyinordertobenetfromtheOBwiththemMSswheneverpossible.Forsimplicity,weconsiderthatthereisasinglesignicantmMSinterfererforagivenfemtocell.Dierentcriteriamaybeconsideredfordecidingwhether24

PAGE 34

Asacasestudy,consideratwo-userscenario,whereanmMSusesasubsetofthefrequencyresourcesofanfMS.Byequating(3.3)and(3.6)andaftersomemanipulation,wemaywrite1+Pm;i which,uponsomefurthermanipulationsimpliesto(BN0)2(Pf;j+Pm;i)+BN0Pf;jIf;j+BN0Pm;iIm;i+BN0Pm;iPf;j=Pm;iIm;iIf;j+BN0Pm;iIm;i+BN0Pm;iIf;j+(BN0)2Pf;j:(3.9) Cancellingsomecommonterms,wemayexpress(3.9)asBN0(Pf;jIf;j+Pm;iPf;j)+(BN0)2Pf;j=Pm;iIm;iIf;j+BN0Pm;iIf;j;(3.10) anddividingbothsidesof(3.10)byPm;i,wehaveBN0Pf;jIf;j SinceN20in(3.11)istypicallynegligiblecomparedtootherterms,itcanbedropped,whichyieldsthefollowinginterferencethreshold26

PAGE 35

Notethat(3.12)givestheULinterferencepowerthresholdtobeusedatanfBSforachievingMSC.IftheULinterferenceislargerthanthisthreshold,itbecomespreferableatanfBS(inthesenseofmaximizingthesumcapacity)toavoidusingtheOBwiththemMS.Inotherwords,CULmsccanbewrittenasCULmsc=8>><>>:CULtot(df;i);ifIm;i
PAGE 36

where,I(thr)fexpressesDLinterferencepowerthresholdfromfBStomMS.IftheDLin-terferenceisgreaterthanI(thr)m,itisworthtogiveupusingOBinordertoachieveMSC.Then,CDLmsccanbewrittenasCDLmsc=8>><>>:CDLtot(df;i);ifIm;i
PAGE 37

TheDLresultsinFig.3.5andFig.3.6showthatsimilartotheUL,thereisacross-overdistancedf;iwherereuseoftheOBnolongerbecomespreferableformaximizingthe29

PAGE 39

Ontheotherhand,theremayalsobescenarioswherecapacityaugmentationatthemMScausesaconsiderablecapacitylossatthefemtocell.ThislosscanbeexpressedasLULf(df;i)=Xj2SfCULf;j(df;i)~CULf;j(df;i);(3.17)31

PAGE 40

wheretheratioofthecapacitylossthatafemtocellobservesis4f=LULf(df;i) NotethatLULf(df;i)dependsonthefemtocellpoweroveranOB,especiallythemMSswhoarevictimofclosedaccesspolicywithasevereinterference(Pm;i6If;j).However,itismostlydrivenbytheratioofOBwhichisdenedasOBoverthetotalfemtocellbandwidth. Inthissection,asecondmetricthatconsidersaminimummacrocellloss(MML)isevaluated.ProposedmetrictakesintoaccountthecapacitylossofanmMSwhoisvictim

PAGE 41

wheretheratioofthecapacitylossthatanmMSobservesis=LULm;i(df;i) ~CULm;i(df;i):(3.20) ThenthedecisionmetricbasedontheMMLcriterioncanbewrittenas wherethr(06thr61)indicatestolerancelevelofanmMStocapacityloss.FormMSsthatarevulnerabletocapacityloss,thriscloseto0.Whenthriscloserto1,themMSismoretoleranttocapacityloss.Inotherwords,thr=1showsthatthemMShasfulltoleranceandOBwillbeemployedforanycase,whereasthr=0indicatesthatthemMShasnotolerancetocapacitylossandOBreusecannotbeemployed.NotethatisindependentfromOBmeaningthatthelossobservedatmMSisonlyduetodegradationofitsSINRvalue.Fig.3.7depictsthefvaluesaccordingtodierentOBratiosforvariousfMSpowers.ItalsoshowsthateectofOBratiohasmoreinuenceonfaftertheIf;jdominatesthePm;i.Fig.3.7alsoshowscorrespondingvaluesforeachfMSpower.AnillustrativeexampleisshowninFig.3.7whethertheOBwillbeusedornotwithrespecttothr=0:5.3.2.3MinimumEectiveInterference

PAGE 42

whereIoTmrepresentsthetotalfMSinterferenceoverthermalnoiselevelatthemBSoverthetransmissionbandwidthB. TheIoTparametermaybeusedtodetectthefemtocellinterferenceatthemMSs.Iftheinterferencelevelislarge,themMSsmaysignalthisinformationtotherelevantfemtocell5,whichmayre-scheduleitsfMSstodierentbandstopreventinterference.TheIoTvalue

PAGE 43

whereIoTm;irepresentstheallfBSinterferencesoverthermalnoiselevelattheithmMSoverthetransmissionbandwidthB.Then,minimumeectiveinterference(MEI)decisionmetricbaseduponIoTcanbewrittenasCDLmei=8>><>>:CDLm;i;ifIoTm;i
PAGE 44

MediancapacitycomparisonbetweenOAandclosedaccessdeploymentschemesispro-videdinTable3.1.NotethatmacrocellcapacityreductionbetweenOAandclosedaccessisnotonlyduetotheincreaseininterferenceobservedbymMSsbutalsobecauseoftheincreaseinnumberofmMSsoverthesamespectrumresource.Table3.1alsoshowsthatmediancapacityismaximizedbyemployingsumcapacitymaximizationmetric(MSC). ThecapacityCDFsofmMSsandfMSswithandwithouttheproposedDSRtechniquesaredepictedinFig.3.8andFig.3.9.OnlythemMSsthatarewithinthevicinityoffemtocellsareconsideredinthecapacityCDFs.Tothisend,thevictimmMSsaredetectedusingoneofthethreemetricsdiscussedintheprevioussection.Therefore,noDSRwithMSC-detected,MML-detected,andMEI-detectedrefertotheschemeswithoutDSRandwherethemMSusersaredetectedbasedononeofthethreedierentcriteria6.ThesuperiormMSCDFcurveseemstobelongtoMEIwhereIthrisequalto5dB,howeverapproximately30%ofdetectedmMSsarefoundtosuerfromlowcapacityalthoughtheircapacityisgood.Ontheotherhand,allmMSsdetectedbyMMLcriteria,wherethr=0:5,suerfromlowcapacity.TheenhancementinmMScapacitycomesattheexpenseoffemtocellusercapacityloss.Fig.3.9showsthecorrespondingcapacitylossinfMSsforeachmetric.ThedierencebetweentheCDFvaluesforagivencapacitymayvarybetween2.5%-3.5%forproposedmetrics. ThecapacitylossthatfMSobservesisheavilydependentontheOBthatfMSdoesnotreuse.Therefore,forthescenariowherethenumberofmMSsinthemacrocellismuchmorethanthenumberoffMSsinafemtocell,theeectofreleasingtheOBisbearabletothefMSs.Ontheotherhand,forascenariowherethenumberofmMSinamacrocelliscomparabletothenumberoffMSsinafemtocell,releasingtheOBaectsthefMSsmoreseverely.Fig.3.10andFig.3.11showmMScapacityimprovementandcorrespondingfMS

PAGE 45

Closed DSR DSR(MML, DSR(MEI, Open(%) Users(%) Access (MSC) All 25:09 25:84 25:48 25:68 31:00 Indoor(78) 31:81 32:76 32:34 32:47 38:85 Outdoor(22) 1:91 2:00 1:91 2:12 3:86 Femto(52) 46:67 47:74 46:80 47:40 39:8(76) Macro(48) 1:03 1:87 1:82 1:92 3:79(24) capacitylosswheremacrocellhas4usersandeachfemtocellhasupto4users.InFig.3.10alldetectedmMSs,accordingtoMMLmethod,suerfromlowcapacitywhereasonly85%ofthedetectedusershavedesperatecapacityinMSCmethod.Thus,thefavorablemMSCDFcurvecanbedecidedbythemetricwhichfavorsonlythemMSswhodeeplysuer(e.g.MML(thr=0.5)).ThefMScapacitylossthatcorrespondstoFig.3.10isillustratedinFig.3.11wherethelossduetothereleasingOBisquiteconsiderable.Variousf;jvaluesarealsomarkedforthreemethodsinFig.3.11whereOBratioisaround25%.Itcanbeseenthatmarkedf;jinFig.3.11arenotexceedingthefvaluesinFig.3.7fortheOBratiois25%.37

PAGE 50

First,powercontrolhasalreadybeenemployedtomacrocellnetworkinordertomitigateso-callednear-fareectforcell-edgeusers[22].Infemtocellnetworks,dependingonthefMSlocation,fMSadaptsitstransmissionpowerinsuchawaythatitcausesminimalinterferencetothenetworkwhilecommunicatingwiththefBS.ThisadaptationallowsfMStoreduceintercellularinterference(amongthefemtocells)andintracellularinterference(betweenfemtocelltomacrocell). Second,spectrumresourceportioningisalsousedforinterferencemitigationintwotirednetworks.Itcanbedoneinthreeways:

PAGE 51

Third,accessingschemeusedinfemtocellnetworkdeploymentdirectlyaectstheinter-ferencelevelintheenvironment.Inclosedaccess(closedsubscribergroup(CSG))femtocell,thenumberofmMSscanjoinaparticularfemtocellnetworkislimited.So,anyuserthatisnotlistedaspermittedusercausesandreceiveshighinterferencefemtocell.Ontheotherhand,foropenaccess(OA)femtocellnetworks,anyuserequipment(UE)receivingbetterlinkqualityfromfemtocellisgrantedasfemtocellusersothatinterferencebetweenfemto-celltomacrocellisreduced.However,therearesomedrawbacksrelatedtoOAthatshouldbementioned[2,3,23] InthissectionOAnetworkisanalyzed.ForOAfemtocellnetwork,anyUEabovethecellselectioncriterionthresholdwillbegrantedassubscribertothebasestationofinterest.Cellselectioncanbedonebasedonreceivedsignalstrength(RSS),capacity,andinterferenceleveloftheusers. InOAfemtocelldeploymentscheme,comparedtoCSGmode,thefemtocellhasmoresubscriberssincesomeofthemMSshand-otothefemtocell.Althoughavailablebandwidth43

PAGE 52

wherePm;iisthereceivedpowerovertransmissionbandwidth(Bm;i=Btot=M)ofithmMSandN0representsthenoisepowerperhertz. Itcanbeconcludedfrom(4.1)thatcapacityimprovementcanbeachievedbyincreasingthetransmissionbandwidthofmMS(Bm;i)viaeitherhavingsmallerMorlargerBtotorincreasingthereceivedpower(Pm;i).However,RSSofthemBSobservedbyindoorusersissignicantlylowerthantheRSSexperiencedbyoutdoorusersduetothewallpenetrationlossasindepictedinFig.2.3whereindoorcoverageispoor.Femtocellscanberemedytoindoorusers'RSSproblembydeployingthemwithinthehouses/ocessothatRSSaswell44

PAGE 54

where~Bfisthebandwidthassignedtofemtocellnetworksand~M(~M
PAGE 55

whereIfisthetotalinterferenceobservedfromallclosebyfemtocellnetworks.Compar-ing(4.4)with(4.2),itisobservedthatthebandwidthavailableperuserimproveswithco-channeldeployment(i.e.,Bm;i>~Bm;i).However,themMSsalsoobserveinterferenceIffromnearbyfemtocellnetworks,whichmaydegradethecapacityifitissignicant.There-fore,whetherthecapacityimprovesornotwithrespecttoadedicatedchannelscenariodependsjointlyon~BfandIf.Ontheotherhand,typicallythereisonlylittleimprovementinaveragemacrocelluserbandwidthwhenco-channeldeploymentsisconsideredinsteadofdedicatedchanneldeployment.Hence,itistypicallythecasethatthecapacityofmacrocelluserswouldbelargerwithdedicatedchanneldeploymentsduetointerferenceproblems. Similarly,comparing(4.4)with(4.1),whetherthecapacityimprovesornotforco-channeldeploymentwithrespecttoano-femtocellscenariodependsjointlyonMandIfem.Ifmoreuserscanbeo-loadedtofemtocells,wehavesmallerM,whichmayoverweightheimpactofinterferenceandimprovethecapacityofremainingmacrocellusers. Ontheotherhand,thecapacityofanfMSuserwithco-channeldeploymentcanbewrittenasbCf;i=Bf whereBf=Btot~Bfem,whichimpliessignicantincreaseinavailablebandwidthperfemtocelluser.ThiscomesattheexpenseofinterferenceImobservedfrommacrocellusersandthemBS.Sincebandwidthaectsthechannelcapacitylinearly,andinterferenceaectsthechannelcapacitythroughthelogarithmfunction,asalsodiscussedbyseveralotherworkintheliterature([2,12,13,25])co-channeldeploymentsoffemtocellstypicallyresultin47

PAGE 56

betteroverallcapacitiescomparedtodedicatedchanneldeployments.TheseobservationswillalsobeveriedthroughcomputersimulationsinSection4.3.4.2OpenAccesswithLoadBalancing(OA-LB)

PAGE 57

However,interferenceandbandwidtharetwoothercriticalparametersaectingthecapacityofMSs,andshouldbealsotakenintoaccount[27].Forexample,forthemMSinFig.4.2,thelinkqualityfromFemtocell-AandthemBSmaybebetterthanthelinkqualityfromFemtocell-B;however,Femtocell-Aisoverloadedwithseveralusers,andthereisonlyasingleuserinFemtocell-B.Therefore,duetotheavailabilityoflargerspectrumresources,Femtocell-BmayprovidebettercapacitytothemMScomparedtoFemtocell-Aorthemacrocell.Thecapacity-maximizingcellselectionforthemMSmaysimplybeformulatedas1^i=argmaxiCm;i;(4.7) whereCm;idenotestheresultingcapacityofthemMSifitmakesahand-otocell-i.Notethat(4.7)doesnotonlytakethelinkqualityintoaccount,butalsotheavailablebandwidthandtheinterference.Theadvantagesof(4.7)include1)Thespectrumismorefairlydistributedamongmacrocellandfemtocellusers,2)Maximumnumberofusersthatmaybeconnectedtoacertainfemtocellwillbelowered,andtheburdenonthebackhauloffemtocellnetworkisdecreased. FortheexamplescenarioinFig.4.1(b),assumingequalbandwidthperfemtocelluserafterthemMShand-otothefemtocell,thecapacityofthefMSandmMSuserswithopenaccesscanberespectivelywrittenas

PAGE 58

Fig.4.3comparesthecapacitiesofindoorusersforthefollowingcases:withnoactivefemtocells,withfemtocellsoperatingindedicatedchannelmode,andwithfemtocellsoper-atinginco-channelmode.Allresultsareobtainedforwallloss(WL)of10dBand20dB.Theintroductionoffemtocells,forbothchannelmodes,presentsasignicantincreaseinthecapacityofindoorusers.Withahighersignalstrengthandmorebandwidth,thisincreaseincapacityisnotsurprising.Amajorfactorimpactingtheperformanceoffemtocellisthelevelofinterferencereceivedfromthemacrocell(incaseofco-channelmode)andfromneighboringfemtocells(inbothcases).Anotherimportantfactoristheavailablebandwidthforfemtocellusers,whichdependsonthenumberofusersservedbythefemtocelland,fordedicatedchannelmode,ontheallocatedportionofthespectrum.Inthissimulation,10%ofthespectrumallocatedtofemtocelloperatingindedicatedmode,andtheresultsinFig.4.3showaclearadvantagetoco-channelmodeoverdedicatedchannelmodeforindoorusers.50

PAGE 59

NotethatinFig.4.3,increaseinthewallpenetrationlossreducestheRSSfromthemBS.Thisresultsinadegradationinthecapacitywhennofemtocellsarepresent.Ontheotherhand,femtocellperformancebenetsfromhigherwallpenetrationlossasitreducestheinterferencefromotherfBSsandthemBS,andprovidesabetterisolationfortheindoorfemtocell.Also,astaircasetypeofbehaviorisobservedinthecapacityCDFofdedicatedchannelassignmentwithWL=10dB.ReasonforthisisthatforseveralindoorusersclosertothemBS,RSSfromthemBSmaystillbestrongerthantheRSSfromthefBSforsmallerWL.Therefore,withSINRbasedmetric,despitethescarcerspectrumresourcesatthemBS,someMSsmaystillprefertoconnecttothemBSevenwhentheyarelocatedindoors.ThisresultsinlowercapacitiesfortheseMSsanddeactivationoffemtocellsthatdonothaveanyremainingusers.51

PAGE 60

Fig.4.4comparesthecapacitiesofoutdoorusersforthesamecasesconsideredinthepreviousgure.Contrarytothecaseofindoorusers,thisgureshowsthatdedicatedchannelmodeoutperformsco-channelmodeforoutdooruserswithWL=20dB.ForthecasewhereWL=10dB,RSSofthesignalcomingfrommBSisstrongenoughtomuchmoreindoorusersregistertomBSwhichresultsinlesscapacity(almostsamecapacitytheonewithnofemtocellsarepresent)comparedtothecasewhereWL=20dB.Superiorityofdedicatedspectrumusagetoco-channelspectrumusagewithWL=20dBcomesasadirectresultoftheseparationofoperatingbandsoffemtocellsandthemacrocellwhichreducesinterferencecausedtooutdoorusers(whoaremainlycommunicatingwiththemacrocell)fromfemtocells.However,interferenceispresentinfemtocellsoperatinginco-channelmode,causingadecreaseinoutdooruserscapacity.Inbothcases,theincreaseinoutdoor52

PAGE 61

WL w/Femtow/FemtoUsers (dB) NoFemto(dc)(cc) All 20 1:1415:7665:42 Indoor 20 0:8517:2774:63 Outdoor 20 1:295:815:27 All 10 1:213:1963:29 Indoor 10 1:073:5072:03 Outdoor 10 1:291:302:87 users'capacityiscausedbytheextrabandwidthavailabletomacrocellusersasaresultofindoorusershandingotofemtocells.Forhigherwallpenetrationloss,thisgainoutweighsthelosscausedbyinterferenceinco-channelmode.Hence,anincreaseisnoticedinthecapacitywithWL=20dBoverWL=10dB.Withdedicatedchannelmode,thehigherwallpenetrationlossallowsmoreindooruserstohand-otofemtocellsleavingmoreavailablebandwidthtothemacrocellusers.Thisexplainsthegaininoutdooruserscapacitywithhigherpenetrationlossindedicatedchannelmodealthoughtheyarenotunderinterferencefromnear-byfemtocells. ThemediancapacitiesoftheresultsinFigs.4.3-4.4aresummarizedinTable4.1,whichconrmthat1)capacityofindoorusersmaybeimprovedsignicantlythroughdeployingfemtocells,especiallyforlargerWLandforco-channeldeployments,and2)capacityofoutdoorusersmayalsobeimprovedduetoo-loadingeect,especiallywithdedicatedchanneldeployments.4.3.2OpenAccesswithLoadBalancing(OA-LB)

PAGE 62

Figs.4.7-4.9illustratethecapacityofthemMSatdierentlocationsandfordierentdmBS.Best-capacityandbest-RSSmetricsarebothconsideredforcellselection,andclosedaccess(closedsubscribergroup(CSG))resultsarealsoincludedforcomparison.WhenthemBSisclosertothefemtocellsasinFig.4.7,theCSGcapacityofthemMSimprovesduetobettersignalquality.Ontheotherhand,thereisonlyminimalgainwithcapacity-maximizingcellselectionduetostrongerinterferencefromthemacrocelltothefemtocell.Still,Fig.4.7showsthatevenifthesignalqualityfromthemBSisbetter,themMSmayachievebettercapacitiesbyswitchingotoFemtocell-2atcertainlocations. ResultsinFig.4.8andFig.4.9showthatasthedistancebetweenthemBSandthefemtocellincreases,capacitymaximizingcellselectionyieldsbettercapacitiescompared54

PAGE 63

Fig.4.10showsthemeancapacityofmMSoverthetrajectoryinFig.4.5,fordierentdmBSandNf2values.ThemeanCSGcapacitydegradesingeneralsincetheRSSbecomes55

PAGE 68

ThreedierenttechniquesareproposedinChapter3forimprovingthecapacityofclosedaccesstypesoffemtocells.ThegainsareanalyzedthroughShannoncapacityformu-lationsandcomputersimulations.Fortheclosedaccessfemtocells,capacityimprovementisachievedthroughnotusingtheOBatafemtocellundercertaincircumstances.AlsoChapter3includesevaluationofbothoftheproposedtechniquesinsystem-levelsimula-tions,andconsideringrealisticspectrumsensingmethodsforassessingthepracticalityoftheproposedDSRapproach. InChapter4,performanceimprovementsthroughdedicatedchannelandco-channelfemtocelldeploymentsarestudiedthroughthehelpofchannelcapacityformulationsandcomputersimulations.Itisshownthatwhileco-channeloperationispreferablefromtheperspectiveofindoorusers,dedicatedchanneloperationispreferableforcausinglessinter-ferencetooutdoorusers.Furthermore,deploymentoffemtocellsimprovesthecapacityofnotonlyindoorusers,butalsooutdoorusersduetoo-loadingeect,especiallyforded-icatedchanneldeploymentswithlargerwallpenetrationloss.Moreover,aloadbalancingmethodisproposed,whereacapacity-basedmetricisusedforcell-selectionratherthanalink-qualitybasedmetric;thisallowsmorefairsharingofthespectrumresourcesamongthe60

PAGE 69

FutureworkincludestheimprovementsinthesimulationtoolaccordingtoparametersgiveninChapter2suchasintroducingthelog-normalshadowingtothebothmacrocellandfemtocellchannelsettings,implementingthesectorizationsuchthateverymacrocellhasthreesectorswithuniversalfrequencyreuse,andintroducingtheinterferencesfromothermacrocells.LTEsignalgeneration,transmission/receptionofthegeneratedsignal,andobtainingthebiterror(BER)curvesbasedonthereceivedsignalarealsoplanningasfuturework.Moreover,theoreticaloptimizationoftheproposedmethodsinChapter3(e.g.,optimizationofIoTandMMLthresholds)forvariousscenariosisconsideredasfuturework.61

PAGE 70

F.Chiussi,D.Logothetis,I.Widjaja,andD.Kataria,\Femtocells[GuestEditorial],"IEEECommunicationsMagazine,vol.48,no.1,pp.24{25,2010.[2] V.Chandrasekhar,J.G.Andrews,andA.Gatherer,\Femtocellnetworks:asurvey,"IEEECommun.Mag.,vol.46,no.9,pp.59{67,Sep.2008.[3] F.Chiussi,D.Logothetis,I.Widjaja,andD.Kataria,\Femtocells[GuestEditorial],"IEEECommunicationsMagazine,vol.47,no.1,pp.66{67,2009.[4] Airvana,\Femtocells:Personalbasestations,"2007,whitePaper.[Online].Available:http://www.airvana.com/les/Femton I.T..Media,\Femtocellmarketstatus,"Feb.2010,whitePaper.[6] \SimulationAssumptionsandParametersforFDDHeNBRFRequirements,"3GPP,3GPPTGSRANR4-092042,May2009.[7] NTTDOCOMO,\Tpfortr36.814onheterogeneousnetwork,"3GPPTSGRANWG1MeetingR1-101278,SanFrancisco,USA,pp.1{5,Feb.2010.[8] FemtoForum,\InterferencemanagementinUMTSfemtocells,"WhitePaper,Dec.2008.[Online].Available:http://www.femtoforum.org/femto/Files/File/InterferenceManagementinUMTSFemtocells.pdf[9] I.Demirdogen,I.Guvenc,andH.Arslan,\Capacityofclosed-accessfemtocellnetworkswithdynamicspectrumreuse,"insubmittedtoProc.IEEEInt.Symp.Personal,In-door,MobileRadioCommun.(PIMRC),Istanbul,Turkey,Sep.2010,pp.1{6.[10] \3rdGenerationPartnershipProject;TechnicalSpecicationGroupRadioAccessNet-works;3GHomeNodeBStudyItemTechnicalReport(Release8),"3GPP,3GPPTR25.820,March2008.[11] S.P.Yeh,S.Talwar,S.C.Lee,andH.Kim,\WiMAXfemtocells:aperspectiveonnetworkarchitecture,capacity,andcoverage,"IEEECommun.Mag.,vol.46,no.10,pp.58{65,Oct.2008.[12] D.L.Perez,A.Valcarce,G.D.L.Roche,E.Liu,andJ.Zhang;,\AccessmethodstoWiMAXfemtocells:Adownlinksystem-levelcasestudy,"inProc.IEEEInt.Conf.Commun.Syst.(ICCS),Guangzhou,China,Nov.2008,pp.1657{1662.62

PAGE 71

H.Claussen,\Performanceofmacro-andco-channelfemtocellsinahierarchicalcellstructure,"inProc.IEEEInt.Symp.Personal,Indoor,MobileRadioCommun.(PIMRC),Athens,Greece,Sep.2007,pp.1{5.[14] NTTDOCOMO,\DownlinkInterferenceCoordinationBetweeneNodeBandHomeeNodeB,"3GPPTSGRANWG1MeetingR1-101225,SanFrancisco,USA,pp.1{8,Feb.2010.[15] Y.Bai,J.Zhou,andL.Chen,\Hybridspectrumsharingforcoexistenceofmacrocellandfemtocell,"inIEEEInternationalConferenceonCommunicationsTechnologyandApplications,Beijing,Oct.2009,pp.1{5.[16] Y.-Y.Li,M.Macuha,E.S.Sousa,T.Sato,andM.Nanri,\Cognitiveinterferencemanagementin3gfemtocells,"inProc.IEEEInt.Symp.Personal,Indoor,MobileRadioCommun.(PIMRC),Tokyo,Japan,Sep.2009,pp.1{5.[17] NTTDOCOMO,\ViewsonDeploymentScenariosinHeterogeneousNetworks,"3GPPTSGRANWG1MeetingR1-094916,Tokyo,Japan,pp.1{4,Nov.2009.[18] \Guidelinesforevaluationofradiointerfacetechnologiesforimt-advanced,"ITU-RReportM.2135,Nov.2008.[19] \3rdGenerationPartnershipProject;TechnicalSpecicationGroupRadioAccessNet-works;FurtherAdvancementsforE-UTRAPhysicalLayerAspects(Release9),"3GPP,3GPPTR36.814,Nov.2009.[20] \ConsiderationsonInterferenceCoordinationinHet-Net,"3GPPTSGRANWG1MeetingR1-100902,SanFrancisco,USA,Feb.2010.[21] S.Sesia,I.Touk,andM.Baker,LTE,TheUMTSLongTermEvolution:FromTheorytoPractice.WileyPublishing,2009.[22] M.Yavuz,F.Meshkati,S.Nanda,A.Pokhariyal,N.Johnson,B.Raghothaman,andA.Richardson,\InterferencemanagementandperformanceanalysisofUMTS/HSPA+femtocells-[femtocellwirelesscommunications],"IEEECommunicationsMagazine,vol.47,no.9,pp.102{109,2009.[23] D.Lopez-Perez,A.Valcarce,G.DeLaRoche,J.Zhang,etal.,\OFDMAfemtocells:aroadmaponinterferenceavoidance,"IEEECommunicationsMagazine,vol.47,no.9,pp.41{48,2009.[24] I.Guvenc,M.R.Jeong,F.Watanabe,andH.Inamura,\Ahybridfrequencyassignmentforfemtocellsandcoverageareaanalysisforco-channeloperation,"IEEECommun.Lett.,vol.12,no.12,pp.880{882,Dec.2008.[25] V.ChandrasekharandJ.G.Andrews,\Uplinkcapacityandinterferenceavoidancefortwo-tiercellularnetworks,"inProc.IEEEGlobalTelecommun.Conf.(GLOBECOM),Washington,DC,Nov.2007,pp.3322{3326.[26] K.Y.Lin,H.P.Lin,andR.T.Juang,\ProposedtextforHOfromfemtocellBStomacroBSorotherfemtocellBS(AWD-femto),"IEEEStandardContributionC802.16m-09/1307,July2009.63

PAGE 72

I.Demirdogen,I.Guvenc,andH.Arslan,\Capacityanalysisofopen-accessfemtocellnetworks,"insubmittedtoProc.IEEEInt.Symp.Personal,Indoor,MobileRadioCommun.(PIMRC)Workshop,Istanbul,Turkey,Sep.2010.64


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003467
035
(OCoLC)
040
FHM
c FHM
049
FHMM
090
XX9999 (Online)
1 100
Demirdogen, Ibrahim.
0 245
Comparison and improvement of different access methods in femtocell networks
h [electronic resource] /
by Ibrahim Demirdogen.
260
[Tampa, Fla] :
b University of South Florida,
2010.
500
Title from PDF of title page.
Document formatted into pages; contains X pages.
502
Thesis (M.S.E.E.)--University of South Florida, 2010.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
3 520
ABSTRACT: A variety of wide band wireless systems have been pushed towards their limits in order to meet growing interest for high data rate in wireless communications.In particular, the limit due to the spectrum scarcity forces communication systems to utilize the spectrum resource at maximum efficiency level. One of the methods that allow effective spectrum employing is to cover multiple systems over same spectrum source by allowing bearable interference to occur between them. Femtocells have been recently introduced as a remedy to spectrum scarcity and coverage problems in current cellular structures. Femtocells are personal use base stations and they share the spectrum in a way that they can coexist with the macrocell. This thesis provides a critical reviews of different access methods in femtocell networks and further introduces improvements related to these access methods. Simulation results validate capacity improvement of proposed techniques compared to the existing access methods.
590
Advisor: Huseyin Arslan, Ph.D.
653
Femtocell
Open Access
Closed Access
Closed Subscriber Group
690
Dissertations, Academic
z USF
x Electrical Engineering
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.3467