首页 Analog Circuit Design (2009).pdf

Analog Circuit Design (2009).pdf

Analog Circuit Design (2009).pdf

上传者: 上攻下伐 2012-05-31 评分1 评论0 下载105 收藏0 阅读量662 暂无简介 简介 举报

简介:本文档为《Analog Circuit Design (2009)pdf》,可适用于电信技术领域,主题内容包含AnalogCircuitDesignArthurHMvanRoermund•HermanCasierMichielSteyaertEditorsA符等。

AnalogCircuitDesignArthurHMvanRoermund•HermanCasierMichielSteyaertEditorsAnalogCircuitDesignSmartDataConverters,FiltersonChip,MultimodeTransmittersABCEditorsDrArthurHMvanRoermundDepartmentofElectricalEngineeringEindhovenUniversityofTechnologyMBEindhovenNetherlandsahmvroermundtuenlDrHermanCasierAvondsterKuurneBelgiumhermancasierieeeorgProfMichielSteyaertDepartmentofElectricalEngineering(ESAT)KatholiekeUniversiteitLeuvenKasteelparkArenbergLeuvenBelgiummichielsteyaertesatkuleuvenbeISBNeISBNDOISpringerDordrechtHeidelbergLondonNewYorkLibraryofCongressControlNumber:cNopartofthisworkmaybereproduced,storedinaretrievalsystem,ortransmittedinanyformorbyanymeans,electronic,mechanical,photocopying,microfilming,recordingorotherwise,withoutwrittenpermissionfromthePublisher,withtheexceptionofanymaterialsuppliedspecificallyforthepurposeofbeingenteredandexecutedonacomputersystem,forexclusiveusebythepurchaseroftheworkCoverdesign:eStudioCalamarSLPrintedonacidfreepaperSpringerispartofSpringerScienceBusinessMedia(wwwspringercom)SpringerScienceBusinessMediaBVPrefaceThisbookispartoftheAnalogCircuitDesignseriesandcontainscontributionsofthespeakersofthethworkshoponAdvancesinAnalogCircuitDesign(AACD),whichwasorganizedbySvenMattissonofEricssonTheworkshopwasheldinLund,Sweden,fromMarchtoApril,Thebookcomprisesthreeparts,coveringadvancedanalogandmixedsignalcircuitdesignfieldsthatareconsideredasveryimportantbythecircuitdesigncommunity:SmartDataConvertersFiltersonChipMultimodeTransmittersEachpartissetupwithsixpapersfromexpertsinthefieldTheaimoftheAACDworkshopistobringtogetheragroupofexpertdesignerstodiscussnewdevelopmentsandfutureoptionsEachworkshopisthenfollowedbythepublicationofabookbySpringerintheirsuccessfulseriesofAnalogCircuitDesignThisbookisnumberinthisseriesThebookscanbeseenasareferenceforallpeopleinvolvedinanalogandmixedsignaldesignThefulllistofthepreviousbooksandtopicsintheseriesisgivennextWeareconfidentthatthisbook,likeitspredecessors,providesavaluablecontributiontoouranalogandmixedsignalcircuitdesigncommunityArthurvanRoermundThetopicscoveredbeforeinthisseries:Pavia(Italy)HighspeedClockandDataRecoveryHighperformanceAmplifiersPowerManagementOostende(Belgium)Sensors,ActuatorsandPowerDriversfortheAutomotiveandIndustrialEnvironmentIntegratedPAsfromWirelinetoRFVeryHighFrequencyFrontEnds(continued)vviPreface(continued)Maastricht(TheNetherlands)HighspeedADConvertersAutomotiveElectronics:EMCIssuesUltraLowPowerWirelessLimerick(Ireland)RFCircuits:WideBand,FrontEnds,DACsDesignMethodologyandVerificationofRFandMixedSignalSystemsLowPowerandLowVoltageMontreux(Swiss)SensorandActuatorInterfaceElectronicsIntegratedHighVoltageElectronicsandPowerManagementLowPowerandHighResolutionADCsGraz(Austria)FractionalNSynthesizersDesignforRobustnessLineandBusdriversSpa(Belgium)StructuredMixedModeDesignMultiBitSigmaDeltaConvertersShortRangeRFCircuitsNoordwijk(TheNetherlands)ScalableAnalogCircuitsHighSpeedDAConvertersRFPowerAmplifiersMunich(Germany)HighSpeedADConvertersMixedSignalDesignPLLsandSynthesizersNice(France)XDSLandotherCommunicationSystemsRFMOSTModelsandBehaviouralModellingIntegratedFiltersandOscillatorsCopenhagen(Denmark)VoltElectronicsMixedModeSystemsLNAsandRFPowerAmpsforTelecomComo(Italy)RFADConvertersSensorandActuatorInterfacesLowNoiseOscillators,PLLsandSynthesizersLausanne(Swiss)RFCMOSCircuitDesignBandpassSigmaDeltaandOtherDataConvertersTranslinearCircuitsVillach(Austria)LowNoisePowerVoltageMixedModewithCADToolsVoltage,CurrentandTimeReferencesEindhoven(Netherlands)LowPowerLowVoltageIntegratedFiltersSmartPowerLeuven(Belgium)MixedModeADDesignSensorInterfacesCommunicationCircuitsScheveningen(TheNetherlands)OpAmpsADCAnalogCADContentsPartISmartDataConvertersLMSBasedDigitalAssistingforDataConvertersBangSupSongPipelinedADCDigitalCalibrationTechniquesandTradeoffsImranAhmedHighResolutionandWideBandwidthCMOSPipelineADConvertersHansVandeVelASignalProcessingViewonTimeInterleavedADCSChristianVogelDACCorrectionandFlexibility,Classification,NewMethodsandDesignsGeorgiRadulov,PatrickQuinn,HansHegt,andArthurvanRoermundSmartCMOSCurrentSteeringDAConvertersforEmbeddedApplicationsMartinClara,DanielGruber,andWolfgangKlatzerPartIIFiltersOnChipSynthesisofLowSensitivityAnalogFiltersLarsWanhammarHighPerformanceContinuousTimeFilterswithOnChipTuningJoseSilvaMartinezandAydınIKarsılayanviiviiiContentsSourceFollowerBasedContinuousTimeAnalogFiltersStefanoD’Amico,MarcelloDeMatteis,andAndreaBaschirottoReconfigurableActiveRCFilterswithHighLinearityandLowNoiseforHomeNetworkingApplicationsJanVandenbussche,JanCrols,andYuichiSegawaOnChipInstantaneouslyCompandingFiltersforWirelessCommunicationsVaibhavMaheshwariandWouterASerdijnBAWICCOIntegrationTunableFiltersatGHzFrequenciesAndreiaCathelin,StephaneRazafimandimby,andAndreasKaiserPartIIIMultimodeTransmittersMultimodeTransmitters:EasierwithStrongNonlinearityEarlMcCuneRBSHighEfficiencyPowerAmplifierResearch–ChallengesandPossibilitiesBoBerglund,UlfGustavsson,JohanThoreback,ThomasLejon,andEricssonABMultiModeTransmittersinCMOSManelCollados,XinHe,JanvanSinderen,andRafRooversChallengesforMobileTerminalCMOSPowerAmplifiersPatrickReynaertMultimodeTransmitterswith†BasedAllDigitalRFSignalGenerationAFrappe,AKaiser,AFlament,andBStefanelliSwitchedModeTransmitterArchitecturesHenrikSjoland,CarlBryant,VandanaBassoo,andMikeFaulknerPartISmartDataConvertersThefirstpartofthisbookcoversthetheme‘SmartDataConverters’Asthenameindicates,itdealswithsmartconvertersthathavesomekindofsmartnessimplementedonchip,tomaketheconverterbetterinperformanceforagivenamountofresourceslikepowerdissipationandareaOnchipsmartnessmightalsoresultinanincreaseinyield,adecreaseindesigneffort,ahigherflexibility,morefunctionalityandorbroaderapplicabilityAlltheseaspectsinturnalsopayoffinlesscostThePartstartswithADconvertersThreetypesofADconvertersachieveconsiderableattentionnowadays,andarethereforeaddressedhere:pipelined,SigmaDelta,andtimeinterleavedADconvertersThefirstpaperdiscussesbothLMSbasedcalibratedpipelineandSigmaDeltaconvertersandalsomakessomecomparisonsbetweenthetwoThesecondpaperfullyfocusesonpipelineconvertersandaddressesseveralcalibrationtechniquesThethirdpaperdiscussesacalibratedpipelineintheapplicationcontextofamultichannel,andthuswideband,frontendofacellularbasestationNextweproceedwithapaperontimeinterleavedconvertersHeretheproblemisintheequalityofthechannelsintermsofgain,time,andmoregenericallyseen:inspectralbehaviourThispaperwilladdresstheproblemfromasignalprocessingpointofview,sofromahigherlevelofabstraction,toshowwhattheoreticalapproachesarepossibletocorrectforlowerlevelinducedchanneldifferences,andwhatarethetradeoffsbetweenthem,onanalgorithmiclevelFinallyweendupwithtwoDApapersThefirstonegivesanoverviewandclassificationofsmartapproachesforCurrentSteeringDAs,astheyareknownnowinliterature,showssolutionsformissingapproaches,andaddressesflexibilityasoneofthefeaturesofsmartconvertersThesecondDApaperalsoaddressesCurrentSteeringDAs,butfocusesmorespecificallyontheembeddingofthesekindsofconvertersinsystemsonchip(SoCs),whichimpliessomeextraconstraintsthatshouldbemetArthurvanRoermundChapterLMSBasedDigitalAssistingforDataConvertersBangSupSongAbstractAggressivedevicescalingdowntothenanometerrangeoffersICdesignersbothopportunitiesandchallengesDigitaldesignersbenefitgreatlyfromthesystemflexibilityandaffordability,butanalogRFdesignersarestrugglingwithflaweddevicesSincescaleddevicesarefasterandsmaller,theincentivetousesuchstrengthsadvantageouslyhaspromptedmanyeffortstoovercomeanalogimperfectionbydigitalmeansDesignersareintroducingmoreDSPfunctionalitytoenhancetheperformanceofanalogRFsystemsMoreintelligenceisbeingbuiltintoanalogRFdesignsasinlinearPA,RFreceiverfrontend,ADCDAC,digitalPLL,etcSuchpervasivedesigntechniqueswithdigitalassistingwillprevailinthefutureSOCdesignAfterabriefoverviewofthetrend,examplesoftheLMSbasedcalibrationalgorithmappliedtothepipelineandCTcascaded†modulatorarediscussedIntroductionCMOSanalogdesignhasevolvedalongwiththedevicescalingforthreedecadessinceearlysInitsearlydays,thesupplyvoltagewashigher,theopamphadhighgainwhiledeviceswereslow,andthecrudelithographylimitedthecapacitormatchingonlyto–blevelThetwostageopampandthesimpleSARwerepredominantlyusedatlowsofkHzrangemostlyforthevoicebandprocessingThe†modulatorwasfeasible,butdigitalfilteringwasverycostlyThischangedinsasCMOSwasaggressivelyscaleddowntowardsthesubmicronrangeInthismiddleperiod,thesupplyvoltagewasloweredfromhigh–to–V,anddeviceswerefastenoughtodigitizethevideobandandbeyondTwoADCarchitecturesstoodout–pipelineforhighspeedcommunicationsandvideo,and†forhighresolutionaudioCascadedsinglestageopampwasadopted,andmanyADCcalibrationtechniquesweredevelopedtoenhancetheresolutionofthepipelinedBSSong()DepartmentofElectricalandComputerEngineering,UniversityofCalifornia,SanDiego,USAemail:songeceucsdeduAHMvanRoermundetal(eds),AnalogCircuitDesign:SmartDataConverters,FiltersonChip,MultimodeTransmitters,DOI,cSpringerScienceBusinessMediaBVBSSongADCtoabovebrangeNowins,CMOSisstillbeingscaleddownfromthesubmicrontothenanometerrange,andthesupplyvoltagealsoapproachessubVTherealadvantagesofsuchscaleddevicesarerawspeed,finelithography,andalmostfreedigitalcircuitryThefinelinelithographyalsomadethebarecapacitormatchingofblevelfeasibleThesedays,analogengineersstartwithfasterandmoreaccuratedevicesthanearliergenerationsdid,andmostdesignsturnouttobealreadyhighspeedandhighresolutionwithlowpowerHowever,acoupleofproblemsshouldbedealtwithWithlowsupplyvoltages,SNRislimitedbythesignalswing,andthelowgaindefeatsanydesignefforttousetheconventionalanalogdesignwisdomaccumulatedoverdecadesInaddition,thedeviceleakagemakesanyaccurateswitchedcapacitordesigndifficultInfact,itappearsthattheanalogdesigntrendisreset,anditstartsoveragainfromthebeginningTwoormultistageopampsareback,buttheirgainisstilllowandnonlinearOldADCdesignssuchasalgorithmic,SAR,andtimeinterleavingarealsobeingrevisitedInordertoavoidusinglowgainnonlinearopamps,thenewbreedsofADCarchitecturesthatusenoopampsstartedtoemergeExamplesarecomparatorbasedpipelineADCsandquantizersbasedontimeresolutionOntheotherhand,theindustryhasgrownwiththepowerfulbroadbanddigitalprocessingthatenablesSOCssuchascellphone,WiFi,TVtuner,:::ThisnewenvironmenthascreatedademandforwidebandADCssuchasIFquantizerswithveryhighSFDRtofacilitatethedigitalchannelfilteringafterquantizingthedesiredspectrumwithlargeblockerchannelsAlsoforhighresolutiongraphicorimaging,highSNRoverdBandlowlevellinearityoverbatsamplingratesoverMSsarerequiredtoresolveevendarkimagesfurtherinmoredetailsItischallengingtomeetsuchdemandswithscaledlowvoltageCMOSTwohighresolutionADCarchitecturesthatcanmeetsuchhighdemandsarethecalibratedpipelinedADCandtheCT†modulatorTheformerisnowwellestablishedenoughtocalibrateeventheopampnonlinearityThelatterexhibitsmanydesirablefeaturesinwirelessapplicationsandgainsmomentumasitrequiresnoantialiasing,andSNRisimprovednotbythecalibrationaccuracybutbythefeedbackInthefollowingsections,afterhighresolutionADCsandtheirfundamentallimitsareoverviewed,anLMSbasedresolutionenhancingtechniqueisintroduced,whicheliminatestheresidualerroraftercalibrationusingthezeroforcingLMSservofeedbackconceptHighResolutionADCsHighresolutionADCssamplingat–MSswith–blinearityhavebeenimplementedmostlywithSAR,†,orpipelinearchitecturesasshownintheresolutionspectrumofFigTheSARisverydesirableforlowvoltageandlowpowerapplicationssinceitusesonlyonecomparatorHowever,thepipelineoffersasignificantspeedadvantagewhilethe†ismorerobustinachievinghighresolutionHighresolutionADCsathighsamplingratesareonlyfeasiblewithscaledLMSBasedDigitalAssistingforDataConvertersFigResolutionvsbandwidthofADCsHighResolutionApplicationstechnologywithlowsupplyvoltages,andtheirperformanceiscommonlycharacterizedbytheirlinearitymeasuredbySFDRorTHDSuchADCswithhighlinearitybutpoorSNRareallowedinsystemsperformingdigitalfilteringTheearliestefforttoenhancetheADCresolutionwasanEPROMbasedcodemappingtechniqueusingaradix<,whichwarrantsmonotonicityandproperaddressingHowever,itwaspossibleonlyatfactorysinceitrequiredexternalprecisioninstrumentsThefirstselfcalibrationconceptfortheSARwasintroducedtomeasurecapacitormismatcherrors,tostorethemdigitally,andtosubtractthemduringthenormaloperation,ThisselfcalibratedSARbasedonthechargeredistributioncapacitorarraywasslow,andtheoversamplingADCcoveredthevoiceoraudiobandbetterAlsoonecriticalflawofthehighresolutionSARwastheslowlyvaryingoffsetofthecomparatorduetothestressinflictedupontheinputdifferentialpairofthecomparatorwhenseveraldecisionsaremaderepeatedlyafteroneinputsamplingFinally,theNyquistrateADCabovethevideobandbecamearealitywhenthepipelinedarchitecturewasintroduced,andthecapacitorarrayMDACasaresidueamplifierenabledthedevelopmentofhighresolutionADCs–TheswitchedcapacitorMDACperformsmultiplefunctionsofsampling,DACsubtraction,andamplificationasaresidueamplifierinthepipelinedADCorasanintegratorintheDT†modulatorFigurecomparestheswitchedcapacitorMDACwiththeCTintegratorTheformerisusedinanopenendedsystem,andtheresidueamplifiershouldsettlewithanabsoluteaccuracyHowever,thelatterrestsinsidethefeedbackloop,anditsgainandnonlinearityerrorsarereducedbytheloopgainOnecriticalfactortoconsideratthesystemlevelistheantialiasingrequirementNyquistrateADCsneedhighorderantialiasingfilterswhenoperatedatclosetotheNyquistratewhileCT†modulatorsneednoantialiasingatallThespeedadvantageofthepipelinedADCoverthe†modulatorhasalwaysbeenbyafactorofto,butthegapwasquicklynarrowedastechnologywasscaledAgoodexampleisthefirstdigitallycalibratedMSs,bADCproduct(MAX)overtakenbytheBSSongFigPipelinevsCT†modulatorPipelineMDACCTModulatorResidueampinopenloopIntegratorinfeedbackHighopampgainLowopampgainGainerrorNogainerrorReducedbyresiduegainReducedbyloopgainDACmismatcherrorDACmismatcherrorAbsolutesettlingLinearsettlingTolerableoffsetOffsetincorrectionrangeAntialiasingfilterNoantialiasingfilterΔΣ†ADCItalsohappenedearlierinswhenthe†modulatorreplacedtheselfcalibratedSARasaudiocodersEventoday,thesamecompetitionbetweenthepipelinedADCandthe†modulatorstillcontinuesThecommonthemeinthiscompetitionforthebestisnowcalibrationTheCT†modulatoralsoneedscalibrationastheoversamplingratioisloweredto–approachingtheNyquistrateforhighspeedoperationAllearliercalibrationwasdoneintheanalogdomainalthoughmeasurederrorswerestoreddigitallyAnefforttoperformtheerrorsubtractioninthedigitaldomainledtothedigitalcalibrationconcept,,buterrormeasurementswerestillperformedinaseparatemeasurementcycleThetermsuchasforegroundorbackgroundisuseddependingonhowtheerrormeasurementisperformed–Thelatestbackgrounderrormeasurementtechniquehasevolvedintoaverysophisticatedone,calledPNditheringThePNsequenceisapseudorandombinarypulsesequencewithanequalprobabilityoforoveralongsampleperiodItwasusedforthepulsemodulationfortheradarjammingduringtheWorldWarII,andalsoforthemilitarysecuritycommunicationsknownasSpreadSpectrumandGlobalPositioningSystem,whicharenowwellknownascommercialsystemssuchasCDMAandGPSThefirstexampleofusingthePNsequencetoenhancetheADCresolutionwastodithertheADCforlowDNL,a

职业精品

现金采购管理制度.doc

材料采购及出入库管理制度.doc

日常采购管理流程.doc

采购管理.ppt

用户评论

0/200
    暂无评论
上传我的资料

精彩专题

相关资料换一换

  • Analog Circuit D…

  • Analog circuit d…

  • Analog Circuit D…

  • Analog Circuit D…

  • Analog Circuit D…

  • Analog Circuit D…

  • Analog_Circuit_D…

  • Analog Circuit D…

  • Analog_Circuit_D…

资料评价:

/ 351
所需积分:1 立即下载

意见
反馈

返回
顶部