简介:Selectionoftestcasesplaysakeyroleinimprovingtestingefficiency.Black-boxtestingisanimportantwayoftesting,anditsvalidityliesontheselectionoftestcasesinsomesense.Areasonableandeffectivemethodabouttheselectionandgenerationoftestcasesisurgentlyneeded.Thisletterfirstintroducessomeusualmethodsonblack-boxtestcasegeneration,thenproposesanewalgorithmbasedoninterfaceparametersanddiscussesitsproperties,finallyshowstheeffectivenessofthealgorithm.
简介:WehavemeasuredthedeepenergyleveloftheInP:Fewhichissemi-insulatorthroughthemethodofOTCS.TheeffectoflightintensityonOTCSmeasurementismainlydiscussed.ThereareelectrontrapofET=0.034eVandholetrapofET=1.13eVinInP:Feunderthestronglightandlowtemperature.ThelocationoftheOTCSpeakofelectrontrap(ET=0.34eV)movestowardsthedirectionofhightemperaturer,whenthelightintensitywasincreased,ETisdifferentunderdifferentlightintensity.Itiscorrectedintermsoftheorythatthestuffratioofthedeepenergylevelisaffectedbythelightintensity.Theexperimentsshowthattheerrorisdecreasedgreatlywiththecorrection.
简介:Inthispaper,wepresentanewdeformablemodelforshapesegmentation,whichmakestwomodificationstotheoriginallevelsetimplementationofdeformablemodels.Themodificationsaremotivatedbydifficultiesthatwehaveencounteredinapplyingdeformablemodelstosegmentationofmedicalimages.Thelevelsetalgorithmhassomeadvantagesovertheclassicalsnakedeformablemodels.However,itcoulddeveloplargegapsintheboundaryandholeswithintheobjects.Suchboundarygapsandholesofobjectscancauseinaccuratesegmentationthatrequiresmanualcorrection.Theproposedmethodinthispaperpossessesaninherentpropertytodetectgapsandholeswithintheobjectwithasingleinitialcontourandalsodoesnotrequirespecificinitialization.Thefirstmodificationistoreplacetheedgedetectorbysomeareaconstraint,andthesecondmodificationutilizesweightedlengthconstrainttoregularizethecurveunderevolution.Theproposedmethodhasbeenappliedtobothsyntheticandrealimageswithpromisingresults.
简介:Thispaperstudiesthelinkageproblembetweentheresultofhigh-levelsynthesisandback-endtechnology,presentsamethodofhigh-leveltechnologymappingbasedonknowledge,andstudiesdeeplyallofitsimportantlinkssuchasknowledgerepresentation,knowledgeutilityandknowledgeacquisition.Itincludes:(1)presentakindofexpandedproductionaboutknowledgeofcircuitstructure;(2)presentaVHDL-basedmethodtoacquireknowledgeoftechnologymapping;(3)providesolutioncontrolstrategyandalgorithmofknowledgeutility;(4)presentahalf-automaticmaintenancemethod,whichcanfindredundanceandcontradictionofknowledgebase;(5)presentapracticalmethodtoembedthealgorithmintoknowledgesystemtodecreasecomplexityofknowledgebase.Asystemhasbeendevelopedandlinkedwiththreekindsoftechnologies,soverifiedtheworkofthispaper.
简介:Aimingatthedefectsofthenodesinthelowenergyadaptiveclusteringhierarchy(LEACH)protocol,suchashighenergyconsumptionandunevenenergyconsumption,atwo-levellinearclusteringprotocolisbuilt.Theprotocolimprovesthewayofthenodesdistributionatrandom.Theterminalnodeswhichhavenotbeenatwo-levelclusterheadintheclustercancompetewiththeprincipleofequivalentpossibility,andonthebasisoftherestenergyofnodesthetwo-levelclusterheadisselectedatlast.Thesinglehopwithintheclusterandsinglehopormultiplehopsbetweenclustersareused.Simulationexperimentresultsshowthattheperformanceofthetwo-levellinearclusteringprotocolappliedtotheHexicorridoragriculturalfieldissuperiortothatoftheLEACHprotocolinthesurvivaltimeofnetworknodes,theratioofsuccess,andtheremainingenergyofnetworknodes.
简介:Inthemodernanaloguedesign,TransistorLevelFaultSimulation(TLFS)playstheim-portantpartsinceeveryfaultinthewholecircuithastobesimulatedatthatlevel.Unfortunately,itisaveryCPUintensivetaskeventhoughitmaintainsthehighaccuracy.Therefore,HighLevelFaultModeling(HLFM)andHighLevelFaultSimulation(HLFS)arerequiredinordertoalleviatetheeffortsofsimulation.Inthispaper,differentHLFMapproachesarereviewedatthedevicelevelduringlasttwodecades.Weclarifytheirdomainsofapplicationandevaluatetheirstrengthsandcurrentlimitations.Wealsoanalyzecausesoffaultsandintroducevarioustestapproaches.
简介:Theresidualelectricallyactivedefectsin(4×1012cm-2(30KeV)+5×1012cm-2(130KeV))si-implantedLECundopedsi-GaAsactivatedbytwo-steprapidthermalannealing(RTA)LABELEDAS970℃(9S)+750℃(12S)havebeeninvestigatedwithdeepleveltransientspec-troscopy(DLTS).TwoelectrontrapsET1(Ec-0.53eV,σn=2.3×10-16cm2)andET2(Ec-0.81eV,σn=9.7×10(-13)cm2)aredetected.Furthermore,thenoticeablevariationsoftrap’scon-centrationandenergylevelintheforbiddengapwiththedepthprofileofdefectsinducedbyionim-plantationandRTAprocesshavealsobeenobserved.The[Asi·VAs·AsGa]and[VAs·Asi·VGa·AsGa]areproposedtobethepossibleatomicconfigurationsofET1andET2,respectivelytoexplaintheirRTAbehaviors.
简介:Throughexperimentsandcomputersimulation,theinfluenceoftheenergylevelsoforganicmaterialsandelectrodematerialsintheorganiclight-emittingdiodes(OLEDs)onthedeviceperformancesisdiscussed.Resultsshowthatthedeviceperformancesareinfluencedbynotonlythecarrierinjectionbarriersattheelectrodeinterfacebutalsothebarriersattheorganicheterojunctioninterface.ThisresultishelpfultotheselectionoftheorganicmaterialsandtheirarrangementintheoptimaldesignofOLEDs.
简介:Doubledataratesynchronousdynamicrandomaccessmemory(DDR3)hasbecomeoneofthemostmainstreamapplicationsincurrentserverandcomputersystems.Inordertoquicklysetupasystem-levelsignalintegrity(SI)simulationflowfortheDDR3interface,twosystem-levelSIsimulationmethodologies,whichareboard-levelS-parameterextractioninthefrequency-domainandsystem-levelsimulationassumptionsinthetimedomain,areintroducedinthispaper.BycomparingtheflowofSpeed2000andPowerSI/Hspice,PowerSIischosenfortheprintedcircuitboard(PCB)board-levelS-parameterextraction,whileTektronixoscilloscope(TDS7404)isusedfortheDDR3waveformmeasurement.Thelabmeasurementshowsgoodagreementbetweensimulationandmeasurement.ThestudyshowsthatthecombinationofPowerSIandHspiceisrecommendedforquicksystem-levelDDR3SIsimulation.
简介:Themicroscopicmechanismoflowlevellaserirradiation(LLLI)mendingrheologyforbloodisanalyzedbythequantumtheory.Theresultsshowthatthelasercanresolvethefibrinclot,sothepropertyofbloodrheologyisimproved,andsomebondsofthecholesterolinbloodarefracturedbythelowlevellaser(LLL).Hencetheratioofmembranecholesteroltomembranephosphatideoftheredcellisreduced,andthenthebloodcirculationcanbeimproved.Theclinicalcuredtestsandsomecorrelativetestdatacanbeexplainedbythetheoreticalanalysesinthispaper.
简介:Accordingtothedipoletransitionselectionrulesandbychoosingpropertransitionchannels,itisfoundthattheinvertedY-typefour-levelschemecanbeclassifiedintoninedifferent2nd-ordersubschemeswhenusingcorrespondinglaserfields,andtheycanbefurtherclassifiedintoeighty3rd-ordersubschemesindetailaccordingtodifferentexcitingintensitiesthatareappliedindifferentchannels.Inthispaper,thestructuremultiformityofinvertedY-typefour-levelschemeisdiscussedandthedemonstrationsonsomesubschemesarealsogiven.Atthesametime,thesubschemeswhichhavebeenstudiedarepointedout,butmostofthemhavenotbeenresearchedtothebestofourknowledge.
简介:Withtheincreaseoftheclockfrequencyandsiliconintegration,powerawarecomputinghasbecomeacriticalconcerninthedesignoftheembeddedprocessorandsystem-on-chip(SoC).Dynamicvoltagescaling(DVS)isaneffectivemethodforlow-powerdesigns.However,traditionalDVSmethodshavetwodeficiencies.First,theyhaveaconservativesafetymarginwhichisnotnecessaryformostofthetime.Second,theyareexclusivelyconcernedwiththecriticalstageandignorethesignificantpotentialfreeslacktimeofthenoncriticalstage.Thesefactorsleadtoalargeamountofpowerwaste.Inthispaper,anovelpipelinestructurewithultra-lowpowerconsumptionisproposed.Itcutsoffthesafetymarginandtakesuseofthenoncriticalstagesatthesametime.Aprototypepipelineisdesignedin0.13mtechnologyandanalyzed.Theresultshowsthatalargeamountofenergycanbesavedbyusingthisstructure.Comparedwiththefixedvoltagecase,50%oftheenergycanbesaved,andwithrespecttothetraditionaladaptivevoltagescalingdesign,37.8%oftheenergycanbesaved.
简介:Classifyingthetextureofgranulesin2Dimageshasarousedmanifoldresearchatten-tionforitstechnicalchallengesinimageprocessingareas.ThisletterpresentsanaggregatetextureidentificationapproachbyjointlyusingGrayLevelCo-occurrenceProbability(GLCP)andBPneuralnetworktechniques.First,upto8GLCP-associatedtexturefeatureparametersaredefinedandcomputed,andtheseconsequentparametersnextserveastheinputsfeedingtotheBPneuralnetworktocalculatethesimilaritytoanyofgivenaggregatetexturetype.Afinitenumberofaggregateimagesof3kinds,witheachcontainingspecifictypeofmineralparticles,areputtotheidentificationtest,experimentallyprovingthefeasibilityandrobustnessoftheproposedmethod.
简介:Thisletterstudiesandanalyzestheworkingfeaturesofmaincircuitoftri-levelboostPowerFactorCorrect(PFC)converterandtheadvantagesoftri-levelswitchconverterinaspectsofbearinghigh-voltageofpowercomponents,overallsystemlossandmagneticcomponentselectionbaseduponthesingle-levelboostPFCswitchconverter.Besides,relyingontheapplicationofmi-croprocessorinpowerconvertertechnologyandDSP(DigitalSignalProcessing)chip'sstrongcal-culatingcapacity,theletterpresentstheadoptionofmodifiedschemeoftri-levelboostPFCconverterunderthecontrolofpredictivecontrolalgorithm.Moreover,theoperatingprincipleandcontrolmethodarespecified,theresultsofcircuittestandanalysisareprovidedandtheadvantagesofpre-dictivecontroltechnology-basedmulti-levelboostPFCconverterisverified.