National Kaohsiung First University of Science and Technology
FirstTech Institutional RepositoryNot a member yet
18584 research outputs found
Sort by
Design and Application of CMOS Microwave Frequency Synthesizer
No full text[[abstract]]由於近??積體電?技術大幅提昇，可攜式個人通訊器材普及，鎖相迴?獲得很大的應用。而鎖相迴?是一個能夠跟蹤輸入訊號相位與頻?的閉迴?自動控制系統，因此在單晶片系統電?上，鎖相迴?可?是很熱門的電?，所以本論文使用了TSMC0.18umCMOS製程來設計出一頻率合成器的晶片。本論文之整數式頻率合成器之架構整體架構採用相位頻率檢測器(PhaseFrequencyDetector，PFD)、電流泵浦(ChargePump，CP)、迴路濾波器(LoopFilter)、電壓控制振盪器(VoltageControlledOscillator，VCO)以及除頻器(Divider)，電路實踐使用台積電CMOST18製成來實現整數式頻率合成器，晶片大小為1070.400*1073.014μ㎡。本論文之整數式頻率合成器，其頻率範圍為4.4~4.7GHz，中心頻率為4.55GHz其可調範圍為300MHz，消耗功率為193.24mW，輸出功率為-0.14dBm，而在距離載波1MHz頻率偏移時為-114.745dBc/Hz。[[abstract]]Asaresultofrecentyears,asubstantialincreaseinintegratedcircuittechnology,Portablepersonalcommunicationequipmentuniversal,Phase-lockedloopforagreatdealofapplication.Thephase-lockedloopisaclosedloopautomaticcontrolsystemcapableoftrackingthephaseandfrequencyoftheinputsignal,Therefore,inthesingle-chipsystemcircuit,Phase-lockedloopcanbesaidtobeverypopularcircuit,SothispaperusestheTSMC0.18umCMOSprocesstodesignafrequencysynthesizerchip.Inthispaper,theoverallstructureoftheintegerfrequencysynthesizeradoptsphasefrequencydetector(PFD),chargepump(CP),loopfilter,voltagecontrolledoscillator(VCO)anddivider,CircuitpracticeusingTSMCCMOST18madetoachieveintegerfrequencysynthesizer,Thewafersizeis1070.400*1073.014μ㎡.Inthispaper,theintegerfrequencysynthesizer,Itsfrequencyrangeof4.4~4.7GHz,Thecenterfrequencyof4.55GHzadjustablerangeof300MHz,Thepowerconsumptionis193.24mW,Theoutputpoweris-0.14dBm,Andat-114.745dBc/Hzatafrequencyoffsetof1MHzfromthecarrier
The Study of Security Protocols in LTE Networks
No full text[[abstract]]本論文針對第四代行動通信系統之長程演進（LongTermEvolution，簡稱LTE）之區域更新（TrackingAreaUpdate，簡稱TAU）、服務請求（ServiceRequest，簡稱SR）以及認證與密碼交換（AuthenticationandKeyAgreement，簡稱AKA）流程研究，並利用偽eNB的架設，誘使受害者依附，待受害者依附後進行TAU或SR流程測試，發現除了原本文獻提到的TAU流程可以對受害者達成阻斷服務攻擊（DenialofService，簡稱DoS）與降等攻擊外，在SR流程也能促發DoS與降等攻擊。此外可利用SR或TAU流程達成主動式國際移動用戶識別碼（InternationalMobileSubscriberIdentity，簡稱IMSI）捕捉，對於IMSI的洩漏程度較大，而透過捕捉到的IMSI可進而研究AKA流程之缺陷，透過受害者的IMSI碼發起的AttachRequest可以獲得用以認證受害者的認證訊息，攻擊者藉由回覆合法的認證訊息讓受者通過偽認證。[[abstract]]Inthispaper,wefocusonthetrackingareaupdate(TAU),servicerequest(SR)andauthenticationandkeyagreement(AKA)oftheLongTermEvolution(LTE)system,andweusethefakeeNBtoinducevictimstodoTAUorSRprocedure.Inthepreviouspaper,theweaknessoftheTAUprocedureisthatthevictimsmightbeforcetotheDoSorthedowngradeattack.Inourtest,wenoticedthattheSRprocedurehasthesameweaknessastheTAUprocedure.Andthisweaknesscangetinternationalmobilesubscriberidentity(IMSI),inthispaper,weproposedanactivecapturemethodwhichtheIMSIwillcertainlybecaptured.WhenthismethodgettheIMSIfromvictims,wecanusethattomakeafakeattachrequest,andtheEPCwillresponsetheauthenticationparametertous.Thisparameterwilldeceivethevictimtodoauthentication
Thermal Pedestrian Benchmark Construction and Feature Analysis
No full text[[abstract]]行人資料庫(PedestrianDataset)於行人偵測演算法開發過程中，扮演著重要與關鍵之角色，其能有效驗證當前行人偵測演算法之有效性(Effectiveness)與侷限(Limitation)，提供研究學者演算法改善之依據，目前文獻中存在諸多完整之可見光行人資料庫，然而可見光感測器主要侷限於環境光源不足時，例如:夜間，霧天等，其無法有效成像，使得其並不適用於全天候行人偵測應用中。基於上述，近幾年來有諸多基於熱影像之行人偵測演算法被提出，但就目前所蒐集之文獻，並無一套完善之熱影像行人資料庫可供驗證既有之熱像行人演算法，因此本研究主要目的在建構一套具有熱像儀特性與行人多樣性之完善熱影像資料庫，所建構之熱像行人資料庫，包含各種背景場景、人形姿態、不同距離與拍攝角度、同時包含靜態背景與動態影像序列，以模擬監控與智慧型車輛應用，可供驗證各種應用之行人偵測演算法使用。此外，基於二值化滑動視窗(BinarySlidingWindow)搭配支援向量機(SupportVectorMachine:SVM)之行人偵測之架構，本論文同時提出一個系統化分析現有文獻中常用熱像特徵之效能，主要分析因素包含:訓練樣本比之例、不同遠近、不同遮蔽情況下之效能，並FPPI曲線形式呈現實驗分析數據。[[abstract]]Pedestriandatasetplaysacrucialpartindevelopmentofpedestriandetectivealgorithm,itcanvalidateeffectivenessandlimitationofcurrentpedestriandetectivealgorithm,providingfoundationofimprovementforresearchers.Therearemanyexistingvisiblepedestriandatasetsinpresentreferences.However,thevisiblesensorscan’tuseinpoorlightconditions,forexample:innight,inmistetc.Insuchenvironment,itcan’tformtheimageproperlyandthatwillmakeitunsuitableforapplicationofall-daypedestriandetection.Accordingly,abundantthermalpedestriandetectionalgorithmshavebeenproposedincurrentyears.Butthereisn’taperfectthermalpedestriandatasetintheliterature.Therefore,themainpurposeofourstudyistoconstructasetofthermalimagingdatasetwithspecificsofthermalimagerandpedestrianmultiplicity.Theconstructedthermalpedestriandatasetincludesvariousbackgrounds,scenes,postures,differentdistance,shotanglesandstaticbackgroundwithdynamicimagingsequences.Thatmakessimulatedsurveillanceandapplicationofintelligentvehiclepossible,providingtovalidatepedestriandetectionalgorithmsinvariousapplications.Furthermore,duetoBinarySlidingWindowframeworkwhichispopularoneforpedestriandetection,ourresearchproposesansystematicwaytoanalyzethefactorsthatinfluencestheperformanceofalgorithms.Thefactorsgenerallyincludesthepositive/negativesampleproportion,differentdistancesandocclusionratio.AllresultsareexhibitedinFPPI-formcurve
Design of Partial Discharge Real-Time Capture System
No full text[[abstract]]局部放電通常發生於絕緣體的內部或表面電場集中區域，會產生化學反應侵蝕絕緣材料使其劣化，並在電場集中區域造成放電效果，最後使絕緣材料崩潰進而導致儀器損毀和危及使用者人身安全。本論文提出一局部放電即時在線監測系統，其系統架構包含FPGA組件、人機介面和類比放大電路等，配合本論文開發的即時在線監測軟體，達到局部放電即時在線監測、記錄與分析的功能。首先將所抓到的局部放電訊號利用FPGA進行資料分析，其分析結果以RS-232傳輸至人機介面及顯示資料。分析的方式包含有FastFourierTransform(FFT)分析、TimeFrequencyMap(T-FMap)分析、PhaseResolvedPartialDischarge(PRPD)分析等，用以分別說明局部放電訊號所對應的特性，藉數據分析來提高系統的穩定性，降低誤判發生機率，並依據實測結果來分析局部放電可能發生的原因，進一步避免儀器設備的損壞與使用者的危險。本系統使用Altera公司的BeMicroCV和BeScope做為硬體基礎，設計類比放大電路倍率和頻寬分別達到10倍和100MHz，並使用高壓局部放電模擬平台來驗證系統正確性。關鍵字:局部放電、在線監測系統、FFT、T-FMap、PRPD[[abstract]]Partialdischargeusuallyoccursintheinsulatororsurfaceelectricfieldconcentrationarea.Itwillproduceachemicalreactiontoerodetheinsulatingmaterialdamagedandcausethedischargeeffectintheelectricfieldconcentratedarea.Theinsulationmaterialdeteriorationresultinstrumentdamageandendangerthesafetyofusersfinally.Thisthesispresentsanon-linepartialdischargereal-timeCaptureSystem.ThesystemarchitectureincludesFPGAcomponents,humanmachineinterface(HMI)andanalogamplifycircuits.Withthereal-timeonlinemonitoringsoftwaredevelopedinthispaper,theon-linemonitoring,recordingandanalysisofpartialdischargecanbeachieved.Inthisthesis,thecapturedsignalofpartialdischargeisfirstlyanalyzedbyFPGA,andtheanalysisresultisthentransmittedbyRS-232toHMIfordisplaying.TheanalysismethodincludesFastFourierTransform(FFT)analysis,TimeFrequencyMap(T-FMap)analysis,andPhaseResolvedPartialDischarge(PRPD)analysis.Theyareusedtodescribethecharacteristicsofpartialdischargesignals,respectively.Bydataanalysistoimprovethestabilityofthesystemandreducetheincidenceofmiscarriageofjustice,basedonthemeasuredresultstoanalyzethepossiblecausesofpartialdischargeandfurthertoavoiddamagingtoequipmentanduserrisk.Thehardwareofthepartialdischargereal-timemonitoringsystemwasdesignedbasedonBeMicroCVandBeScope,andthemagnificationrateandbandwidthofanalogamplifycircuitdesignedtoachieve10and100MHzrespectively.Finally,thehigh-voltagepartialdischargesimulationplatformwasusedtoverifythefeasibilityandpracticabilityofsystem.Keywords:PartialDischarge,On-LineMonitoringSystem,FFT,T-FMap,PRP
Design and Implementation of Novel Interleaved High Step-up Converter for Fuel Cell Applications
No full text[[abstract]]本論文提出使用變壓器之倍壓箝位交錯順向式高昇壓轉換電路，變壓器為順向模式，且輸出電容為串聯形式，可實現整體電路之高昇壓轉換比。利用箝位電路之倍壓技術與無損耗被動箝位功能。如此不但提升電壓增益轉換，且變壓器之漏感能量可回收至輸出端，同時可抑制開關突波電壓，使轉換電路效率提高。且功率開關為低電壓應力，元件可選用低導通內阻(Rds(ON))之MOSFET，減少導通損耗，架構之效率藉此提高。所提出新型交錯式高昇壓轉換電路做實際測試，以驗證高效率之可行性與電路特性，其規格為輸入電壓24V、輸出電壓380V與額定功率1kW。最高效率為300W之96.24%，滿載1kW效率為92.57%。[[abstract]]Thispaperpresentnovelinterleavedhighstep-upconverterwithtransformerandvoltagedoublerofclampcircuit.Transformerisaforwardrectifiedoutput,andtheoutputcapacitorisinseries,whichachievethehighboostconversionratiooftheoverallcircuit.Theconverterutilizevoltagemultipiliertechniquesofclampcircuitandthepassivelosslessclampedfunction.Notonlytoenhancethevoltagegainconversion,butalsotheleakageenergyofidealtransformertoberecycledtooutputterminal,whichalleviatesthevoltagespikeacrossswitch,Theconvertercanlifttheconversionefficiency.Thepowerswitchislowvoltagestressandcomponentsareavailablelowconductionresistance(Rds(ON))ofMOSFET.Toreducetheconductionloss,andtheefficiencyofthestructuretoimprove.Thespecificationoftheproposedconvertersforperformanceverificationandcircuitcharacteristicsareshownasbelow:Theproposedconverteristestedwithinputvoltage24V,outputvoltage380Vandmaximumpower1kW.Theproposedconverterhighestefficiencyis96.24%inthe300W,1kWfullloadefficiencyis92.57%
The Relationship between Event Marketing of Cycling around The Island and Corporate Image –A Case Study of E.SUN BANK
No full text[[abstract]]為探究單車環島活動事件行銷與企業形象關係之議題，本研究以玉山銀行所共舉辦的「玉山單車環島．千里傳愛」活動為例，透過相關文獻整理，並以玉山銀行之一般客戶為研究主體，採用實驗法之程序，將活動舉辦模式分為：「自主環島」、「受贊助者自行環島」及「受贊助者招募環島」三類進行實驗，再輔以敘述性統計、ANOVA變異數分析方法等，探討「事件行銷型態」、「單車環島態度」及「企業形象」之間的關係。研究結果發現，玉山銀行所舉辦的單車環島事件行銷，在企業形象上是有正向影響。而在不同的活動模式中，「自主環島」的成效優於「贊助環島」，即表示此運動事件行銷，是企業行銷時值得參考的行銷方式；另「贊助環島」模式中，「受贊助者招募環島」比起「受贊助者自行環島」在企業形象上，雖沒有顯著上的差異，但影響略高，故研究者建議，若企業未來欲舉辦單車環島活動時，可考慮以自主再加上廣邀客戶或民眾共同參與之方式進行，或許可達到擴大規模與影響力之效果，進而塑造更佳的企業形象。在單車環島態度方面，研究結果顯示：單車環島對於玉山銀行一般客戶而言，是個正向有意義的體育性活動，且廣被民眾接受與認同。但從數據中也發現，雖然「事件行銷型態」及「單車環島態度」分別對企業形象都有顯著關係，但兩者在企業形象上並沒有顯著的交互作用。[[abstract]]ToexploretherelationshipbetweenthemarketingeventofcyclingaroundTaiwanandcorporateimage,theresearchtakestheeventnamed“CyclingaroundTaiwan,ConveyingLoveinthousandmiles”heldbyE.SunBankasanexample.Throughcommunicationsaddressedinrelatedliterature,takinggeneralcustomersasthesubject,adoptingproceduresofexperimentalmethod,theeventisdividedinto3types.Theyare“CyclingaroundTaiwanforspontanuouscustomers”,“CyclingaroundTaiwanforthosewhorecievedsponsorship”and“CyclingaroundTaiwanforthosewhorecuitedbysponsoredones.”Analyzetheexperimentin3typessupplementedwithnarrativestatistics,ANOVAvarianceanalysismethods,etc.,todiscusstherelationshipamong“thetypesofmarketingevent”,“attitudesforcyclersofthisevent”and“corporateimage.”TheresultshowstheeventheldbyE.Sunbankdidpositiveeffectonitscorporateimage.However,differenttypesledtodifferentoutcomes.Thebeneficialresultfrom“CyclingaroundTaiwanforspontanuouscustomers”isbetterthanthatfrom“CyclingaroundTaiwanforthosewhorecivedsponsorship.”Thatmeansthismarketingeventaboutsportsisworthtakingasreferenceforcorporatemarketing.Abouttypesofsponsoredones,compared“CyclingaroundTaiwanforthosewhorecivedsponsorship”and“CyclingaroundTaiwanforthosewhorecuitedbysponsoredones,”thereisnoobviousdifferenceoncorporateimage.However,thelattertypehasbitbetterbeneficialeffect.Thustheresearcheradvisesthecorporatecanconsiderthetypeof“CyclingaroundTaiwanforspontanuouscustomers”andwidelyinvitecustomersorthepublictoparticipate.Thismayexpandtheeventscaleandinfluenceandthenformabettercorporateimage.Abouttheattitudeforcyclersofthisevent,theresultshowscyclingaroundTaiwanforgeneralcustomersofE.SunBankisapositiveandmeaningfulsportactivity,andwidelyacceptedandrecoginized.ThoughthedatapresentsthereisaremarkablerelationshipwithcorporateimagerespectivelyonthetypeofcyclingaroundTaiwanandtheattitudeforcyclersofthisactivity,thereisnosignificantinteractioneffectwithcorporateimageforbothofthem
Information Quantity Satisfaction of Knowledge Community
No full text[[abstract]]網?的快速發展，也造成虛擬社群的興起，虛擬社群是藉著網路進行溝通與知識交流，也往往不會面對面溝通，同時成員間彼此並不熟識。由一群有共話題、想法、嗜好的人而群聚，跨越時間和空間的限制，透過網路方式，與成員進行互動和交流，發展過程中虛擬的社群也漸漸導向專業社群的型態，成員間彼此分享資訊與互相學習彼此的專業知?。進而讓專業社群中擁有足夠的資訊產出，然而是?麼樣的因素會讓使用者對專業社群中的資訊產出感到滿意?本研究以自我規範學習與社會規範學習作為核心，探討社群使用者對專業社群的資訊產出滿意度。使用問卷調查法蒐集相關看法，透過社群媒體管道進行發放，一共回收有效問卷共266份，研究結果發現，（1）自我規範學習並非有正向影響使用者對資訊社群的資訊產出滿意度。（2）社會規範學習正向影響使用者對資訊社群的資訊產出滿意度。（3）自我能力評估和創新氣氛皆對自我規範學習與社會規範學習有正向影響。過去學術界中將學習策略套用在公司組織中，然而本研究為將學習策略套用在專業社群中，此研究的結果顯示出新的貢獻與驗證，學習策略會對社群的資訊產出滿意造成影響。[[abstract]]TherapiddevelopmentofInternet,alsoledtotheemergenceofvirtualcommunities,operatingthroughanetworkofvirtualcommunities,noentityisavirtualform.Sharedbyagroupoftopics,ideas,interests,andbunching,acrosstimeandspaceconstraints,throughthenetwork,interactandcommunicatewithmembers,virtualcommunitieswillbecomeorientedinthedevelopmentpatternsofprofessionalcommunity,membersshareinformationwitheachotherandlearneachother'sprofessionalknowledge.Whichfactorsmakesuserssatisfywithinformationoutput?Focusingonself-regulatedlearningandsocial-regulatedlearning,thisresearchdiscussesusers’satisfactionabouttheinformationfromspecializedsocialgroups.Usequestionnairestocollectrelevantviewscollectedthroughsocialmediatoatotalof266validquestionnaires，theresearchersfoundthat（1）self-regulatedlearningdoesn'thaveapositiveimpactontheuser'ssatisfactionaboutthecommunity'sinformationoutput;（2）social-regulatedlearningaffectspositivelytheuser'ssatisfactionabouttheinformationcommunity'sinformationoutput.（3）Benefitframesandclimateforinnovationhaveapositiveimpactonself-regulatedlearningandsocial-regulatedlearning.Inthepast,academicrelationswereappliedtocorporateorganizations.However,thisstudyappliedthelearningrelationshiptotheprofessionalcommunity.Theresultsofthisstudyshowednewcontributionsandvalidation,andthelearningrelationshipwillhaveanimpactonthesatisfactionofthecommunity'sinformationoutput
Does capitalization size affect the pricing of over-trading stocks？
No full text[[abstract]]本研究是以探討投資人因過度自信行為造成股價過度交易現象，股票價格被推升上來，在出現主升段訊號就先買進，然後股價會因市場量比價先行，漲波段後會盤整拉回，觀察確定後再買進，股票會呈現持續走揚趨勢，本論文在研究這段期間是否股本越小者，其股票投資組合報酬率會越大，股本越大者，其股票投資組合報酬率相對較小，本文以臺灣股票上市、上櫃公司為研究樣本，研究期間為2015年4月01日至2017年1月31日，研究結果顯示，平均累積異常報酬率最高落在股本規模次高(30億-100億)的區間，平均累積異常報酬率第二高的才是落在小股本(5億-10億)區間，但並非股本規模最小的(5億以下)報酬率是最高的，其中原因，股本規模最小之個股，容易被操作，演變正常投資行為產生不正常操作結果。[[abstract]]Thispaperstudieshowinvestors’overconfidenceaffectstheirover-tradingbehaviorandthusstockreturnsanomaly.Themainpurposeofthispaperistotestwhethersmall(large)capstocksexperiencehigher(lower)stockreturnsinthefuture.UsingasampleofTWSE-andOTC-listedstocksinTaiwanfromApril2015toJanuary2017,wefindnoevidencethatthesmallest-capstocks(cap<0.5NTbillions<cap<10NT$billions)experiencethehigheststockreturnsinthefuture.Thesefindingssupportthehypothesisthatthesmallest-capstocksaremorelikelytobemanipulated,whichinturnleadstoareturnsanomaly
The Effects of Reducing Property Speculation on Housing Prices
No full text[[abstract]]過去文獻在討論房價指數時多半使用經濟變數來當做變數，然而當經濟景氣不好時，政府出手救經濟卻導致近幾年房價大漲進而迫使政府打房，但打房政策在相關文獻大多是討論財政部的奢侈稅及實價登錄居多，但結果都顯示為不顯著。後來財政部與中央銀行陸續推出了一連串打房政策，其中效果是否加成或是減緩將是本文研究目的。本研究利用迴歸分析來針對國泰房地產指數之全國與六大都會區進行實證分析，以2009年第一季至2016年第四季為研究區間，實證結果顯示財政部的房地合一稅是有效抑制房價上漲，但實價登錄卻反而使房價上漲，此外央行打房政策利率調整與貸款成數限制並沒有顯著的效果。[[abstract]]Theaimofthisstudyistoprovideacomprehensivedescriptionofthedependencepatternofhousepriceindex(Cathay)andoverallwithaspecialmunicipalitybystudyingarangeoftheconditionalpricedistributionbyusingregressionanalysis.OurdatacoverCathayIndex.Thesampleperiodspansroughly7yearsfromfirstquarterof2009tothefourthquarterof2016.Ourempiricalresultshowthatthecombinedhousingandreal-estatetaxweresignificantlyinfluentthereducingpropertyspeculationonhousingprices.However,theactualpriceregistrationkeepsthehousingpricesrising.Inaddition,thehousingpolicy,adjustingtherateandmortgagerate,arenotaffectingevidently
The Relationship between Investor 's Sentiment ,Investment Decision and Stock Return
No full text[[abstract]]本研究主要探討投資人情緒、投資決策和市場報酬率三者間的關係，投資人在面對不同投資人情緒時是否會影響其投資決策、投資人情緒是否會影響市場報酬率或是投資決策是否會影響市場報酬率。本研究還探討投資人情緒是否存在不對稱效果，同時還研究不同市場狀況的條件下，三者間關係的變化。實證結果發現，在不加入市場狀況的條件下，投資人情緒對於投資人投資決策無顯著的影響，代表其中不存在著不對稱效果，但是市場報酬率在投資人情緒較悲觀時有顯著的負影響。接著加入不同市場狀況的條件後，投資人情緒對於投資人投資決策有不同的顯著影響，存在不對稱效果。[[abstract]]Thisstudydiscussestherelationshipbetweeninvestor’ssentiment,transactionbehaviorandmarketreturn.Whetherinvestorswillaffecttheirtradingbehaviorinthefaceofdifferentinvestorsentiment,whethertheinvestor'smoodwillaffectthemarketreturnrateorthetransactionbehavior.Whetheritwillaffectthemarketrateofreturn.Thisstudyalsoexploreswhetherthereisanasymmetriceffectininvestorsentiment.Theempiricalresultsshowthatinvestors'mooddoesnothavenosignificanteffectonthebehaviorofinvestorswithoutaddingtothemarketconditions.Thereisnoasymmetriceffectontheinvestor'sbehavior,butthemarketreturnsaresignificantwhentheinvestor'smoodismorepessimisticNegativeimpact.Andaddtheconditionsofdifferentmarketconditions,theinvestorsentimentforthebehaviorofinvestorshavedifferentsignificantimpact,thereisasymmetryeffect