英语电子信息工程9.ppt

《英语电子信息工程9.ppt》由会员上传分享，免费在线阅读，更多相关内容在应用文档-天天文库。

1、Unit9DigitalSignalProcessingPassageABasicConceptsofDSPPassageBDigitalSignalProcessorsPassageCComparisonofDSPandASPPassageABasicConceptsofDSPWedon’tspeakinadigitalsignal.Adigitalsignalisalanguageof1sand0sthatcanbeprocessedbymathematics.Wespeakinreal-world,analogsigna

2、ls.Analogsignalsarerealworldsignalsthatweexperienceeveryday-sound,light,temperature,andpressure.Adigitalsignalisanumericalrepresentationoftheanalogsignal.Itmaybeeasierandmorecosteffectivetoprocessthesesignalsinthedigitalworld.Intherealworld,wecanconvertthesesignalsin

3、todigitalsignalsthroughtheanalog-to-digitalconverter,processthesignals,andifneeded,bringthesignalsbackouttotheanalogworldthroughthedigital-to-analogconverter.1.Theessentialsofanalog-to-digitalanddigital-to-analogconversionThefirstessentialstepinanalog-to-digital(A

4、/D)conversion(asshowninFigure9.1)istosampleananalogsignal.Thisstepisperformedbyasampleandholdcircuit,whichsamplesatregularintervalscalledsamplingintervals.Thelengthofthesamplingintervalisthesameasthesamplingperiod,andthereciprocalofthesamplingperiodisthesamplingfrequ

5、encyfs.AccordingtotheNyquisttheorem,asignalwithamaximumfrequencyofWHz(calledaband-limitedsignal)mustbesampledatleast2Wsamplespersecondtoensureaccuraterecording.Whenthisminimumisnotrespected,distortioncalledaliasingoccurs.Aliasingcauseshighfrequencysignalstoappearaslo

6、werfrequencysignals.Tobesurealiasingwillnotoccur,samplingisalwaysprecededbylowpassfiltering.Thelowpassfilter,calledtheanti-aliasingfilter,removesallfrequenciesabovehalftheselectedsamplingrate.Figure9.1Analog-to-DigitalConversionsAfterabriefacquisitiontime,duringwhic

7、hasampleisacquired,thesampleandholdcircuitholdsthesamplesteadyfortheremainderofthesamplinginterval.ThisholdtimeisneededtoallowtimeforanA/Dconvertertogenerateadigitalcodethatbestcorrespondstotheanalogsample.TheA/Dconverterchoosesaquantizationlevelforeachanalogsample.

8、AnN-bitconverterchoosesamong2Npossiblequantizationlevels.Thelargerthenumberoflevels,thesmallerthequantizationerrors,calculatedasth