Waveform coding
- 1. TERM PAPER ON WAVEFORM CODING Department of Electronics and Communication Engineering Assessable Activity Based Learning (AAL) By: ALAPAN RANJAN BANERJEE ROLL NO :- 03 YEAR :- 3RD YEAR REGISTRATION NO. :- 161220110132 OF 2016-2017 ST. THOMAS’ COLLEGE OF ENGINEERING AND TECHNOLOGY Affiliated to MAULANA ABUL KALAM AZAD UNIVERSITY OF TECHNOLOGY ,WEST BENGAL
- 2. INTRODUCTION Althoughhumansare well equippedforanalogcommunications,analogtransmissionisnotparticularly efficient.Whenanalogsignalsbecome weakbecause of transmissionloss,itishardto separate the complex analogstructure fromthe structure of randomtransmissionnoise.If youamplifyanalogsignals, it alsoamplifiesnoise,andeventuallyanalogconnectionsbecometoonoisytouse.Digital signals, havingonly"one−bit"and"zero−bit"states,are more easilyseparatedfromnoise.Theycanbe amplifiedwithoutcorruption.Digital codingismore immune tonoise corruptiononlong−distance connections.Also,the world'scommunicationsystemshave convertedtoadigital transmissionformat calledpulse code modulation(PCM).PCMisa type of codingthat is called"waveform"codingbecause it createsa codedform of the original voice waveform.Thisdocumentdescribesata highlevel the conversionprocessof analogvoice signalstodigital signals. ANALOG TO DIGITAL CONVERSION Sampling:take samplesattime nT – T: samplingperiod; – fs= 1/T: samplingfrequency Quantization:mapamplitudevaluesintoasetof discrete valueskQ – Q: quantizationinterval orstep size BinaryEncoding – Converteachquantizedvalue intoabinarycodeword
- 3. QUANTIZATION A Q-levelquantizercomparesthe discrete-timeinputx(nT)withitsfixeddigital levels.Itassignsanyone of the digital level tox(nT) withitsfixeddigital levels.Itthenassignsanyone of the digital level tox(nT) whichresultsinminimumdistortionorerror.Thisiscalledquantizationerror.Thusthe otputof a quantizerisa digital level. Uniformquantization :- o the quantizationlevelsare uniformlyspaced o Applicable whenthe signal isinafinite range (fmin,fmax) o The entire datarange is dividedintoLequal intervalsof length Q(knownas quantizationinterval orquantizationstep-size) Q=(fmax-fmin)/L o Interval i is mappedtothe middle value of thisinterval o We store/sendonlythe index of quantizedvalue o There are twotypesof uniformquantization.Theyare Mid-Rise type andMid-Tread type. Non-uniformquantization:- If the quantizercharacteristicsisnonlinearandstepsize isnotconstantinsteadif it is variable,dependedonthe amplitudeof inputsignal thenthe quantizationisknownas nonuniformquantization. The non-uniformquantizationispracticallyachievedbythe processcalled COMPANDING. COMPANDING:- o The processof compressionof the signal atthe transmitterandexpansionatthe receiveris calledascompanding(compressionandexpansion).
- 4. o In companding,the higheramplitudeanalogsignalsare compressedprecedingtransmission and thenexpandedinthe receiver. o Compandingisperformedwiththe helpof three functional blocks,whichare Compressor, UniformquantizerandExpanderandthe diagramin shownbelow o The compressorwill compressthe dynamicrange of the signal sothat highdynamicrange signal can be passedthroughcomponentsof low dynamicrange capability,the uniform quantizerwill undergothe quantizationprocessof the compressedsignal andthe lastlythe expanderwillundergo expansionandinvertthe compressionfunctiontoreconstructthe original signal. o The expanderhascomplementarycharacteristicsasthatof compressorso that the compressorinputisequal toexpanderoutputinordertoreproduce the signal atthe receiver. o belowillustratesthe input-outputcharacteristicsandcurvesof the compandingprocess, and itcan be seenthatcompandinghaslinearcharacteristics. Ideally,we needalinearcompressioncharacteristicsforsmall amplitudesof the inputsignal anda logarithmiccharacteristicselsewhere.Inpractice,thisisdone byusingfollowingtwomethods: 1. A−law Companding 2. u−law Companding
- 5. A−lawCompander Limitingthe linearsamplevaluestotwelve magnitude bits,the A−law compressionis definedbythisequation,whereA isthe compressionparameter(A=87.7in Europe),and x isthe normalizedintegertobe compressed. u−lawCompander Limitingthe linearsamplevaluestothirteenmagnitude bits,the u−law (u−law andMu− laware usedinterchangeablyinthisdocument)compressionisdefinedbythisequation, where mis the compressionparameter(m=255 in the U.S. and Japan) andx isthe normalizedintegertobe compressed. A−lawstandardis primarilyusedbyEurope andthe restof the world.u−law isusedby NorthAmerica and Japan PULSE CODE MODULATION (PCM) In Pulse Code Modulation,the message signal isrepresentedbyasequence of codedpulses.This message signal isachievedbyrepresentingthe signal indiscrete form inbothtime andamplitude.
- 6. Basic Elementsof PCMcommunication :- o PCMtransmitter :- Low PassFilter Thisfiltereliminatesthe highfrequencycomponentspresentinthe inputanalog signal whichisgreaterthanthe highestfrequencyof the message signal,to avoidaliasingof the message signal. Sampler Thisis the technique whichhelpstocollectthe sample dataatinstantaneous valuesof message signal,soasto reconstructthe original signal.The sampling rate mustbe greaterthan twice the highestfrequencycomponent of the message signal,inaccordance withthe samplingtheorem. Quantizer Quantizingisaprocessof reducingthe excessive bitsandconfiningthe data. The sampledoutputwhengiventoQuantizer,reducesthe redundantbitsand compressesthe value. Encoder The digitization of analog signal is done by the encoder. It designates each quantized level by a binary code. The sampling done here is the sample-and-hold process. These three sections (LPF, Sampler, and Quantized) will act as an analog to digital converter.Encodingminimizesthe bandwidthused. o PCMtransmissionpath:- RegenerativeRepeater Thissectionincreasesthe signal strength.The outputof the channel alsohas one regenerativerepeatercircuit,tocompensate the signal lossandreconstruct
- 7. the signal,andalsoto increase itsstrength. Here isthe blockdiagramof regenerativefilter. o PCMreceiver:- Decoder The decodercircuitdecodesthe pulse codedwaveformtoreproduce the original signal.Thiscircuitactsas the demodulator. ReconstructionFilter Afterthe digital-to-analogconversionisdone bythe regenerative circuitandthe decoder,alow-passfilterisemployed,calledasthe reconstructionfiltertoget back the original signal. Application:- o The PCM isusedinthe satellitetransmissionsystem. o It isusedin space communication. o It isusedin telephony. o The compact disc (CD) isa recentapplicationof PCM. Advantagesof PCMcommunication:- o PCMprovideshighnoise immunity. o It ispossible toregenerate the codedsignalalongthe transmissionpath. o It ispossible tostore the signal. Disadvantagesof PCMcommunication:- o The encoding,decodingandquantizingcircuitryof PCMis complex. o PCMrequiresalarge bandwidthascomparedto othersystems.
- 8. DIFFERENTIAL PULSE CODE MODULATION (DPCM) For the samplesthatare highlycorrelated,whenencodedbyPCMtechnique,leave redundant informationbehind. PCMrequiresalarge bandwidthascomparedtoothersystems. Toprocessthis redundantinformationand tosave bandwidthDPCMis introduced. Basic Elementsof DPCMcommunication:- o DPCMtransmitter :- The predictorproducesthe assumedsamplesfromthe previousoutputsof the transmittercircuit.The inputtothispredictoristhe quantizedversionsof the input signal.Predictorinputisthe sumof quantizeroutputandpredictoroutput. o DPCM receiver:- The predictorassumesa value,basedonthe previousoutputs.The inputgiventothe decoderisprocessedandthatoutputis summedupwiththe outputof the predictor,to obtaina betteroutput.
- 9. Advantagesof DPCMcommunication:- o As the differencebetweenx(nT) andx*(nT) isbeingencodedandtransmittedbythe PCM technique asmall differance voltageistobe quantizedandencoded,thiswill needless numberof quantizationlevelsandhence lessnumberof bitstorepresentthem o Thus signalingrate andbandwidthof aDPCMwill be lessthanthat of PCM Disadvantagesof DPCMcommunication:- o Highbit rate o Practical usage is limited o Need the predicator circuitto be used which is very complex DELTA MODULATION The type of modulation,where the samplingrate ismuchhigherandin whichthe step size after quantizationisof a smallervalue Δ,sucha modulationistermedasdeltamodulation. If Mk > Mk-1 take 1 If Mk < Mk-1 take 0 Basic Elementsof deltamodulation :- o Transmitter :- The predictorcircuitin DPCMisreplacedbya simple delaycircuitinDM. From the above diagram, we have the notationsas− x(nTs)=oversampledinput ep(nTs)=summeroutputandquantizerinput eq(nTs) =quantizeroutput= v(nTs)v(nTs) xˆ(nTs) = outputof delaycircuit u(nTs) = inputof delaycircuit
- 10. o Receiver:- From the above diagram,we have the notationsas – vˆ(nTs) isthe inputsample uˆ(nTs) isthe summeroutput x¯(nTs) isthe delayedoutput A binary sequence will be givenasan inputtothe demodulator.The stair-case approximatedoutputisgiventothe LPF. Advantage :- o Since.Deltamodulationtransmit1bit forone sample,thereforethe signalingrate and transmissionbandwidthisquite small. o Very easydesignof the modulatorandthe demodulator Disadvantage :- o If signal changesveryfast“overloadslope noise”occurs. o If signal changesveryslow“granularnoise”occurs. ADDAPTIVE DELTA MODULATION In orderto overcome the quantizationerrors due toslope overloadandgranularnoise,the stepsize (Δ) ismade adaptive tovariationsinthe inputsignal x(t).Particularlyinthe steepsegmentof the signal x(t), the stepsize isincreased.Andthe stepisdecreasedwhenthe inputisvaryingslowly.Thismethodis knownas Adaptive DeltaModulation(ADM).The adaptive deltamodulatorscantake continuous changesinstepsize or discrete changesinstepsize.
- 11. Basic Elementsof deltamodulation:- o Transmitter :- The logicfor step size control isaddedinthe diagram.The stepsize increasesor decreasesaccordingtoa specifiedrule dependingonone bitquantizeroutput. o Receiver:- The receiverhastwoportions.The firstportionproducesthe stepsize fromeach incomingbit.Exactlythe same processisfollowedasthatintransmitter.The previous inputandpresentinputdecide the stepsize.Itisthenappliedtothe secondportioni.e., an accumulatorwhichbuildsupstaircase waveform.The low passfilterthensmoothens out the staircase waveformtoreconstructthe original signal. Advantages:- o The signal to noise ratioof ADMis betterthanthat of DM because of the reductionin slope overloaddistortionandidle noise. o Because of the variable stepsize ,the dynamicrange of ADM is widerthanDM. o Utilizationof bandwidthisbetterinADMthan DM.
- 12. Disadvantages :- o Implementationof stepsize control logiccircuitisnoteasy. COMPARISONBETWEEN PCM, DELTA MODULATION, ADAPTIVE DELTA MODULATION AND DEFFERENTIAL PULSE CODE MODULATION DIFFERENCES BETWEEN WAVEFORMCODING AND SOUCECODING :- The waveformcodersare generallydesignedtobe signal independent. The waveformcodersare differentfromthe sauce coders.The source codersdependon parameterizationof the analogsignal inaccordance withan appropriate model forgenerationof signal. REFERENCES :- https://electronicspost.com https://www.quora.com https://www.tutorialpoint.com https://www.wikipedia.com