Analog/Digital CodingOverviewAnalog SignalsSlide 4ProblemsDigital SignalsSlide 7Analog to Digital CodingSlide 9Slide 10Slide 11Slide 12CD A/D ConversionDigital to Analog CodingQuestions?Analog/DigitalAnalog/DigitalCodingCodingBobby GeevarugheseBobby GeevarugheseECE-E443ECE-E4434/12/054/12/05OverviewOverview Analog SignalsAnalog Signals Digital SignalsDigital Signals A/D CodingA/D Coding D/A CodingD/A CodingAnalog SignalsAnalog SignalsSound carried as pattern of pressure Sound carried as pattern of pressure waveswavesChange in pressure is measured in each Change in pressure is measured in each instant and voltage will vary in same wayinstant and voltage will vary in same wayAnalog SignalsAnalog SignalsProblemsProblemsDegradationDegradationSize limitationsSize limitationsNoiseNoiseVarious amounts will alter reconstruction of Various amounts will alter reconstruction of original signaloriginal signalDigital SignalsDigital SignalsFinite valuesFinite valuesConsists of patterns of bits of informationConsists of patterns of bits of informationAdvantage over analog signals in that Advantage over analog signals in that precise signal level is not necessaryprecise signal level is not necessaryDigital SignalsDigital SignalsAnalog to Digital CodingAnalog to Digital CodingSampling TheoremSampling TheoremMust sample at 2x the BW to perfectly reconstruct the Must sample at 2x the BW to perfectly reconstruct the original signal and avoid aliasingoriginal signal and avoid aliasingProblem arises because values are still in continuous Problem arises because values are still in continuous range and can take on any valuerange and can take on any valueQuantizationQuantizationSamples are “rounded off” to the nearest Samples are “rounded off” to the nearest amplitude valueamplitude valueAnalog to Digital CodingAnalog to Digital CodingAnalog to Digital CodingAnalog to Digital CodingCompact DiscCompact DiscFalling and rising edges of a pit correspond to Falling and rising edges of a pit correspond to the falling and rising edges of the pulse trainthe falling and rising edges of the pulse trainLight sensor recreates sequence of 1’s and 0’s Light sensor recreates sequence of 1’s and 0’s to convert back to original signalto convert back to original signalCD A/D ConversionCD A/D ConversionSampling rate = 44,100 samples/secondSampling rate = 44,100 samples/second16 bits/sample (word length) needed to 16 bits/sample (word length) needed to represent different quantization levelsrepresent different quantization levelsQuantization levels = 65,536Quantization levels = 65,536Sampling at this rate with this much bit Sampling at this rate with this much bit resolution enables perfect reconstruction of resolution enables perfect reconstruction of original signaloriginal signalDigital to Analog CodingDigital to Analog Coding Complete reversal of ADCComplete reversal of