3-1©2005 Raj JainCSE473sWashington University in St. LouisData Data TransmissionTransmissionRaj JainProfessor of CSE Washington University in Saint LouisSaint Louis, MO [email protected] slides are available on-line at:http://www.cse.wustl.edu/~jain/cse473-05/3-2©2005 Raj JainCSE473sWashington University in St. LouisOverviewOverview Time Domain and Frequency Domain Decibels  Data vs Signal Attenuation, Delay Distortion, Noise, Capacity3-3©2005 Raj JainCSE473sWashington University in St. LouisTransmission TerminologyTransmission Terminology Guided Media: Wire, Fiber , coaxUnguided Media: Air, Vacuum, sea water Direct Link Point to Point vs Point to MultipointT RT RTR1 R2 R33-4©2005 Raj JainCSE473sWashington University in St. LouisLine DuplexityLine Duplexity Simplex: Transmit or receive, e.g., TelevisionHalf-Duplex: Transmit and receive alternately, e.g., Police Radio Full Duplex: Transmit and receive simultaneously,e.g., TelephoneT RT/R T/R3-5©2005 Raj JainCSE473sWashington University in St. LouisAnalog vs Digital SignalsAnalog vs Digital SignalsAnalog: Digital:3-6©2005 Raj JainCSE473sWashington University in St. LouisPeriodic vs AperiodicPeriodic vs Aperiodic Periodic: Signal pattern repeats over time Aperiodic: Not periodicTS(t+T) = s(t) for all tPeriod = T3-7©2005 Raj JainCSE473sWashington University in St. LouisExamples of Periodic SignalsExamples of Periodic Signals Sine Wave: Square Wave:3-8©2005 Raj JainCSE473sWashington University in St. LouisFrequency, Period, and PhaseFrequency, Period, and Phase A Sin(2πft + θ), Period T=1/f, Frequency in Hertz3-9©2005 Raj JainCSE473sWashington University in St. LouisSine WaveSine Wave Peak Amplitude (A): Maximum strength of signal in volts Frequency (f): Hertz (Hz) or cycles per second Period = time for one repetition (T)T = 1/f Phase (φ): Relative position in time3-10©2005 Raj JainCSE473sWashington University in St. LouisWavelengthWavelengthDistance occupied by one cycleDistance between two points of corresponding phase in two consecutive cyclesWavelength = λAssuming signal velocity vλ= vTλf = vc = 3*108 m/s (speed of light in free space) = 300 m/µsDistanceAmplitude3-11©2005 Raj JainCSE473sWashington University in St. LouisTime and Frequency DomainsTime and Frequency DomainsFrequencyAmplitudeFrequencyAmplitudeFrequencyAmplitudef3fAAf 3fA/3A/33-12©2005 Raj JainCSE473sWashington University in St. LouisFrequency Domain ConceptsFrequency Domain Concepts Fundamental Frequency: All other frequency components are multiple of fundamental frequency f Period = 1/f Spectrum: Range of frequencies Absolute Bandwidth: Width of the spectrumAbsolute Bandwidth = 3f-f = 2f Effective Bandwidth: Narrow band of frequencies containing most of the energy DC Component: Constant or zero frequencyA+B sin (2πft+θ)3-13©2005 Raj JainCSE473sWashington University in St. LouisSignal with DC ComponentSignal with DC Component3-14©2005 Raj JainCSE473sWashington University in St. LouisFrequency Components of Square WaveFrequency Components of Square WaveFreq.AmplitudeAmplitudeFrequencyfA3f 5fA/3A/5Freq.f 3fA/35fA/57fA/7Σκ=1,3,5,A/k sin (2πkft)3-15©2005 Raj JainCSE473sWashington University in St. LouisData vs SignalData vs SignalTelephoneModemCODECDigital TransceiverAnalogDigitalDataSignalAnalogDigitalAnalogAnalogDigitalDigitalData SignalMediumData3-16©2005 Raj JainCSE473sWashington University in St. LouisAnalog Data Example: Speech and MusicAnalog Data Example: Speech and Music3-17©2005 Raj JainCSE473sWashington University in St. LouisAnalog Data Example 2: TelevisionAnalog Data Example 2: Television30 Screens/secInterlacing: Odd lines every 1/60 sand even lines every 1/60 s483 lines/screen3-18©2005 Raj JainCSE473sWashington University in St. LouisVideo SignalVideo SignalUSA - 483 lines scanned per frame at 30 frames per second525 lines but 42 lost during vertical retraceSo 525 lines x 30 scans = 15750 lines per second63.5µs per line11µs for retrace, so 52.5 µs per video lineMax frequency if line alternates black and whiteHorizontal resolution is about 450 lines giving 225 cycles of wave in 52.5 µsMax frequency of 4.2MHz3-19©2005 Raj JainCSE473sWashington University in St. LouisAttenuation Attenuation and Dispersion (Delay Distortion)and Dispersion (Delay Distortion)Distance3-20©2005 Raj JainCSE473sWashington University in St. LouisDigital TransmissionDigital Transmission Repeaters are used to regenerate digital signal Signal attenuation is overcome Noise is not amplified Low cost LSI/VLSI technology Longer distances over lower quality lines Capacity utilizationHigh bandwidth links economicalHigh degree of multiplexing easier with digital techniques Security & Privacy: Encryption3-21©2005 Raj JainCSE473sWashington University in St. LouisDecibelsDecibels Attenuation = Log10 PinPoutExample 1: Pin = 10 mW, Pout=5 mWAttenuation = 10 log 10(10/5) = 10 log 102 = 3 dBExample 2: Pin = 100mW, Pout=1 mWAttenuation = 10 log 10(100/1) = 10 log 10100 = 20 dBBelPinPoutdecibelAttenuation = 10 Log10 VinVoutdecibelAttenuation = 20 Log103-22©2005 Raj JainCSE473sWashington University in St. LouisNoise Noise  Additional signals inserted between transmitter and receiver Thermal Noise:Due to thermal agitation of electronsUniformly distributedWhite noise Intermodulation Noise:Signals that are the sum and difference of original frequencies sharing a medium3-23©2005 Raj JainCSE473sWashington University in St. LouisNoise (Cont)Noise (Cont) Crosstalk Noise: A signal from one line is picked up by another Impulse Noise: Irregular pulses or spikese.g., External electromagnetic interferenceShort durationHigh amplitude3-24©2005 Raj JainCSE473sWashington University in St. LouisChannel CapacityChannel Capacity Capacity = Maximum data rate for a channel Nyquist Theorem: Bandwidth = BData rate <2 B Bi-level Encoding: Data rate = 2 × Bandwidth05V Multilevel: Data rate = 2 × Bandwidth × log 2MExample: M=4, Capacity = 4 × Bandwidth3-25©2005 Raj JainCSE473sWashington University in St. LouisShannon's TheoremShannon's Theorem Bandwidth = B HzSignal-to-noise ratio = S/N  Maximum number of bits/sec = B log2(1+S/N) Example: Phone wire bandwidth = 3100 HzS/N = 30 dB 10 Log 10S/N = 30Log 10S/N = 3S/N = 103= 1000Capacity = 3100