Chapter 5Chapter 5Data EncodingData Encoding

ReviewReview

Information: Information: Numeric Data, Numeric Data, characters, voice, pictures, characters, voice, pictures, codes or any massage that can codes or any massage that can be read by and has meaning to be read by and has meaning to human and machine.human and machine.

ReviewReview

• For transmission:For transmission:– Information must be converted into binary Information must be converted into binary

first.first.• ASCII tableASCII table• UnicodeUnicode

– Information must be Information must be encoded into encoded into

electromagnetic signals.electromagnetic signals. (Analog or digital) (Analog or digital)

ReviewReview

• Digital Signal:Digital Signal:– A digital signal is a sequence of discrete A digital signal is a sequence of discrete

discontinuous voltage pulses.discontinuous voltage pulses.

• Each pulse is a signal elementEach pulse is a signal element

• In its simplest form each signal element In its simplest form each signal element represents a binary 0 or 1.represents a binary 0 or 1.

Data EncodingData Encoding

Both analog and digitalBoth analog and digital information can information can be be encoded as either analog or digitalencoded as either analog or digital. . (Function of media and communication )(Function of media and communication )

• Digital data, digital signalDigital data, digital signal

• Digital data, analog signalDigital data, analog signal

• Analog data, digital signalAnalog data, digital signal

• Analog data, analog signalAnalog data, analog signal

Terminology (digital signal)Terminology (digital signal)

• Unipolar encodingUnipolar encoding: If the signal : If the signal elements all have the same algebraic elements all have the same algebraic signs, all positive or all negative, the signs, all positive or all negative, the signal is called unipolar.signal is called unipolar.

• Polar encodingPolar encoding: One logical state is : One logical state is represented by positive voltage and the represented by positive voltage and the other by the negative voltage level.other by the negative voltage level.

Terminology (digital signal)Terminology (digital signal)

• Data rate:Data rate: The rate in bits per second that the data The rate in bits per second that the data is transmitted. is transmitted. (R)(R)

• Bit durationBit duration: The amount of time for one bit : The amount of time for one bit transmission transmission (1/R)(1/R)

• Modulation rate:Modulation rate: The rate at which the signal The rate at which the signal level is changed. level is changed. (baud rate, signal levels per (baud rate, signal levels per second)second)

TerminologyTerminology

• Encoding scheme:Encoding scheme: The mapping from data bits to The mapping from data bits to signal elementssignal elements

• Spectrum:Spectrum: The spectrum of a signal is the range The spectrum of a signal is the range of frequencies that it contains.of frequencies that it contains.

• Absolute bandwidth:Absolute bandwidth: The width of the spectrum The width of the spectrum• Effective bandwidth:Effective bandwidth: The are of the bandwidth The are of the bandwidth

where most of the energy of the signal is where most of the energy of the signal is concentrated.concentrated.

TerminologyTerminology

• DC (direct current)component:DC (direct current)component: A A component of a signal with the frequency of component of a signal with the frequency of zero.zero.

• ExampleExample– S(t)=1+(4/S(t)=1+(4/)sin(2 )sin(2 ft) + …. ft) + ….

Evaluation of Various Encoding Evaluation of Various Encoding Techniques (affecting factors)Techniques (affecting factors)

• Signal spectrum:Signal spectrum:– Lack of high frequency components means less Lack of high frequency components means less

bandwidth required for transmission bandwidth required for transmission

– DC component:DC component: It is desirable to have no DC It is desirable to have no DC component. (easier implementation)component. (easier implementation)

• ClockingClocking: The beginning and end of each bit : The beginning and end of each bit position must be determined.position must be determined.– Providing separate clocking information.Providing separate clocking information.

– Implementation of some other ways of synchronizationImplementation of some other ways of synchronization

Evaluation of Various Encoding Evaluation of Various Encoding Techniques (affecting factors)Techniques (affecting factors)

• Error detection:Error detection:– To detect errors more quickly, some error To detect errors more quickly, some error

detection techniques must be built into detection techniques must be built into signaling encoding methods. signaling encoding methods.

• Signal interference and noise immunity:Signal interference and noise immunity:– Some signal encoding techniques provide better Some signal encoding techniques provide better

error rate (BER) than otherserror rate (BER) than others

• Cost and complexityCost and complexity

Data EncodingData Encoding

Digital data, analog signalDigital data, analog signal

• A modem converts digital data to analog A modem converts digital data to analog datadata– Amplitude –shift keying Amplitude –shift keying (ASK)(ASK)– Frequency –shift keying Frequency –shift keying (FSK)(FSK)– Phase –shift keying Phase –shift keying (PSK)(PSK)

Data EncodingData Encoding

Analog data, Digital signalsAnalog data, Digital signals

• Pulse code modulation Pulse code modulation (PCM)(PCM)

– Samples analog data periodicallySamples analog data periodically– Quantizing (limiting the possible values to Quantizing (limiting the possible values to

discrete set of values) the samplesdiscrete set of values) the samples

Data EncodingData Encoding

Digital data, digital signalDigital data, digital signal

• Simplest form of digital encodingSimplest form of digital encoding

– Two voltage level requiredTwo voltage level required– It can be enhanced to improve performance.It can be enhanced to improve performance.

Digital–to-Digital Encoding Digital–to-Digital Encoding SchemesSchemes

• UnipolarUnipolar– Uses only one level of voltage Uses only one level of voltage

(almost obsolete)(almost obsolete)

• PolarPolar– Uses two level of voltageUses two level of voltage

• BipolarBipolar– Uses theree level of voltageUses theree level of voltage

Unipolar EncodingUnipolar Encoding

• Presence and absence of a voltage Presence and absence of a voltage level is used for two binary digits.level is used for two binary digits.

• The The absenceabsence of voltage could represent of voltage could represent zerozero..

• A constant A constant positive voltagepositive voltage could could represent represent 11..

UnipolarUnipolar

AmplitudeAmplitude

0 1 0 0 0Time

Unipolar Encoding IssuesUnipolar Encoding Issues

• Synchronization: A major issue:– Example: For a bit rate of 1000 bps, the

receiving device must measure each bit for 0.005 s.

• DC Component: • The average amplitude of a unipolar encoded

signal is not zero. – This creates a DC component ( a component with zero

frequency). – DC component can not travel through some media

that can not handle DC component

Polar EncodingPolar Encoding

Polar encoding uses Polar encoding uses tow voltage tow voltage levelslevels (positive and negative) (positive and negative)

Polar

NRZ RZ Biphase

NRZ-L NRZ-IManchester Differential

Manchester

Variation of Nonreturn to Zero Variation of Nonreturn to Zero (NRZ)(NRZ)

NRZ-L, Nonreturn to Zero-level (polar)

• The level of the signal depends on the type of the bit it represents (a positive voltage usually represents bit 0 and negative voltage represents the bit 1 (or vice versa)

– The problem exist when receiver needs to interpret long streams of 1 or zero.

Or NRZ-I (Nonreturn to Zero Invert on ones)

0 1 0 0

Time

111

Amplitude

Nonreturn to Zero-Level

0

Variation of Nonreturn to Zero Variation of Nonreturn to Zero (NRZ)(NRZ)

NRZ-I (Nonreturn to Zero Invert on ones)

• An inversion of voltage level represents a 1 bit.

• The transition between a positive and negative voltage represents a 1 not the voltage level itself.

• A 0 is represented by no change

• Still a string of zeros is a problem.

Time

Amplitude

Nonreturn to Zero, invert on ones

0 1 0 0 0 111 0

0 1 0 0Time

111

Amplitude

Nonreturn to Zero-LevelNonreturn to Zero, invert on ones

0 1 0 0 0 111

0

0

Return to ZeroReturn to Zero

• One solutionOne solution to to synchronization synchronization issueissue of NRZ-L and NRZ-I is of NRZ-L and NRZ-I is using RZusing RZ (Return to Zero) encoding schemes.(Return to Zero) encoding schemes.– It uses three valuesIt uses three values: positive, negative : positive, negative

and zero.and zero.– In RZ, the signal changes during each bit.In RZ, the signal changes during each bit.– A 1 bit is represented by positive-to zero A 1 bit is represented by positive-to zero

and a 0 bit by negative-to-zero.and a 0 bit by negative-to-zero.

0 1 0 0

Time

111

Return to Zero

These transitions can be used for synchronization

It requires two signal changes to encode one bit.(uses more bandwidth)

NRZ pros and consNRZ pros and cons

• ProsPros– Easy to engineerEasy to engineer– Make good use of bandwidthMake good use of bandwidth

• ConsCons– dc componentdc component– Lack of synchronization capabilityLack of synchronization capability

• Used for magnetic recordingUsed for magnetic recording• Not often used for signal transmissionNot often used for signal transmission

Polar

NRZ RZ Biphase

NRZ-L NRZ-IManchester Differential

Manchester

Biphase EncodingBiphase Encoding

• The most popular encoding to deal with the The most popular encoding to deal with the synchronization problem.synchronization problem.

• The signal changes at the middle of the bit The signal changes at the middle of the bit interval and continues to the opposite pole interval and continues to the opposite pole (dose not return to zero).(dose not return to zero).

• Types of biphase encodingTypes of biphase encoding::– ManchesterManchester

– Differential ManchesterDifferential Manchester

Biphase EncodingBiphase Encoding

• Manchester Encoding:Manchester Encoding:– The inversion at the middle of each bit is The inversion at the middle of each bit is

used for both synchronization and bit used for both synchronization and bit representationrepresentation

• i.e. Transition serves as clock and datai.e. Transition serves as clock and data

– Low to high represents oneLow to high represents one– High to low represents zeroHigh to low represents zero– Used by IEEE 802.3Used by IEEE 802.3

0 1 0 0 1 1 1 0

Zero One

Manchester Encoding

Differential EncodingDifferential Encoding

• Data represented by changes rather Data represented by changes rather than levelsthan levels

• More reliable detection of transition More reliable detection of transition rather than levelrather than level

• In complex transmission layouts it is In complex transmission layouts it is easy to lose sense of polarityeasy to lose sense of polarity

Biphase EncodingBiphase Encoding

• Differential Manchester:Differential Manchester:– Transition at the middle of bit interval is Transition at the middle of bit interval is

used for clocking only.used for clocking only.– Transition at the start of a bit period Transition at the start of a bit period

represents zero.represents zero.– No transition at start of a bit period No transition at start of a bit period

represents one.represents one.– Note: this is a differential encoding schemeNote: this is a differential encoding scheme– Used by IEEE 802.5.Used by IEEE 802.5.

Differential Manchester Encoding

0 1 0 0 1 1 1 0

Presence of transition at the beginning of the bit interval represents zero.Absence of transition at the beginning of the bit interval represents one.

Biphase Pros and ConsBiphase Pros and Cons

• ConCon– At least one transition per bit time and possibly At least one transition per bit time and possibly

twotwo– Maximum modulation rate is twice NRZMaximum modulation rate is twice NRZ– Requires more bandwidthRequires more bandwidth

• ProsPros– Synchronization on mid bit transition (self Synchronization on mid bit transition (self

clocking)clocking)– No dc componentNo dc component– Error detectionError detection

• Absence of expected transitionAbsence of expected transition

Multilevel BinaryMultilevel Binary

Use more than two levelsUse more than two levels

• Bipolar-AMI Bipolar-AMI (Alternate mark inversion)(Alternate mark inversion)

• Pseudoternary Pseudoternary (variation of Bipolar-(variation of Bipolar-AMI)AMI)

Bipolar EncodingBipolar Encoding

• Uses there voltage levels Uses there voltage levels – Positive, negative, and zeroPositive, negative, and zero

• Zero levelZero level represents represents binary 0binary 0

• One’sOne’s are represented by are represented by alternating alternating positive and negative voltagespositive and negative voltages

Types of Bipolar EncodingTypes of Bipolar Encoding

• Bipolar Alternate Mark Inversion Bipolar Alternate Mark Inversion (AMI) (AMI)

– Bipolar 8-zero substitution (B8ZS)Bipolar 8-zero substitution (B8ZS)

– High density bipolar 3 High density bipolar 3 (HDB3)(HDB3)

Types of Bipolar EncodingTypes of Bipolar Encoding

AM I B8ZS H D B3

B ipolar

Bipolar Alternate Mark Bipolar Alternate Mark Inversion (AMI)Inversion (AMI)

• Mark comes from telegraphy (meaning 1)Mark comes from telegraphy (meaning 1)

• Zero voltage represents zeroZero voltage represents zero

• Binary 1’s are represented by alternating Binary 1’s are represented by alternating positive and negative voltagespositive and negative voltages

0 1 0 0 1 1 1 0

Bipolar Alternate mark inversion (AMI)

Bipolar-AMI and PseudoternaryBipolar-AMI and Pseudoternary

Types of Bipolar EncodingTypes of Bipolar Encoding

• Pros:Pros:– DC component is zeroDC component is zero– A long sequence of 1’s is always A long sequence of 1’s is always

synchronized. synchronized. – Lower bandwidthLower bandwidth– Easy error detectionEasy error detection

• ConsCons– No mechanism for synchronization of long string of No mechanism for synchronization of long string of

zeroszeros

Variation of AMIVariation of AMI

• Bipolar 8-zero substitutionBipolar 8-zero substitution (B8ZS) (B8ZS)• (implemented in US)(implemented in US)• High Density bipolar 3High Density bipolar 3 (HDB3) (HDB3)• (implemented in Europe)(implemented in Europe)

– In both methods the original pattern is In both methods the original pattern is modified in the case of multiple modified in the case of multiple consecutive zeros.consecutive zeros.

Bipolar 8-zero substitutionBipolar 8-zero substitution (B8ZS)(B8ZS)

• It works similar to BMIIt works similar to BMI

• Whenever 8 or more consecutive Whenever 8 or more consecutive zeros occurs, signal level is forced to zeros occurs, signal level is forced to change. change.

PseudoternaryPseudoternary

• One represented by absence of line One represented by absence of line signalsignal

• Zero represented by alternating positive Zero represented by alternating positive and negativeand negative

• No advantage or disadvantage over No advantage or disadvantage over bipolar-AMIbipolar-AMI

Trade Off for Multilevel BinaryTrade Off for Multilevel Binary

• Not as efficient as NRZ Not as efficient as NRZ – Each signal element only represents one bitEach signal element only represents one bit

– In a 3 level system could represent logIn a 3 level system could represent log223 = 1.58 3 = 1.58

bitsbits– Receiver must distinguish between three Receiver must distinguish between three

levels levels (+A, -A, 0)(+A, -A, 0)

– Requires approx. 3dB more signal power for Requires approx. 3dB more signal power for same probability of bit errorsame probability of bit error

ScramblingScrambling

• Use scrambling to replace sequences that would Use scrambling to replace sequences that would produce constant voltageproduce constant voltage

• Filling sequence Filling sequence – Must produce enough transitions to syncMust produce enough transitions to sync– Must be recognized by receiver and replace with originalMust be recognized by receiver and replace with original– Same length as originalSame length as original

• No dc componentNo dc component• No long sequences of zero level line signalNo long sequences of zero level line signal• No reduction in data rateNo reduction in data rate• Error detection capabilityError detection capability

B8ZSB8ZS

• Bipolar With 8 Zeros SubstitutionBipolar With 8 Zeros Substitution• Based on bipolar-AMIBased on bipolar-AMI• If octet of all zeros and last voltage pulse If octet of all zeros and last voltage pulse

preceding was positive encode as 000+-0-+preceding was positive encode as 000+-0-+• If octet of all zeros and last voltage pulse If octet of all zeros and last voltage pulse

preceding was negative encode as 000-+0+-preceding was negative encode as 000-+0+-• Causes two violations of AMI codeCauses two violations of AMI code• Unlikely to occur as a result of noiseUnlikely to occur as a result of noise• Receiver detects and interprets as octet of all Receiver detects and interprets as octet of all

zeroszeros

HDB3HDB3

• High Density Bipolar 3 ZerosHigh Density Bipolar 3 Zeros

• Based on bipolar-AMIBased on bipolar-AMI

• String of four zeros replaced with one or String of four zeros replaced with one or two pulsestwo pulses

B8ZS and HDB3B8ZS and HDB3

Digital Data, Analog SignalDigital Data, Analog Signal

• Public telephone systemPublic telephone system– 300Hz to 3400Hz300Hz to 3400Hz– Use modem (modulator-demodulator)Use modem (modulator-demodulator)

• Amplitude shift keying (ASK)Amplitude shift keying (ASK)

• Frequency shift keying (FSK)Frequency shift keying (FSK)

• Phase shift keying (PK)Phase shift keying (PK)

Digital to Analog Encoding

Amplitude Shift KeyingAmplitude Shift Keying

• Values represented by different amplitudes of Values represented by different amplitudes of carriercarrier

• Usually, one amplitude is zeroUsually, one amplitude is zero– i.e. presence and absence of carrier is usedi.e. presence and absence of carrier is used

• Susceptible to sudden gain changesSusceptible to sudden gain changes• InefficientInefficient• Up to 1200bps on voice grade linesUp to 1200bps on voice grade lines• Used over optical fiberUsed over optical fiber

Modulation Techniques (ASK)Modulation Techniques (ASK)

)2cos()( tAts fc

0)( tsBinary 1

Binary 0

ASK

Modulation Techniques(ASK)Modulation Techniques(ASK)

Frequency Shift KeyingFrequency Shift Keying

• Values represented by different Values represented by different frequencies (near carrier)frequencies (near carrier)

• Less susceptible to error than ASKLess susceptible to error than ASK

• Up to 1200bps on voice grade linesUp to 1200bps on voice grade lines

• High frequency radioHigh frequency radio

• Even higher frequency on LANs using Even higher frequency on LANs using co-axco-ax

Modulation Techniques (ASK)Modulation Techniques (ASK)

)2cos()(1tAts f Binary 1

Binary 0)2cos()(2tAts f

f1 and f2 are offset from fc by equal but opposite amount

FSK on Voice Grade LineFSK on Voice Grade Line

FSK

Modulation Techniques(FSK)Modulation Techniques(FSK)

Phase Shift KeyingPhase Shift Keying

• Phase of carrier signal is shifted to Phase of carrier signal is shifted to represent datarepresent data

• Differential PSKDifferential PSK– Phase shifted relative to previous Phase shifted relative to previous

transmission rather than some reference transmission rather than some reference signalsignal

Modulation Techniques (PSK)Modulation Techniques (PSK)(Differential PSK)(Differential PSK)

)2cos()( tAts fc

Binary 1

Binary 0)2cos()( tAts fc

The phase shift is is in reference to previous bit transmittedRather than to some constant reference signal.

PSK

PSKConstellation

Quadrature PSKQuadrature PSK

• More efficient use by each signal More efficient use by each signal element representing more than one bitelement representing more than one bit– e.g. shifts of e.g. shifts of /2 (90/2 (90oo))– Each element represents two bitsEach element represents two bits– Can use 8 phase angles and have more Can use 8 phase angles and have more

than one amplitudethan one amplitude– 9600bps modem use 12 angles , four of 9600bps modem use 12 angles , four of

which have two amplitudeswhich have two amplitudes

Modulation Techniques (PSK)Modulation Techniques (PSK)(Differential QPSK)(Differential QPSK)

)4/2cos()( tAts fc

Binary 11

Binary 10)4/32cos()( tAts fc

)4/52cos()( tAts fc

)4/72cos()( tAts fc

Binary 00

Binary 01

4-PSK

4-PSKConstellation

8-QAM Signal

8-PSKConstellation

Have a great day .See you on Friday.

PSKBandwidth

4-QAM and 8-QAMConstellation

Bandwidth for ASK

Bandwidth for FSK

16-QAMConstellation

Bit Rate and Baud Rate

Bit Rate and Baud Rate

Modulation Techniques(FSK)Modulation Techniques(FSK)

Performance of Digital to Analog Performance of Digital to Analog Modulation SchemesModulation Schemes

• BandwidthBandwidth– ASK and PSK bandwidth directly related to bit rateASK and PSK bandwidth directly related to bit rate– FSK bandwidth related to data rate for lower FSK bandwidth related to data rate for lower

frequencies, but to offset of modulated frequency frequencies, but to offset of modulated frequency from carrier at high frequenciesfrom carrier at high frequencies

– (See Stallings for math)(See Stallings for math)

• In the presence of noise, bit error rate of PSK In the presence of noise, bit error rate of PSK and QPSK are about 3dB superior to ASK and QPSK are about 3dB superior to ASK and FSKand FSK

Analog Data, Digital SignalAnalog Data, Digital Signal

• DigitizationDigitization– Conversion of analog data into digital dataConversion of analog data into digital data– Digital data can then be transmitted using NRZ-LDigital data can then be transmitted using NRZ-L– Digital data can then be transmitted using code Digital data can then be transmitted using code

other than NRZ-Lother than NRZ-L– Digital data can then be converted to analog signalDigital data can then be converted to analog signal– Analog to digital conversion done using a codecAnalog to digital conversion done using a codec– Pulse code modulationPulse code modulation– Delta modulationDelta modulation

Pulse Code Modulation(PCM) Pulse Code Modulation(PCM) (1)(1)

• If a signal is sampled at regular intervals at a If a signal is sampled at regular intervals at a rate higher than twice the highest signal rate higher than twice the highest signal frequency, the samples contain all the frequency, the samples contain all the information of the original signalinformation of the original signal– (Proof - Stallings appendix 4A)(Proof - Stallings appendix 4A)

• Voice data limited to below 4000HzVoice data limited to below 4000Hz• Require 8000 sample per secondRequire 8000 sample per second• Analog samples (Pulse Amplitude Modulation, Analog samples (Pulse Amplitude Modulation,

PAM)PAM)• Each sample assigned digital valueEach sample assigned digital value

Pulse Code Modulation(PCM) Pulse Code Modulation(PCM) (2)(2)

• 4 bit system gives 16 levels4 bit system gives 16 levels• QuantizedQuantized

– Quantizing error or noiseQuantizing error or noise– Approximations mean it is impossible to recover Approximations mean it is impossible to recover

original exactlyoriginal exactly

• 8 bit sample gives 256 levels8 bit sample gives 256 levels• Quality comparable with analog transmissionQuality comparable with analog transmission• 8000 samples per second of 8 bits each gives 8000 samples per second of 8 bits each gives

64kbps64kbps

Nonlinear EncodingNonlinear Encoding

• Quantization levels not evenly spacedQuantization levels not evenly spaced

• Reduces overall signal distortionReduces overall signal distortion

• Can also be done by compandingCan also be done by companding

Delta ModulationDelta Modulation

• Analog input is approximated by a Analog input is approximated by a staircase functionstaircase function

• Move up or down one level (Move up or down one level () at each ) at each sample intervalsample interval

• Binary behaviorBinary behavior– Function moves up or down at each Function moves up or down at each

sample intervalsample interval

Delta Modulation - exampleDelta Modulation - example

Delta Modulation - OperationDelta Modulation - Operation

Delta Modulation - PerformanceDelta Modulation - Performance

• Good voice reproduction Good voice reproduction – PCM - 128 levels (7 bit)PCM - 128 levels (7 bit)– Voice bandwidth 4khzVoice bandwidth 4khz– Should be 8000 x 7 = 56kbps for PCMShould be 8000 x 7 = 56kbps for PCM

• Data compression can improve on thisData compression can improve on this– e.g. Interframe coding techniques for videoe.g. Interframe coding techniques for video

Analog Data, Analog SignalsAnalog Data, Analog Signals

• Why modulate analog signals?Why modulate analog signals?– Higher frequency can give more efficient Higher frequency can give more efficient

transmissiontransmission– Permits frequency division multiplexing (chapter 8)Permits frequency division multiplexing (chapter 8)

• Types of modulationTypes of modulation– AmplitudeAmplitude– FrequencyFrequency– PhasePhase

Analog Analog ModulationModulation

Spread SpectrumSpread Spectrum

• Analog or digital dataAnalog or digital data• Analog signalAnalog signal• Spread data over wide bandwidthSpread data over wide bandwidth• Makes jamming and interception harderMakes jamming and interception harder• Frequency hopingFrequency hoping

– Signal broadcast over seemingly random series of Signal broadcast over seemingly random series of frequenciesfrequencies

• Direct SequenceDirect Sequence– Each bit is represented by multiple bits in transmitted Each bit is represented by multiple bits in transmitted

signalsignal– Chipping codeChipping code

Required ReadingRequired Reading

• Stallings chapter 5Stallings chapter 5

ReviewReview

Atmospheric and Atmospheric and Extraterrestrial NoiseExtraterrestrial Noise

–Lightning:Lightning: It is a major source of noise, caused It is a major source of noise, caused by the static discharge of thunderclouds.by the static discharge of thunderclouds.

•Several million voltsSeveral million volts

•Currents exceeding 20,000 amps.Currents exceeding 20,000 amps.

–Solar NoiseSolar Noise: Ionized gases of the sun produces : Ionized gases of the sun produces a wide range of frequencies that penetrate the a wide range of frequencies that penetrate the Earth’s atmosphere.Earth’s atmosphere.–Cosmic NoiseCosmic Noise Radiation of noise by distant stars Radiation of noise by distant stars penetrating the Earth’s atmosphere.Long haul penetrating the Earth’s atmosphere.Long haul telecommunications service (1500 km support telecommunications service (1500 km support 20,000 to 60,000 voice channels)20,000 to 60,000 voice channels)

•An alternative to fiber optic and coaxial cableAn alternative to fiber optic and coaxial cable•Short point-to-point links between buildings Short point-to-point links between buildings (closed-circuit TV or data link)(closed-circuit TV or data link)