Data and Computer Data and Computer Communications Communications

(1)

Data and Computer Data and Computer

Communications Communications

Chapter 5 – Signal Encoding Chapter 5 – Signal Encoding

Techniques

(2)

Signal Encoding Techniques Signal Encoding Techniques

Even the natives have difficulty mastering this peculiar vocabulary

—The Golden Bough, Sir James George Frazer

(3)

Signal Encoding Techniques

(4)

Digital Data, Digital Signal Digital Data, Digital Signal

 Digital signal Digital signal



discrete, discontinuous voltage pulses discrete, discontinuous voltage pulses



each pulse is a signal element each pulse is a signal element



binary data encoded into signal elements binary data encoded into signal elements

(5)

Some Terms Some Terms

 unipolar unipolar

 polar polar

 data rate data rate

 duration or length of a bit duration or length of a bit

 modulation rate modulation rate

 mark and space mark and space

(6)

Interpreting Signals Interpreting Signals

 need to know need to know



timing of bits - when they start and end timing of bits - when they start and end



signal levels signal levels

 factors affecting signal interpretation factors affecting signal interpretation



signal to noise ratio signal to noise ratio



data rate data rate



bandwidth bandwidth



encoding scheme encoding scheme

(7)

Comparison of Encoding Comparison of Encoding

Schemes Schemes

 signal spectrum signal spectrum

 clocking clocking

 error detection error detection

 signal interference and noise immunity signal interference and noise immunity

 cost and complexity cost and complexity

(8)

Encoding Schemes

(9)

Nonreturn to Zero-Level Nonreturn to Zero-Level

(NRZ-L) (NRZ-L)

 two different voltages for 0 and 1 bits two different voltages for 0 and 1 bits

 voltage constant during bit interval voltage constant during bit interval



no transition I.e. no return to zero voltage no transition I.e. no return to zero voltage



such as absence of voltage for zero, constant such as absence of voltage for zero, constant positive voltage for one

positive voltage for one



more often, negative voltage for one value more often, negative voltage for one value and positive for the other

and positive for the other

(10)

Nonreturn to Zero Inverted Nonreturn to Zero Inverted

 nonreturn to zero inverted on ones nonreturn to zero inverted on ones

 constant voltage pulse for duration of bit constant voltage pulse for duration of bit

 data encoded as presence or absence of signal data encoded as presence or absence of signal transition at beginning of bit time

transition at beginning of bit time



transition (low to high or high to low) denotes binary 1 transition (low to high or high to low) denotes binary 1



no transition denotes binary 0 no transition denotes binary 0

 example of differential encoding since have example of differential encoding since have



data represented by changes rather than levels data represented by changes rather than levels



more reliable detection of transition rather than level more reliable detection of transition rather than level



easy to lose sense of polarity easy to lose sense of polarity

(11)

NRZ Pros & Cons NRZ Pros & Cons

 Pros Pros



easy to engineer easy to engineer



make good use of bandwidth make good use of bandwidth

 Cons Cons



dc component dc component



lack of synchronization capability lack of synchronization capability

 used for magnetic recording used for magnetic recording

 not often used for signal transmission not often used for signal transmission

(12)

Multilevel Binary Multilevel Binary

Bipolar-AMI Bipolar-AMI

 Use more than two levels Use more than two levels

 Bipolar-AMI Bipolar-AMI



zero represented by no line signal zero represented by no line signal



one represented by positive or negative pulse one represented by positive or negative pulse



one pulses alternate in polarity one pulses alternate in polarity



no loss of sync if a long string of ones no loss of sync if a long string of ones



long runs of zeros still a problem long runs of zeros still a problem



no net dc component no net dc component



lower bandwidth lower bandwidth



easy error detection easy error detection

(13)

Multilevel Binary Multilevel Binary

Pseudoternary Pseudoternary

 one represented by absence of line signal one represented by absence of line signal

 zero represented by alternating positive zero represented by alternating positive and negative

and negative

 no advantage or disadvantage over no advantage or disadvantage over bipolar-AMI

bipolar-AMI

 each used in some applications each used in some applications

(14)

Multilevel Binary Issues Multilevel Binary Issues

 synchronization with long runs of 0’s or 1’s synchronization with long runs of 0’s or 1’s



can insert additional bits, cf ISDN can insert additional bits, cf ISDN



scramble data (later) scramble data (later)

 not as efficient as NRZ not as efficient as NRZ



each signal element only represents one bit each signal element only represents one bit

• receiver distinguishes between three levels: +A, -A, 0 receiver distinguishes between three levels: +A, -A, 0



a 3 level system could represent log a 3 level system could represent log

₂₂

3 = 1.58 bits 3 = 1.58 bits



requires approx. 3dB more signal power for same requires approx. 3dB more signal power for same probability of bit error

probability of bit error

(15)

Manchester Encoding Manchester Encoding

 has transition in middle of each bit period has transition in middle of each bit period

 transition serves as clock and data transition serves as clock and data

 low to high represents one low to high represents one

 high to low represents zero high to low represents zero

 used by IEEE 802. used by IEEE 802.

(16)

Differential Manchester Differential Manchester

Encoding Encoding

 midbit transition is clocking only midbit transition is clocking only

 transition at start of bit period representing 0 transition at start of bit period representing 0

 no transition at start of bit period representing 1 no transition at start of bit period representing 1



this is a differential encoding scheme this is a differential encoding scheme

 used by IEEE 802.5 used by IEEE 802.5

(17)

Biphase Pros and Cons Biphase Pros and Cons

 Con Con



at least one transition per bit time and possibly two at least one transition per bit time and possibly two



maximum modulation rate is twice NRZ maximum modulation rate is twice NRZ



requires more bandwidth requires more bandwidth

 Pros Pros



synchronization on mid bit transition (self clocking) synchronization on mid bit transition (self clocking)



has no dc component has no dc component



has error detection has error detection

(18)

Modulation Rate

(19)

Scrambling Scrambling

 use scrambling to replace sequences that would use scrambling to replace sequences that would produce constant voltage

produce constant voltage

 these filling sequences must these filling sequences must



produce enough transitions to sync produce enough transitions to sync



be recognized by receiver & replaced with original be recognized by receiver & replaced with original



be same length as original be same length as original

 design goals design goals



have no dc component have no dc component



have no long sequences of zero level line signal have no long sequences of zero level line signal



have no reduction in data rate have no reduction in data rate



give error detection capability give error detection capability

(20)

B8ZS and HDB3

(21)

Digital Data, Analog Signal Digital Data, Analog Signal

 main use is public telephone system main use is public telephone system



has freq range of 300Hz to 3400Hz has freq range of 300Hz to 3400Hz



use modem (modulator-demodulator) use modem (modulator-demodulator)

 encoding techniques encoding techniques



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



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



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

(22)

Modulation Techniques

(23)

Amplitude Shift Keying Amplitude Shift Keying

 encode 0/1 by different carrier amplitudes encode 0/1 by different carrier amplitudes



usually have one amplitude zero usually have one amplitude zero

 susceptible to sudden gain changes susceptible to sudden gain changes

 inefficient inefficient

 used for used for



up to 1200bps on voice grade lines up to 1200bps on voice grade lines



very high speeds over optical fiber very high speeds over optical fiber

(24)

Binary Frequency Shift Binary Frequency Shift

Keying Keying

 most common is binary FSK (BFSK) most common is binary FSK (BFSK)

 two binary values represented by two different two binary values represented by two different frequencies (near carrier)

frequencies (near carrier)

 less susceptible to error than ASK less susceptible to error than ASK

 used for used for



up to 1200bps on voice grade lines up to 1200bps on voice grade lines



high frequency radio high frequency radio



even higher frequency on LANs using co-ax even higher frequency on LANs using co-ax

(25)

Multiple FSK Multiple FSK

 each signalling element represents more each signalling element represents more than one bit

than one bit

 more than two frequencies used more than two frequencies used

 more bandwidth efficient more bandwidth efficient

 more prone to error more prone to error

(26)

Phase Shift Keying Phase Shift Keying

 phase of carrier signal is shifted to phase of carrier signal is shifted to represent data

represent data

 binary PSK binary PSK



two phases represent two binary digits two phases represent two binary digits

 differential PSK differential PSK



phase shifted relative to previous transmission phase shifted relative to previous transmission rather than some reference signal

rather than some reference signal

(27)

Quadrature PSK Quadrature PSK

 get more efficient use if each signal get more efficient use if each signal element represents more than one bit element represents more than one bit



eg. shifts of eg. shifts of   /2 (90 /2 (90

^o^o

) )



each element represents two bits each element represents two bits



split input data stream in two & modulate onto split input data stream in two & modulate onto carrier & phase shifted carrier

carrier & phase shifted carrier

 can use 8 phase angles & more than one can use 8 phase angles & more than one amplitude

amplitude



9600bps modem uses 12 angles, four of 9600bps modem uses 12 angles, four of which have two amplitudes

which have two amplitudes

(28)

QPSK and OQPSK QPSK and OQPSK

Modulators

(29)

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

 bandwidth bandwidth



ASK/PSK bandwidth directly relates to bit rate ASK/PSK bandwidth directly relates to bit rate



multilevel PSK gives significant improvements multilevel PSK gives significant improvements

 in presence of noise: in presence of noise:



bit error rate of PSK and QPSK are about 3dB bit error rate of PSK and QPSK are about 3dB superior to ASK and FSK

superior to ASK and FSK



for MFSK & MPSK have tradeoff between for MFSK & MPSK have tradeoff between

bandwidth efficiency and error performance

(30)

Quadrature Amplitude Quadrature Amplitude

Modulation Modulation

 QAM used on asymmetric digital subscriber line QAM used on asymmetric digital subscriber line (ADSL) and some wireless

(ADSL) and some wireless

 combination of ASK and PSK combination of ASK and PSK

 logical extension of QPSK logical extension of QPSK

 send two different signals simultaneously on send two different signals simultaneously on same carrier frequency

same carrier frequency



use two copies of carrier, one shifted 90 use two copies of carrier, one shifted 90

^°^°



each carrier is ASK modulated each carrier is ASK modulated



two independent signals over same medium two independent signals over same medium



demodulate and combine for original binary output demodulate and combine for original binary output

(31)

QAM Modulator

(32)

QAM Variants QAM Variants

 two level ASK two level ASK



each of two streams in one of two states each of two streams in one of two states



four state system four state system



essentially QPSK essentially QPSK

 four level ASK four level ASK



combined stream in one of 16 states combined stream in one of 16 states

 have 64 and 256 state systems have 64 and 256 state systems

 improved data rate for given bandwidth improved data rate for given bandwidth



but increased potential error rate but increased potential error rate

(33)

Analog Data, Digital Signal Analog Data, Digital Signal

 digitization is conversion of analog data digitization is conversion of analog data into digital data which can then:

into digital data which can then:



be transmitted using NRZ-L be transmitted using NRZ-L



be transmitted using code other than NRZ-L be transmitted using code other than NRZ-L



be converted to analog signal be converted to analog signal

 analog to digital conversion done using a analog to digital conversion done using a codec

codec



pulse code modulation pulse code modulation



delta modulation delta modulation

(34)

Digitizing Analog Data

(35)

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

 sampling theorem: sampling theorem:



“ “ 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 information frequency, the samples contain all information

in original signal”



eg. 4000Hz voice data, requires 8000 sample eg. 4000Hz voice data, requires 8000 sample per sec

per sec

 strictly have analog samples strictly have analog samples



Pulse Amplitude Modulation (PAM) Pulse Amplitude Modulation (PAM)

 so assign each a digital value so assign each a digital value

(36)

PCM Example

(37)

PCM Block Diagram

(38)

Non-Linear Coding

(39)

Companding

(40)

Delta Modulation Delta Modulation

 analog input is approximated by a analog input is approximated by a staircase function

staircase function



can move up or down one level ( can move up or down one level (   ) at each ) at each sample interval

sample interval

 has binary behavior has binary behavior



since function only moves up or down at each since function only moves up or down at each sample interval

sample interval



hence can encode each sample as single bit hence can encode each sample as single bit



1 for up or 0 for down 1 for up or 0 for down

(41)

Delta Modulation Example

(42)

Delta Modulation Operation

(43)

PCM verses Delta Modulation PCM verses Delta Modulation

 DM has simplicity compared to PCM DM has simplicity compared to PCM

 but has worse SNR but has worse SNR

 issue of bandwidth used issue of bandwidth used



eg. for good voice reproduction with PCM eg. for good voice reproduction with PCM

• want 128 levels (7 bit) & voice bandwidth 4khz want 128 levels (7 bit) & voice bandwidth 4khz

• need 8000 x 7 = 56kbps need 8000 x 7 = 56kbps

 data compression can improve on this data compression can improve on this

 still growing demand for digital signals still growing demand for digital signals



use of repeaters, TDM, efficient switching use of repeaters, TDM, efficient switching

 PCM preferred to DM for analog signals PCM preferred to DM for analog signals

(44)

Analog Data, Analog Signals Analog Data, Analog Signals

 modulate carrier frequency with analog data modulate carrier frequency with analog data

 why modulate analog signals? why modulate analog signals?



higher frequency can give more efficient transmission higher frequency can give more efficient transmission

