Applied Network Research Group Department of Computer Engineering, Kasetsart University
Data Encoding
Surasak Sanguanpong nguan@ku.ac.th http://www.cpe.ku.ac.th/~nguan
Last updated: 11 July 2000
Data Encoding - 2/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Encoding scheme
voice
Telephone analog
digital
Modem analog
analog
CODEC digital
digital Digital transmitter
digital
Analog data, Analog signal
Digital data, Analog signal Digital data, Digital signal Analog data, Digital signal
Analog transmission means a transmitting analog signals without regard to their content; the signal may represent analog data (e.g. voice) or digital data (e.g. binary data). Digital transmission is concerned with a transmitting binary signal
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Encoding and Modulation
Encoder
Encoder Decoder
DecoderModulator
Modulator DemodulatorDemodulator digital
or analog
digital or analog
digital
analog
g(t)
m(t)
fc
s(f) x(t)
t
fc f g(t)
m(t)
x(t)
s(t)
For digital signal, a data source g(t), which may be either digital or analog, is encoded into a digital signal x(t).
Analog transmission uses a continuous constant-frequency signal known as the carrier signal. The frequency of the carrier signal is chosen to be compatible with the transmission medium. Data is transmitted using a carrier signal by modulation, which is the process of encoding source data onto a carrier signal with frequency fc
Data Encoding - 4/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Why encoding?
z Three factors determine successfulness of receiving signal
z S/N
z data rate
z bandwidth
z More factor can be used to improve
z
encoding scheme
With other factors held constant, the following statements are true.
• An increase in data rate increases bit error rate.
• An increase S/N decreases bit error rate.
• An increase in bandwidth allows an increase in data rate [Stalling, p98].
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Encoding evaluation factors
z Signal spectrum
z Clocking
z Error detection
z Signal interference& noise immunity
z Cost and complexity
Five factors are used to evaluate the various encoding scheme:
Signal spectrum : a lack of high-frequency components means that less bandwidth is required for transmission. No dc component is desirable.
Clocking : suitable encoding provide some synchronization mechanism to determine the beginning and end of each bit position.
Error detection : some error detection can be built into the encoding scheme.
Signal interference & noise immunity : some encoding scheme has superior performance in the presence of noise.
Cost and complexity : higher signaling rate to achieve a greater data rate results expensive devices.
Data Encoding - 6/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Digital data, Digital signal
0 1 0 0 1 1 0 0 0 1 1
NRZ NRZI Bipolar -AMI
Pseudoternary Manchester Differential Manchester
Definition of Digital Signal Encoding Formats Nonreturn-to-Zero-Level (NRZ-L)
0 = high level 1 = low level l
Nonreturn to Zero Inverted (NRZI)
0 = no transition at beginning of interval (one bit time) 1 = transition at beginning of interval
Bipolar-AMI
0 = no line signal
1 = positive or negative level, alternating for successive ones Pseudoternary
0 = positive or negative level, alternating for successive zeroes 1 = no line signal
Manchester
0 = transition from high to low in middle of interval 1 = transition from high to low in middle of interval Differential Manchester
Always a transition in middle of interval 0 = no transition at beginning of interval 1 = transition at beginning of interval
[Stallings, p99,100]
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Scrambling techniques
1 1 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0
0 0 0
0 0 0
V B 0 V B
V B 0 0 V B 0 0 V
Bit value
B8ZS Bipolar-AMI
HDB3
To maintain synchronization for the receiver’s clock using bipolar.
B8ZS
Same as bipolar AMI, except that any string of eight zeros is replaced by a string with two code violations.
• If an octet of all zeros occur and the last voltage pulse preceding this octet was positive, the eight zeros of the octet are encode as 00+-0-+
• If an octet of all zeros occur and the last voltage pulse preceding this Oct was positive, the eight zeros of the octet are encode as 00-+0+-
HD3B
Same as bipolar AMI, except that any string of four zeros is replaced by a string with one code violation. The scheme replace strings of four zeros with the sequence B00V. [Stallings, p106]
Data Encoding - 8/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Spectral density
-0.5 0 0.5
1 1.5
0 0.5 1 1.5
NRZ-L, NRZI
B8ZS,HDB3
AMI, Pseudoternary
Manchester,
Differential Manchester
Meansquarevoltageperunitbandwidth
Normalized frequency (f/r)
NRZ make efficient use of bandwidth. most of the frequency in NRZ and NRZI signals are between dc and half the bit rate.
Manchester& Different Manchester has the bulk of the energy between one-half and one times the bit rate. Thus the bandwidth is reasonably narrow and contain no dc component.
AMI make use of bandwidth less than the bandwidth of NRZ [Stallings, p102].
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Digital data, Analog signal
z
Transmitting digital data through PSTN
z
Modem is used to convert digital data to analog signal and vice versa
The most familiar use is for transmitting digital data through the public telephone network. The telephone network was designed to support analog signals in the voice-frequency range 300-3400 Hz. It is not suitable for handling digital signal. Modem is used to convert digital to analog signal and vice versa.
Data Encoding - 10/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Modulation techniques
0 0 1 1 0 1 0 0 0 1 0
ASK
FSK PSK
There are three basic modulation techniques for transforming digital data into analog signals:
Amplitude-shift keying (ASK) Frequency-shift keying (FSK) Phase-shift keying (PSK)
Applied Network Research Group Department of Computer Engineering, Kasetsart University
ASK
V
d(t) V
c(t) V
ASK(t)
fc fc-f0
fc-3f0 fc+f0 fc+3f0 Signal
power
Frequency
frequency spectrum
In ASK, the amplitude of a single-frequency known as the carrier frequency is switched between two levels at a rate determined by the bit rate of the transmitted binary data signal. Bandpass filter is used to limit the band of frequencies based on Nyquist’s theorem. [Halsall, p.59]
Data Encoding - 12/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
FSK
Carrier 2 Datasignal
Carrier 1
v
d(t) v
1(t)
v
2(t) v
FSK(t)
f1 Signal
power
Frequency
frequency spectrum
f2
In FSK, two fixed amplitude carrier signals are used, one for a binary 0 and the other for a binary 1. The different between the two carriers is known as the frequency shift. The modulation operation is equivalent to summing together the outputs of two separate ASK modulators.
FSK is the modulation method that was used in all early low bit rate modems.
[Halsall p.62]
Applied Network Research Group Department of Computer Engineering, Kasetsart University
FSK in modem
400 980 (1070)
1850 (2225) 1180
(1270) 1650 (2025)
3400 Amplitude
Frequency(Hz) PSTN bandwidth
The above figure illustrate the frequency assignments that are used for two types of FSK modem to provide a full-duplex 300 bps link between two DTEs. One set of frequency assignments is defined by EIA and the other by ITU-T (V.21)
Modulator
Space=1180 Demodulator Space=1180 D T E
D T E Modulator
Space=1070 Mark=1270 Demodulator Space=2025 Mark=2225
Demodulator Space=1070 Mark=1270
Modulator Space=2025
Mark=2225
EIA frequency assignment
Data Encoding - 14/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
PSK
DataSignal
Carrier
Phase coherent
Differential vc(t)
vc(t)
vPSK(t)
v’PSK(t)
180=0 0=1
phase diagram bit rate = signaling rate
In PSK, the phase of the carrier signal is shift to represent data. Two type of PSK are used.
Phase coherent PSK : Used two fixed carrier signals to represent a binary 0 and 1 with a 180° phase different. The disadvantage of this scheme us that a reference carrier signal is required at the receiver against with the phase of the received signal is compared, this required more complex demodulation circuit.
Differential PSK : A phase shift of 90° relative to the current signal indicates a binary 0 is the next bit while a phase shift 270° indicates a binary 1. The demodulation circuitry need determine only the magnitude of each phase shift rather than its absolute value. [Halsall, p.64]
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Multilevel modulation method 0
1 0 10
0
0 ° +90 ° +180 ° +270 °
11
bit rate = n x signaling rate
More sophisticated modulation methods are used which involve either multiple signal levels or a mix of the basic scheme, particularly amplitude and phase. More bit rate can be achieved if signaling element represent more than one bit.
Data Encoding - 16/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Multilevel modulation method
+90°°°°=01
0°°°°=00 +270°°°°=11 +180°°°°=11
4-PSK phase diagram 16-QAM phase diagram
QPSK ( Quadrature PSK or 4-PSK) : Four different phase changes (0°, 90
° , 180° , 270° ) to enables each phase change to convey 2 bits (bit rate=2*signaling rate).
QAM (Quadrature Amplitude Modulation or 16-QAM) : Phase and amplitude changes, 16 levels per signal element and hence 4 bit symbols (bit rate=4*signaling rate).
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Bit rate and Baud rate
z
Bit rate
:A number of bits that are transmitted in a secondz
Baud rate
:A number of line signal changed variation per secondIf a modem transmits 1 bit for every signal change bit rate = baud rate
If a signal change represents 2 or more or n bits bit rate = baud rate*n
The relationship between bit transfer rate and baud rate depends on the number of bit values that are encoded in a signal. When each signal represent one bit, the bit and baud rate are the same. When a signal encodes multiple bits, the bit rate is a multiple of the baud rate.
In modem, encoding techniques are employed to make a signal change represent 2 or more bits.
The term baud comes from Baudot, who developed an encoding scheme for the French Telegraph system in 1877.
Data Encoding - 18/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Analog data, Digital signal
z
Two principle techniques used
z PCM
z DM
Analog voice signal
Sampling clock
signalPAM PCM
signal
Sampling Circuit Sampling
Circuit
Quantizer
&
compander Quantizer
&
compander
Digitized voice signal
Digitization is a process of converting analog data into digital data. The digital signal is converted back into analog signal at the receiver.
The device used for converting analog data into digital form, and recovering the original analog data is known as CODEC (Coder-Decoder).
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Nyquist theorem
“ In a perfectly noiseless channel, if f is the maximum frequency the medium can transmit, the receiver can completely reconstruct a signal by sampling it2*ftimes per second”
Nyquist, 1920
In 1920, Harry Nyquist developed his classic theory. Nyquist showed that original signal must be sampled at a maximum rate of greater than twice the h i g h e s t f r e q u e n c y c o m p o n e n t t o se n d t o r e c e i ve r t o c o m p l e t e l y reconstruction. For example, to convert a voice signal which 4 kHz highest frequency into digital form, it must be sampled at a rate of 8000 times per second.
Data Encoding - 20/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
PCM
z
Sampling signal based on nyquist theorem
3.2 3.9
2.8 3.4 1.2
4.2
3 4 3 3
11
4
011 100 011 011 001 100
Original signal
PAM pulse
PCM pulse with quantized error
011100011011001100
PCM output
PCM is based on the sampling theorem. The original signal is assumed to be band limited with a bandwidth of B. Signal is sampled at a rate 2B. Samples signal are represents as narrow pulse whose amplitude is proportional to the value of the original signal and is known as PAM (Pulse Amplitude Modulation).
The amplitude of each PAM pulse is approximated by an n-bit integer,. In the sample above, n=3. Thus 8=23 levels are used for approximating the PAM pulses.
Applied Network Research Group Department of Computer Engineering, Kasetsart University
PCM conversion process
A B C D
A B C D Analog
voice signal
Sampling
clock PAM
signal PCM
signal
Sampling Circuit Sampling
Circuit
Quantizer
&
compander Quantizer
&
compander
Digitized voice signal
The process starts with an analog signal, which is sampled by PAM sample.
the resulting pulse are quantized to produced PCM pulses and then encoded to produce bit stream. At the receiver end, the process is reversed to reproduce the analog signal. [Halsall, p.69]
Data Encoding - 22/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Nonlinear encoding
0 1 2 3 4 5 6 7 12 13 14 15
Strong signal Weak signal
0 1 2 3 4 5678 8
9
9 10 10
11 11
12 13 14 Without nonlinear encoding With nonlinear encoding15
Quantization levels are not necessary equally spaced. The problem with equal spacing is that the mean absolute error for each sample is the same, regardless the signal level. Lower amplitude values are relatively more distorted.
[Stallings, p.118]
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Companding process
z Implement nonlinear encoding via companding process
z Companding =
Compressing Expanding
z Implement nonlinear encoding via companding process
z Companding =
Compressing Expanding
linear quantization interval
11100100 11100100 11100100
11100100 11
10010011 10010011 10010011 100100
11 10 01 00
11 10 01 00
Network Compressor
circuit Compressor
circuit Linear
ADC Linear
ADC Linear
DAC Linear
DAC Expander
circuit Expander
circuit
vi
vi v’o
vo
Prior to the input signal being sampled and converted by ADC into a digital form, it is passed through a circuit known as a compressor. Similarly, at the destination, the reverse operation is perform on the output of the DAC by a circuit known as expander. [Halsall, p. 71]
Data Encoding - 24/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Analog Data, Analog signal
z
Use Modulation techniques
z
Need a high frequency for effective transmission
z
Modulation permits frequency division multiplex
Modulation has been defined as the process of combining an input signal and a carrier signal.
Applied Network Research Group Department of Computer Engineering, Kasetsart University
Analog Modulation
Carrier
Modulating sine-wave signal
Amplitude-modulated wave
Frequency-modulated wave
Phase-modulated wave
Amplitude modulation is the simplest form of modulation. The modulated signal has constant frequency but its amplitude is vary with the input signal.
The envelope of the resulting signal is 1+nax(t) and as long as na<1, the envelope is an exact reproduction of the original signal. If na>1 , the envelope will cross the time axis and information is lost. [Stallings, p.123]
Data Encoding - 26/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
AM Spectrum
carrier lower sideband upper sideband
•
each sideband contains the complete spectrum of s(t) !
M(f)
0 f
c- B f
Discrete carrier term Lower sideband Upper sideband
f
c+ B f
c2 2
[ ]
casex t f t
s t na f t f t
s t f t n
f f t n
f f t
m
m c
c a
c m
a
c m
( ) cos
( ) cos cos
( ) cos cos ( ) cos ( )
=
= +
= + − + +
2
1 2 2
2 2 2
π
π π
π π π
The resulting signal has a component at the original carrier frequency plus a pair of components each spaced fmHz from the carrier.
From the equation above, it can be seen that AM involves the multiplication of the input signal by the carrier. [Stallings, p.123]
Applied Network Research Group Department of Computer Engineering, Kasetsart University
AM power saving
z
SSB (Single sideband)
z
DSBSC (Double sideband suppresses carrier)
z
VSB (Vestigial sideband)
It should be clear that s(t) contains unnecessary components, since each of the sideband contains the complete spectrum. SSB, DSBSC and VSB are methods to save power and bandwidth.
The disadvantage of suppressing the carrier is that the carrier can be used for synchroni zat ion purposed. A const ant carrier provides a cl ocking mechanism.A compromise approach is VSB, which uses one sideband and a reduced-power carrier. [Stallings, p.124]
Data Encoding - 28/28
Applied Network Research Group Department of Computer Engineering, Kasetsart University
FM and PM
Carrier
Modulating sine-wave signal
Amplitude-modulated (DSBTC) wave
Frequency-modulated wave Phase-modulated wave