Transcript ICN lecture6_ Digital-Digital & Analog

Digital-to-digital Conversion
A digital signal is a sequence of discrete discontinuous
voltage pulses. Each pulse is a signal element (symbol).
Binary data are transmitted by encoding each data bit
into signal elements.
Basic Ideas of conversion:
BINARY DATA
to
DIGITAL SIGNAL.
Line coding and decoding
There is a variety of language that can be used to
describe the process of transforming binary data into
a digital signal. Line coding is another term used to
describe digital encoding
Digital Transmission
Data element – The actual information/data/message to be delivered
Signal element – The right format of signals or suitable signal types
used to carry the data element (actual data)
Ratio of Data/Signal
r=d/s
d = data element
s = signal element
The higher the r the better the transmission efficiency
Data element (Actual information/message)
Signal element (how the information is ‘carried’)
Both elements can be
specified in terms of:
Time interval.
Both measured in secs
Rate.
Data rate: Bits per
second (bps).
Signal rate: Symbols
per second or Baud.
The relationship between
the date rate and the
signal rate will indicate
how efficiently the
bandwidth is being used.
Requirements for digital signaling
Adequate noise immunity. (The receiver can determine a level above the
noise. This will either be above a threshold, in relation to previous symbols
or relative to 0V.)
Synchronization. (The receiver knows when to sample the signal.)
Security. (Only the receiver is able to decode the signal)
Effect of lack of synchronization
Adequate Noise immunity
This effectively means maintaining an adequate
signal to noise ratio. As the ability to control noise
is limited the mechanisms focus on signal
viability. Two broad approaches are considered:

Minimize attenuation mechanisms. The
significant factor here is to remove any DC bias
because coupling will severely attenuate this
component of the signal.

Use transitions and changes as mechanisms to
encode the data.
Line coding schemes
UNIPOLAR
NRZ scheme (Non Return to Zero)
Main problem of NRZ:



No timing information is carried to provide correct synchronisation at
the receiver. (Not knowing when one bit ended and next bit starts)
DC Component – when voltage level in a digital signal is constant for a
while. The spectrum creates a very low frequencies which cannot
allow to be passed by certain devices/systems (e.g. transfomer/
telephone)
Baseline Wandering: Running average of received signal power.
(BI)POLAR
RZ scheme: (Return to Zero) at the middle of the bit
No DC component problem
Main problem of RZ:
 Requires two signal
changes to encode a
data bit (low r), which
needs extra bandwidth
 3 signal levels results in
more complexity
 Baseline Wandering can
still exists.
POLAR Bi-phase: Manchester and differential Manchester



No DC component
problem
No Baseline
Wandering
Timing information
is provided
Duration of signal bit divided into two halves. (two signal levels)
The transition at the middle of the bit is used for synchronization.
Only drawback: r = 0.5 (or 1/2): The minimum bandwidth of Manchester
and differential Manchester is 2 times that of NRZ (1/2bit/baud).
BIPOLAR schemes: AMI* and pseudoternary
In bipolar encoding, we use three levels: positive, zero, and
negative.
 No DC component
problem
 No Baseline
Wandering
 Signal rate ‘r’ can be
improved
 Need to provide
Timing information
via:


AMI - Alternate Mark Inversion
Start/stop bits
Control bits prior to
actual data
Analog to Digital Conversion/Encoding
WCB/McGraw-Hill
 The McGraw-Hill Companies, Inc., 1998
Analog-to-Digital Conversion
A-D conversion requires Sampling method; where a series of pulse-trains
is applied to the analog signal. This process is called Pulse Amplitude
Modulation (PAM).
Hence, it is require to apply sufficient sample of pulse-trains or sampling
frequency to the analog signal:
fs = 2 X Signal_Bandwidth
(according to Nyquist Sampling Theorem).
After suitable sampling, each pulse will be represented a binary number:
the process of converting this pulse-level to binary level in called
Quantisation. Number of M bits used for Quantisation is a crucial factor.
The final resultant is the digital representation of the analog signal in
a binary coding representation via the line-coding or binary encoding.
The entire A-D process is called Pulse Code Modulation (PCM)
Nyquist’s Sampling Theorem
To convert an analogue signal to a digital signal, the sampling
frequency involved (frequency of pulse-trains) needs to be:
Sampling
frequency
fs = 2 × W
Bandwidth
of signal
A band-limited signal of finite energy, which has no frequency
components higher than W Hertz, may be completely described
by specifying the values of the signal at instants of time separated
by (1/2W) seconds or can be recovered from a knowledge of its
samples taken at a rate of 2W samples per second.
Nyquist’s Sampling Theorem
Message signal
Frequency Content
Frequency Content of the
sampled message signal
Message bandwidth
fs = 2 × W
Sampling frequency
Nyquist’s Sampling Theorem
Sampling frequency = message bandwidth
Message signal cannot be recovered from the sampled signal !!
Components of PCM encoder
Three different sampling methods for PCM
Pulse Amplitude Modulation (PAM) Sampling
Pulse Amplitude Modulation
(PAM) - Sampling
Duty cycle =
T / Ts
Quantized PAM Signal
WCB/McGraw-Hill
 The McGraw-Hill Companies, Inc., 1998
Quantizing Using
Sign and Magnitude
WCB/McGraw-Hill
 The McGraw-Hill Companies, Inc., 1998
Quantized PAM Signal
PCM
 The McGraw-Hill Companies, Inc., 1998
From Analog to PCM
 The McGraw-Hill Companies, Inc., 1998
From Analog to PCM
WCB/McGraw-Hill
 The McGraw-Hill Companies, Inc., 1998
From Analog to PCM
WCB/McGraw-Hill
 The McGraw-Hill Companies, Inc., 1998
From Analog to PCM
WCB/McGraw-Hill
 The McGraw-Hill Companies, Inc., 1998
Extra Notes on
Line coding:
MULTILEVEL. In mBnL schemes, a pattern of m data
elements is encoded as a pattern of n signal elements.
Multilevel: 2B1Q scheme
(2bits/baud)
Multilevel: 8B6T scheme (8/6 bits/baud).
Converts bytes of data into 6 digits of ternary code (36 =
729 possible states)
Multitransition: MLT-3 scheme
Summary of line coding schemes