Module – 2: Waveform Coding Techniques

A message signal can be originated from a digital or analog source. If the message signal is
derived from a digital source (e.g., digital computer), then from inception it is in the right
form for processing in digital communication system. If, however the message signal happens
to be analog in nature, as in a speech or video signal, then it has to be converted into digital
form before it can be transmitted by digital means. The block called formatter does this job.
The sampling process is the first process performed in analog to digital convertion. Two other
processes, Quantizing an encoding, are also involved in this conversion. Encoding schemes
like Pulse Code Modulation (PCM), Differential Pulse Code modulation (DPCM), Delta
Modulation (DM), Adaptive Delta Modulation (ADM) and so on are employed for this
purpose. This operation of encoding gives rise to digital signals. Though, almost all of these
coders carry the word “modulation” in their names, there is nothing in their bahaviour
resembling modulation. The are simply encoding techniques for analog sources. One should
conceptually be clear that even after passing through these so called “modulator” blocks the
digitised message sigal remains a base band signal, because it does not undergo any
frequency translation in any part of those blocks.

2.1 Pulse Code Modulation

In pulse-code modulation (PCM), a message signal is represented by a sequence of coded
pulses, which is accomplished by representing the signal in discrete form in both time and
amplitude. The essential operations in the transmitter of a PCM are sampling, Quantizing and
encoding as shown in Figure 2.1(a). The low-pass filter prior to sampling is included to
prevent aliasing of the message signal. The quantizing and encoding operations are usually
performed in the same circuit, which is called analog-to-digital converter. The basic
operations in the receiver are regeneration of impaired signals, decoding and reconstruction
of the train of quantized samples as shown in Figure 2.1(c). Regeneration also occurs at
intermediate points along the transmission path as necessary, as indicated in Figure 2.1(b).
The various operations of PCM are given as follows.

Figure 2.1: Basic elements of a PCM system (a) transmitter (b) transmission path (c) receiver
2.1.1 Encoding:
In combining the process of sampling and quantization, the specification of a continuous
message (baseband) signal becomes limited to a discrete set of values, but not in the form
best suiyed to transmission over a telephone line or rado path. To exploit the adavantages of
samling and quantization for the porpose of making the transmitted signal more robust to
noise, interference and othe channel impairements, we require the use of an encoding process
to translate the discrete set of sample values to a more appropriate form of signal. During
encoding, each of the descrete sample is represented by a codewore, where each element in
the codeword is known as code element or binary symbol (bit). The stream of such
codewords is called as a code.
In a binary code, each binary symbol may be either of two distinct values or kinds,
such as the presence or absence of a pulse. The two binary symbols of a binary code are
customarily denoted as 0 and 1. The binary represention is advantageous because a binary
symbol withstands a relatively high level of noise and is easy to regenerate. Suppose that, in a
binary code, each code word consists of n bits, where n denotes number of bits per sample.
Then using such a code, we may represent a total of 2n distinct number. For example, a
sample quantized into one of 256 levels may be represented by an 8-bit code word.
There are several ways of establishing a one-to-one correspondence between
representation levels and code words. A convenient method is to express the ordinal number
of the representation level as a binary number. In a binary number system, each digit has a
place-value that is a power of 2, as given in Table 2.1 for the case of four bits per sample.
Table2.1: Binary number system for n = 4 bits/sample

Ordinary Number of Level Number Expressed as Binary

Representation level sum of powers of 2 Number
0 0000
1 23 + 22 + 21 + 20 0001
2 23 + 22 + 21 + 20 0010
3 23 + 22 + 21 + 20 0011
4 23 + 22 + 21 + 20 0100
5 23 + 22 + 21 + 20 0101
6 23 + 22 + 21 + 20 0110
7 23 + 22 + 21 + 20 0111
8 23 + 22 + 21 + 20 1000
9 23 + 22 + 21 + 20 1001
10 23 + 22 + 21 + 20 1010
11 23 + 22 + 21 + 20 1011
12 23 + 22 + 21 + 20 1100
13 23 + 22 + 21 + 20 1101
14 23 + 22 + 21 + 20 1110
15 23 + 22 + 21 + 20 1111

There are several waveforms for repregentation of binary sequences produced by analog-to-
digital conversion. Figure 2.2 repregents two of such formats. In Figure 2.2(a), binary symbol
1 is represented by a pulse of constant duration for duration of one bit, and symbol 0 is
represented by switching off the pulse for the same duration. This format is known as non-
return to zero (NRZ) unipolar signal or on-off signal. In Figure 2.2(b), symbol 1 and 0 are
represented by pulses of positive and negetive amplitudes, respectively, with each bit
occupying one complete bit duration. The second format is known as NRZ polar signal.

Figure 2.2: Two binary waveforms (a) Nonreturn to zero unipolar (b) nonreturn to zero polar

2.1.2 Regeneration
The most important feature of PCM system lies in the ability to control the effect of distortion
and noise produced by transmitting a PCM signal through a channel. This capability is
accomplished by reconstructing the PCM signal by means of a chain of regenerative repeaters
located at sufficiently close spacing along the transmission route. As illustrated in Figure 2.3,
three basic functions are performed by a regenerative repeater: equalization, timing and
decision making. The equalizer shapes the received pulses so as to compensate for the effects
of amplitudes and phase distortions produced by non-ideal transmission characteristics of the
channel. The timing circuitry provides a periodic pulse train, derived from the received
pulses, for sampling the equalized pulses at the instants of time where the signal-to-noise is a
maximum. Each sample so extracted is compared to a predetermined threshold in the
decision-making device. In each bit interval, a decision is then made whether the received
symbol is a 1 or a 0 on the basis of whether the threshold is exceeded or not. If the threshold
is exceeded, a clean new pulse representing symbol 1 is transmitted to the next repeater.
Otherwise, another clean new pulse representing symbol 0 is transmitted. In this way, the
accumulation of distortion and noise in a repeater span is completely removed, provided that
the disturbance is not too large to cause an error in the decision-making process. Ideally,
except for delay, the regenerated signal is exactly the same as the signal originally
transmitted. In practice, however, the regenerated signal departs from the original signal for
two main reasons:
1. The unavoidable presence of channel noise and interference causes the repeater to
make wrong decision occasionally, thereby introducing bit errors into the regenerated
signal.
2. If the spacing between received pulses deviates from its assigned value, a jitter is
introduced into the regenerated pulse position, thereby causing distortion.
 Figure 2.3: Block diagram of regenerative repeater

2.1.3 Decoding
The first operation in the receiver is to regenerate (i.e., reshape and clean up) the received
pulses one last time. These clean pulses are then regrouped into code words and decoded (i.e.,
mapped back) into a quantized signal. The decoding process involves generating a pulse the
amplitude of which is the linear sum of all the pulses in the code word, with each pulse being
weighted by its place value (20, 21, 22, …, 2n–1) in the code, where n is number of bits per
sample.

2.1.4 Reconstruction filters:

The final operation in the receiver is to recover the message signal by passing the decoder
output through a low-pass reconstruction filter whose cut-off frequency is equal to the
message bandwidth W. Assuming that the transmission path is error free, the recovered signal
includes no noise with the exception of the distortion introduced by the quantization process.

2.1.5 Bandwidth of PCM

Minimum Bandwidth of PCM

Maximum Bandwidth of PCM:
Let us consider a PCM signal is transmitted using non return to zero unipolar wave as shown
in Figure 2.16(a) and each sample of duration Ts is represented by a code word of size n bits.
Hence the bit duration is:

Tb = Ts/n (2.1)

or, correspondingly, the bit rate is

 bits samples  bits

rb  nrs   (2.2)
 sample sec  sec

where rs is the sampling rate. We know Fourier transform of rectangular function is a sinc
function as shown in Figure 2.4 (the spectrum is depicted after passing through a LPF).
From Figure 2.4 it is known that PCM signal is a base band signal and the bandwidth
of PCM is:

1
( B ) max  Hz (2.3)
Tb

Figure 2.4: spectrum of a rectangular function of duration Tb

2.2 Differential Pulse Code Modulation
When a voice or video signal is sampled at a rate higher than the Nyquist rate as usually done
in PCM, the resulting sampled signal is found to exhibit a high degree of correlation between
adjacent samples. The meaning of this high correlation is that, in an average sense, the signal
does not change rapidly from one sample to the next, and as a result, the difference between
adjacent samples has a variance that is smaller than the variance of the signal itself. When
these highly correlated samples are encoded, as in the standard PCM system, the resulting
encoded signal contains redundant information. This means that symbols that are not
absolutely essential to the transmission of information are generated as a result of the
encoding process. By removing this redundancy before encoding, we obtain a more efficient
coded signal, which is the basic idea behind differential pulse code modulation.
Now if we know the past behaviour of a signal up to a certain point in time, we may
use prediction to make an estimate of a future value of the signal. Suppose then a baseband
signal m(t) is sampled at the rate fs = 1/Ts to produce the sequence {m[n]} whose samples are
Ts seconds apart. The fact that it is possible to predict future values of the signal m(t) provides
motivation for the differential quantization scheme shown in Figure 2.5 (a). In this scheme,
the input signal to the quantizer is defined by:

e [n ]  m [n ]  mˆ [n ]   (2.4)

which is the difference between the unquantized input sample m[n] and a prediction of it,
denoted by m̂ [n]. This predicted value is produced by using a linear prediction filter whose
input, as we will see, consists of a quantized version of the input samples m[n]. The
difference signal e[n] is the prediction error, since it is the amount by which the prediction
filter fails to predict the input exactly. By encoding the quantizer output, as in Figure 2.5 (a),
we obtain a variant of PCM known as differential pulse code modulation (DPCM).
The quantizer output may be expressed as

e q[ n ]  e [ n ]  q [ n ]   (2.5)

Where q[n] is the quantization error. According to Figure 2.5 (a), the quantizer output eq[n] is
ˆ [n ]  to produce the prediction-filter input
added to the predicted value m

mq [n ]  mˆ [n ]  eq [n ]   (2.6)

Substituting Eq. (2.5) into Eq. (2.6), we get

mq [n ]  mˆ [n ]  e [n ]  q [n ]   (2.7)

ˆ [n ]  e [n ]   is equal to the input

However, from Eq. (2.4) we observe that the sum term m
sample m[n]. Therefore, we may simplify Eq. (2.7) as

mq [n ]  m [n ]  q [n ]   (2.8)
which represents a quantized version of the input sample m[n]. That is, irrespective of the
properties of the prediction filter, the quantized sample mq[n] at the prediction filter input
differs from the original input sample m[n] by the quantization error q[n]. Accordingly. If the
prediction is good, the variance of the prediction error e[n] will be smaller than the variance
of m[n], so that a quantizer with given number of levels can be adjusted to produce a
quantization error with a smaller variance than would be possible if the input sample m[n]
were quantized directly as in a standard PCM system.

Figure 2.5: DPCM system (a) Transmitter (b) Receiver

The receiver for reconstructing the quantized version of the input is shown in Figure 2.5(b). It
consists of a decoder to reconstruct the quantized error signal. The quantized version of the
original input is reconstructed from the decoder output using the same prediction filter used
in the transmitter of Figure 2.5(a). In the absence of channel noise, we find that the encoded
signal at the receiver input is identical to the encoded signal at the transmitter output.
Accordingly, the corresponding receiver output is equal to mq[n], which differs from the
original input m[n] only by the quantization error q[n] incurred as a result of quantizing the
prediction error e[n]. From the foregoing analysis we observe that, in a noise-free
environment, the prediction filter in the transmitter and receiver operate on the same
sequence of samples, mq[n]. It is with this purpose in mind that a feedback path is added to
the quantizer in the transmitter as shown in Figure 2.5 (a).

2.3 Delta Modulation

In delta modulation (DM), an incoming signal is oversampled (i.e., at a rate much higher than
the Nyquist rate) to purposely increase the correlation between adjacent samples of the signal.
This is done to permit the use of a simple quantizing strategy for constructing the encoded
signal.
In its basic form, DM provides a staircase approximation to the oversampled version
of the message signal, as illustrated in Figure 2.5 (a). The difference between the input and
the approximation is quantised into only two levels, namely, ± ∆, corresponding to positive
and negative differences. Thus if the approximation falls below the signal at any sampling
epoch, it is increased by ∆. If on the other hand, the approximation lies above the signal, it is
diminished by ∆. Providing that the signal does not change too rapidly from the sample to
sample, we find that the staircase approximation remains within ± ∆ of the input signal.

Figure 2.5: Illustration of delta modulation

Let m(t) denote the input (message) signal, and mq(t) demote its staircase approximation. For
convenience of presentation, we adopt the following notation that is commonly used in the
digital signal processing literature:

m[n] = m(nTs), n = 0, ± 1, ± 2,… (2.9)

Where Ts is the sampling period and m(nTs) is a sample of the signal m(t) taken at time t =
nTs, and likewise for the samples of the other continuous time signals, We may then
formalize the basic principles of the delta modulation in the following set of discrete time
relations:

e[n] = m[n] – mq[n – 1] (2.10)

eq[n] = ∆ sgn(e[n]) (2.11)

mq[n] = mq[n – 1] + eq[n] (2.12)

Where e[n] is an error signal representing the difference between the present sample m[n] of
the input signal ad the latest approximation mq[n – 1] to it, eq[n] is the quantized version of
e[n], and sgn(.) is the signum function. Finally, the quantizer output mq[n] is coded to
produce the DM signal.
 Figure 2.6 (a) illustrate the way in which the staircase approximation mq(t) follows
variation in the input signa m(t) ain accordance with Eq. (2.10) – (2.12), and Figure 2.6 (b)
displays the corresponding binary sequence at the delta modulator output. It is apparent that
in a delta modulation system the rate of information transmission is simply equal to the
sampling rate fs = 1/Ts.
The pronciple virtue of DM is its simplicity. It may be generated by applying the
sampled version of the incomig message signal to a modulator that onvolves a comparator,
quantizer and accumulator interconnected as shown in Figure 2.7 (a). The block labeled z –1
iside the accumulator represents a unit delay, that is, a delay equal to one sampling period.
(The variable z is commonly used in the z – transform, which is basic to the analysis of
discrete time signals and syatems). Details of the modulator follows directly from Eq. (2.10)
– (2.12). The comparator computes the difference between its two inputs. The quantizer
consists of a hard limiter with an input – output relation that is a scaled version of the signum
function. The quantizer output is then applied to an accumulator, producig the result

n n
mq [n ]    sgn  e[i ]   eq [i ] (2.13)
i 1 i 1

Which is obtained by solving Eq. (2.11) and (2.12) for mq[n]. Thus, at the sampling instant
nTs, the accumulator increments the approximation by a step ∆ in a positive or negative
direction, depending on the algebraic sign of the error sample e[n]. If the input sample m[n] is
greater than the most recent approximation mq[n], appositive increment + ∆ is applied to the
approximation. If, on the other hand, the input sample is smaller, a negative increment – ∆ is
applied to the approximation. In this way, the accumulator does the best it can track the input
samples by one step (of amplitude + ∆ or – ∆) at a time. In the receiver show in Figure 2.6
(b), the staircase approximation mq(t) is reconstructed by passing the sequence of positive and
negative pulses, produced at the decoder output through an accumulator in a manner similar
to that used in the transmitter. The out of band quantization noise in the high frequency
staircase waveform mq(t) is rejected by passing it through a low-pass filter, as in Figure 2.7
(b), with a bandwidth equal to the original message bandwidth.

(a)
 (b)
Figure 2.7: DM system (a) Transmitter (b) Receiver

Delta modulation is subject to two types of quantization error: slope overload

distortion and granular noise.
Slope overload distortion: From Eq. (2.12), the digital equivalent of integration in the sense
that it represents the accumulation of positive and negative increments of magnitude ∆. Also,
denoting the quantization error by q[n], as shown by:

mq[n] = m[n] + q[n] (2.14)

we observe from Eq. (2.10) that the input to the quantizer is

e[n] = m[n] – m[n – 1] – q[n – 1] (2.15)

Thus except for the quantization error q[n – 1], the quantizer input is a first backward
difference of the input signal, which may be viewed as a digital approximation to the
derivative of the input signal or, equivalently, as the inverse of the digital integration process.
If we consider the maximum slope of the original input waveform m(t), it is clear that in order
for the sequence of samples {mq[n]} to increase as fast as input sequence of samples {m[n]}
in a region of maximum slope of m(t), we require that the condition

 dm(t )
 max   (2.16)
Ts dt

be satisfied. Otherwise, we find that the step-size ∆ is too small for the staircase
approximation mq(t) to follow a steep segment of the input waveform m(t), with the result that
mq(t) falls behind m(t), as illustrated in Figure 2.7. This condition is called slope overload,
and the resulting quantization error is called slope-overload distortion (noise). Note that since
the maximum slope of the staircase approximation mq(t) is fixed by the step size ∆, increases
and decreases in mq(t) tend to occur along straight lines. For this reason, a delta modulator
using a fixed step size is often referred to as a linear delta modulation
Granular noise: In contrast to slope-overload distortion, granular noise occurs when step size
∆ is too large relative to the local slope characteristics of the input waveform m(t), thereby
causing the staircase approximation mq(t) to hunt around a relatively flat segment of the input
waveform; this phenomenon is also illustrated in Figure 2.7. Granular noise is analogous to
quantization noise in a PCM system.
 We thus see that there is a need to have a large step-size to accommodate a wide
dynamic range, whereas a small step size is required for the accurate representation of
relatively low level signals. It is therefore clear that the choice of the optimum step size that
minimizes the mean square value of the quantization error in a linear delta modulator will be
the result of a compromise between slope-overload distortion and granular noise. To satisfy
such requirement, we need to make the delta modulator “adaptive” in the sense that the step
size is made to vary in accordance with the input signal.

Figure 2.7: Illustration of the two different forms of quantization error in delta modulation

2.4 Adaptive Delta Modulation

To overcome the quantization errors due to slope overload and granular noise, the step size
(δ) is made adaptive to variations in the input signal m(t). Particularly in the steep segment of
the signal m(t), the step size is increased. When the input is varying slowly, the step size is
reduced. Then the method is called Adaptive Delta Modulation (ADM). The ADM can take
continuous changes in step size or discrete changes in step size.
Figure 2.8 (a) shows the transmitter and 2.8 (b) shows receiver of ADM. The logic for
step size control is added in the diagram. The step size increases or decreases according to
certain rule depending on one bit quantizer output.
For example if one bit quantizer output is high (1), then step size may be doubled for
next sample. If on bit quantizer output is low, then step size may be reduced by one step.
Figure 2.9 shows the waveforms of adaptive delta modulator and sequence of bits
transmitted.
In the receiver of adaptive delta modulator shown in Figure 2.8 (b) the first part
generates the step size from each incoming bit. Exactly the same process is followed as that
in transmitter. The previous input and present input decided the step size. It is the given to an
accumulator which builds up staircase waveform. The low-pass filter then smoothens out the
staircase waveform to reconstruct the smooth signal.
 Logic for Step
Size Control

Sampled Message + e[n ] One Bit eq [n ] ADM

Signal m[n ]
 Quantizer
Encoder
 Wave
mq [n  1]
+

+

Delay (Ts )
mq [n ]

Accumulator

(a)

Input
 + Low Pass Output
Filter
+
Logic for Step
Delay (Ts )
Size Control

Accumulator

(b)

Figure 2.8: Adaptive delta modulations (a) Transmitter (b) Receiver