Chapter 8

Pulse Amplitude Modulation

SUMMARY: In this chapter we discuss serial pulse amplitude modulation. This method
provides us with a new channel dimension every T seconds if we choose the pulse according
to the so-called Nyquist criterion. The Nyquist criterion implies that the required channel
bandwidth W is larger than 1/(2T ).

8.1 Harry Nyquist

Figure 8.1: Harry Nyquist.

Harry Nyquist (1889 - 1976) was born in Nilsby, Sweden. He received a Ph.D. in physics
at Yale University in 1917. As an engineer at Bell Labs., he did important work on thermal
noise, the stability of feedback amplifiers, telegraphy, facsimile, television, and other important
communications problems. In 1927 Nyquist determined that the number of independent pulses
that could be put through a telegraph channel per unit time is limited to twice the bandwidth of
the channel. Nyquist published his results in ”Certain topics in Telegraph Transmission Theory”
(1928). This rule is essentially a dual of what is now known as the Nyquist-Shannon sampling
theorem. In 1969 he was awarded the U.S. National Academy of Engineering’s fourth Founder’s
Medal ”in recognition of his many fundamental contributions to engineering.” (from Wikipedia).

113
114 CHAPTER 8. PULSE AMPLITUDE MODULATION

8.2 Problem Description

p(t)

@

@

@ t
0 T 2T
n w (t)
impulses
a0 , a1 , · · · , a K −1 sa (t) ? r (t)
- p(t) - + -


+1 +1 +1 @
@ s (t)
a
6 6 6
B @
@

B
t
 @@ t
B
0 T 2T 3T 0 T B 2T  3T 4T 5T

? @
@
−1

Figure 8.2: A serial pulse-amplitude modulation system.

We have observed before that a channel with a bandwidth of W Hz can accommodate roughly
2W T dimensions each T seconds. We considered building-block waveforms of finite duration,
therefore the bandwidth constraint could only be met approximately. Here we will investigate
whether it is possible to get a new dimension every 1/(2W ) seconds. We allow the building
blocks to have a non-finite duration. Moreover all these building blocks are time-shifts of a
”pulse”. The subject of this chapter is therefore called serial pulse transmission.
Consider figure 8.2. We there assume that the transmitter sends a signal s(t) that consists
of amplitude-modulated time shifts of a pulse p(t) by an integer multiple k of the so-called
modulation interval T , hence

∑
sa (t) = ak p(t − kT ). (8.1)
k=0,K −1

The vector of amplitudes a = (a0 , a1 , · · · , a K −1 ) consists of symbols ak , k = 0, K − 1 taking

values in the alphabet A. We call this modulation method serial pulse-amplitude modulation
(PAM). We are now interested in the property of a pulse that makes its shifts orthonormal.
8.3. ORTHONORMAL PULSES: THE NYQUIST CRITERION 115

8.3 Orthonormal Pulses: the Nyquist Criterion

8.3.1 The Nyquist result
If we have the possibility to choose the pulse p(t) ourselves we can choose it in such a way that
all time shifts of the pulse form an orthonormal base. In that case the pulse p(t) has to satisfy
∫ ∞ {
′ 1 if k = k ′ ,
p(t − kT ) p(t − k T )dt = (8.2)
−∞ 0 if k ̸ = k ′ ,

for integer k and k ′ . This is equivalent to

∫ ∞ {
1 if k = 0,
p(τ ) p(τ − kT )dτ = p(t) ∗ p(−t)|t=kT = h(kT ) = (8.3)
−∞ 0 if k ̸ = 0,

1
where h(t) = p(t) ∗ p(−t). This time-domain restriction on the pulse p(t) is called the zero-
forcing (ZF) criterion. Later we will see why.

THEOREM 8.1 The frequency-domain equivalent to (8.3) which is known as the Nyquist crite-
rion for zero intersymbol interference is

1 ∑
Z( f ) = H ( f + m/T )
T
m=··· ,−1,0,1,2,···
1 ∑
= ∥P( f + m/T )∥2 = 1 for all f ,
T
m=··· ,−1,0,1,2,···

where H ( f ) = P( f )P ∗ ( f ) = ∥P( f )∥2 is the Fourier transform of h(t) = p(t) ∗ p(−t). Note
that ∥P( f )∥ is the modulus of P( f ). Moreover Z ( f ) is called the 1/T -aliased spectrum of
H ( f ).
Later in this section we will give the proof of this theorem. First we will discuss it however
and consider an important consequence of it.

8.3.2 Discussion
• Since p(t) is a real signal, the real part of its spectrum P( f ) is even in f , and the imaginary
part of this spectrum is odd. Therefore the modulus ∥P( f )∥ of P( f ) is an even function
of the frequency f .

• If the bandwidth W of the pulse p(t) is strictly smaller than 1/(2T ) (see figure 8.3), then
the Nyquist criterion, which is based on H ( f ) = ∥P( f )∥2 , can not be satisfied. Thus
no pulse p(t) that satisfies a bandwidth-W constraint, can lead to orthogonal signaling if
W < 1/(2T ).
116 CHAPTER 8. PULSE AMPLITUDE MODULATION

H( f ) Z( f )

A
A
A
A
A
A

A
A
A
A
A
A

A AA
A
A
A
A
A


− 2T
1 1 f − 2T
3
− T1 − 2T
1 1 1 3 f
2T 2T T 2T

Figure 8.3: The spectrum H ( f ) = ∥P( f )∥2 corresponding to a pulse p(t) that does not satisfy
the Nyquist criterion.

H( f ) Z( f )
T 1

− 2T
1 1 f − 2T
3
− T1 − 2T
1 1 1 3 f
2T 2T T 2T

Figure 8.4: The ideally bandlimited spectrum H ( f ) = ∥P( f )∥2 . Note that P( f ) is a spectrum
with the smallest possible bandwidth satisfying the Nyquist criterion.

• The smallest possible bandwidth W of a pulse that satisfies the Nyquist criterion is 1/(2T ).
The ”basic” pulse with bandwidth 1/(2T ) for which the Nyquist criterion holds has a so-
called ideally bandlimited spectrum, which is given by
{ √
T if | f | < 1/(2T ),
P( f ) = (8.4)
0 if | f | > 1/(2T ),
√
and T /2 for | f | = 1/(2T ). The corresponding H ( f ) = ∥P( f )∥2 and 1/T -aliased
spectrum Z ( f ) can be found in figure 8.4.
The basic pulse p(t) that corresponds to the ideally bandlimited spectrum is shown in
figure 8.5. It is the well-known sinc-pulse

0.8

0.6

0.4

0.2

−0.2

−0.4
−5 −4 −3 −2 −1 0 1 2 3 4 5

Figure 8.5: The sinc-pulse that corresponds to T = 1, i.e. p(t) = sin(πt)/(π t).
8.3. ORTHONORMAL PULSES: THE NYQUIST CRITERION 117

H( f ) Z( f )
T 1
@ @ @ @ @ @ @
@ @ @ @ @ @ @
@ @ @ @ @ @ @
− 2T
1 1 f − 2T
3
− T1 − 2T
1 1 1 3 f
2T 2T T 2T

Figure 8.6: A spectrum H ( f ) with a positive excess bandwidth satisfying the Nyquist criterion.

1 sin(π t/T )
p(t) = √ . (8.5)
T π t/T

• A pulse with a bandwidth larger than 1/(2T ) can also satisfy the Nyquist criterion as can
be seen in figure 8.6. Note that the so-called excess bandwidth, i.e. the bandwidth minus
1/(2T ), can be larger than 1/(2T ). Square-root raised-cosine pulses p(t) have a spectrum
H ( f ) = |P( f )|2 that satisfies the Nyquist criterion and an excess bandwidth that can be
controlled.

At he end of this subsection we are ready to state the implication of the previous discussion.

RESULT 8.2 The smallest possible bandwidth W of a pulse that satisfies the Nyquist criterion
)
is W = 2T
1
. The sinc-pulse p(t) = √1 sin(πt/T
πt/T has this property. Note that this way of serial
T
pulse transmission leads to exactly 1
T = 2W dimensions per second.

8.3.3 Proof of the Nyquist result

We will next give the proof of result 8.1, the Nyquist result.
Proof: Since H ( f ) is the Fourier transform of h(t), the condition (8.3) can be rewritten as
∫ ∞ {
1 if k = 0,
h(kT ) = H ( f ) exp( j2π f kT )d f = (8.6)
−∞ 0 if k ̸= 0.

We now break up the integral in parts, a part for each integer m, and obtain
∞ ∫
∑ (2m+1)/2T
h(kT ) = H ( f ) exp( j2π f kT )d f
m=−∞ (2m−1)/2T
∞ ∫ 1/2T
∑ m
= H( f +) exp( j2π f kT )d f
m=−∞ −1/2T T
∫ [ ∞ ]
1/2T ∑ m
= H ( f + ) exp( j2π f kT )d f
−1/2T m=−∞
T
∫ 1/2T
= T Z ( f ) exp( j2π f kT )d f (8.7)
−1/2T
118 CHAPTER 8. PULSE AMPLITUDE MODULATION

where we have defined Z ( f ) by

∞
11 ∑ m
Z( f ) = H ( f + ). (8.8)
T m=−∞ T

Observe that Z ( f ) is a periodic function in f with period 1/T . Therefore it can be expanded in
terms of its Fourier series coefficients · · · , z −1 , z 0 , z 1 , z 2 , · · · as
∞
∑
Z( f ) = z k exp( j2π k f T ) (8.9)
k=−∞

where ∫ 1/2T
zk = T Z ( f ) exp(− j2π f kT )d f. (8.10)
−1/2T

If we now combine (8.7) and (8.10) we obtain that

h(−kT ) = z k , (8.11)

for all integers k. Condition (8.3) now tells us that only z 0 = 1 and all other z k are zero. This
implies that
Z ( f ) = 1, (8.12)
or equivalently
∞
1 ∑ m
H ( f + ) = 1. (8.13)
T m=−∞ T
2

8.3.4 Receiver implementation

If we use serial PAM with orthonormal pulses p(t − kT ), k = 0, K − 1, then we can use a
single matched-filter m(t) = p(T p − t) for the computations of the correlations of the received
waveform r (t) with all building-block waveforms, i.e. with all pulses. Assume that the delay T p
is chosen in such a way that m(t) is (effectively1 ) causal. The output of this filter m(t) when r (t)
is the input signal is
∫ ∞ ∫ ∞
u(t) = r (α)m(t − α)dα = r (α) p(T p − t + α)dα. (8.14)
−∞ −∞

At time t = T p + kT we see at the filter output

∫ ∞
rk = u(T p + kT ) = r (α) p(α − kT )dα, (8.15)
−∞
8.4. EXERCISES 119

r (t) H rk
-  HH -
p(T p − t) 


?
sample at t = T p + kT

Figure 8.7: The optimum receiver front-end for detection of serially transmitted orthonormal
pulses.

which is what the optimum receiver should determine, i.e. the correlation of r (t) with the pulse
p(t − kT ). This leads to the very simple receiver structure shown in figure 8.7. Processing the
samples rk , k = 1, K , should be done
∑ in the usual way.
When there is no noise r (t) = k=0,K −1 ak p(t − kT ). Then at time t = T p + kT we see at
the filter output
∫ ∞
rk = u(T p + kT ) = r (α) p(α − kT )dα
−∞
∫ ∞ ∑
= ak ′ p(α − k ′ T ) p(α − kT )dα
−∞ k ′ =0,K −1
∑ ∫ ∞
= a k′ p(α − k ′ T ) p(α − kT )dα = ak , (8.16)
k ′ =0,K −1 −∞

by the orthonormality of the pulses. Conclusion is that there is no intersymbol interference

present in the samples. In other words the Nyquist criterion forces the intersymbol interference
to be zero (zero-forcing (ZF)).

8.4 Exercises
1. Consider the pulse p(t) with spectrum P( f ) as in Fig. 8.8. We want to apply this pulse in
a pulse-amplitude-modulation system. We are now interested in finding out whether there
exists a T such that this pulse and all its shifted versions form an orthonormal base (set of
building-block waveforms), i.e. for integer k
∫ {
1 for k = 0,
p(t) p(t − kT )dt =
0 voor k ̸= 0.

(a) Check first that pulse p(t) has unit energy. Use Parseval.
(b) Find out (in a graphical manner) whether the pulse-spectrum satisfies the Nyquist
criterion for T = 0.02 sec. .
1 Note that p(t) has a non-finite duration in general.
120 CHAPTER 8. PULSE AMPLITUDE MODULATION

0.1 P( f )

-75 -25 25 75 f (Hz) →

Figure 8.8: The frequency-spectrum P( f ) of the pulse p(t).

(c) Assume that the pulse-spectrum satisfies the Nyquist criterion for T = α. Is this also
true then for T = 2α? Why?

2. Consider the two pulses

sin(πt)
p1 (t) =
πt
sin(2π t) − sin(π t)
p2 (t) = , (8.17)
πt
and determine the Fourier spectra P1 ( f ) and P2 ( f ) of these pulses first. Then show that the
set of all shifted versions of these two pulses {· · · , p1 (t +1), p2 (t +1), p1 (t), p2 (t), p1 (t −
1), p2 (t − 1), p1 (t − 2), p2 (t − 2), · · · } is a set of building blocks (energy 1, orthogonal).
Shifts are over multiples of 1 second.
With these two pulses, multi-pulse transmission is possible, hence
∑
s(t) = ak1 p1 (t − k) + ak2 p2 (t − k)
k=0,K −1

is a linear combination of shifted versions of two pulses.

3. The “square-root raised cosine pulse” p(t) has spectrum

{ √
T cos( π 2f T ) for − T1 ≤ f ≤ T1 ,
P( f ) = .
0 everywhere else.

(a) Check first whether p(t) has unit energy.

(b) Show that p(t) satisfies he Nyquist criterion for modulation-interval size T .
(c) Demonstrate that p(t) ”has a shorter tail” than the sinc-pulse corresponding to T .