DIGITAL MODULATION METHODS IN AN

ADDITIVE WHITE GAUSSIAN NOISE CHANNEL

1
"There are two voices in you at all times. One
howls what the world expects of you, while
the other gently whispers all that you, in your
heart, truly want. Happiness comes to those
who have learned how to drown out the
white noise of the world and pay close
attention to what really matters.“
Heau Taplln
2
Noise pollution is a relative
thing. In a city, it's a jet plane
taking off. In a monastery, it's
a pen that scratches.
Robert Orber

3
CONTENTS:

 Introduction
 Geometric Representation of Signal
Waveforms
 EX 8.1.1 and EX 8.1.2

4
Introduction :

5
Introduction :

6
AWGN Channel :

AWGN:
o Additive
o White
o Gaussian
o Noise
7
AWGN Channel :
AWGN: Additive, White, Gaussian, Noise

Figure 8.18
Model for the
received signal
passed
through an
AWGN
channel.

8
AWGN Channel :
AWGN: Additive, White, Gaussian, Noise

9
AWGN Channel :
AWGN: Additive, White, Gaussian, Noise

10
AWGN Channel :
AWGN: Additive, White, Gaussian, Noise

11
AWGN Channel :
AWGN: Additive, White, Gaussian, Noise
clearvars; close all; clc;
LW = 'LineWidth'; FS = 'FontSize';
L = 1000; x = randn(L,1);
[xc,lags] = xcorr(x,20,'coeff');
stem(lags,xc,'filled',LW,2); grid;

clearvars; close all; clc;

LW = 'LineWidth'; FS = 'FontSize';
a = 0.95; N = 28;
n = 0:N-1;
lags = -(N-1):(N-1);
x = a.^n;
c = xcorr(x);
stem(lags,c ,'filled',LW,2); grid;

12
AWGN Channel :
AWGN: Additive, White, Gaussian, Noise

13
AWGN Channel :
 The AWGN channel is a good model for many satellite and
deep space communication links.
 It is not a good model for most terrestrial links because of
multipath, terrain blocking, interference, etc.
 However, for terrestrial path modeling, AWGN is commonly
used to simulate background noise of the channel under
study, in addition to multipath, terrain blocking, interference,
ground clutter and self interference that modern radio
systems encounter in terrestrial operation.
[https://en.wikipedia.org/wiki/Additive_white_Gaussian_noise]

14
Baseband Vs Passband Channels :

15
Baseband Vs Passband Channels :

16
Introduction :
o Throughout this chapter, we study only one-
shot communications, i.e., transmission of
only one signal corresponding to a single
message, with no transmission following
this single transmission.
o We study sequential transmission of
digital data and inter-symbol interference in
Chapter 10. 17
Introduction :

18
§8.1 GEOMETRIC REPRESENTATION
OF SIGNAL WAVEFORMS

19
Geometric Representation Of Signal Waveforms :

20
Geometric Representation Of Signal Waveforms :

21
Geometric Representation Of Signal Waveforms :

22
 GRAM-SCHMIDT
ORTHOGONAL-IZATION PROCEDURE

23
RECALL – Inner Product :

24
RECALL – Energy of a Signal :

25
Geometric Representation Of Signal Waveforms :

Dimensionality of a signal space : The maximum number of

independent signals in the space.

26
Geometric Representation Of Signal Waveforms :

27
Geometric Representation Of Signal Waveforms :

𝑐21 𝜓1 (𝑡) is the component of form 𝝍𝟏 (𝒕)

present in 𝑠2 (𝑡). By removing 𝑐21 𝜓1 (𝑡)
from 𝑠2 (𝑡) makes 𝑠2 (𝑡)  to 𝜓1 (𝑡).

28
Geometric Representation Of Signal Waveforms :
Inner product of 𝒅𝟐 (𝒕) and 𝝍𝟏 𝒕 can be computed as:
−∞ −∞

න 𝑑2 (𝑡) 𝜓1 𝑡 𝑑𝑡 = න 𝑠2 𝑡 − 𝑐21 𝜓1 (𝑡) 𝜓1 𝑡 𝑑𝑡

−∞ −∞
−∞ −∞

= න 𝑠2 𝑡 𝜓1 𝑡 𝑑𝑡 − 𝑐21 න 𝜓1 𝑡 𝜓1 𝑡 𝑑𝑡
−∞ −∞
= 𝑐21 − 𝑐21 = 0

where we made use of Eq. (8.1.2) and the fact that energy of 𝜓1 𝑡 is equal to 1.

 𝑑2 (𝑡) is orthogonal to 𝜓1 𝑡 .
29
Geometric Representation Of Signal Waveforms :

where 𝑑𝑘 (𝑡) is orthogonal to all 𝜓𝑖 𝑡 with

𝑖 = 1,2,3, ⋯ 𝑘 − 1.

30
Geometric Representation Of Signal Waveforms :

31
EX. 8.1.1

32
Geometric Representation Of Signal Waveforms :

Example 8.1.1
Let us apply the Gram-Schmidt procedure
to the set of four waveforms illustrated in
Figure 8.1(a).

33
Geometric Representation Of Signal Waveforms :
ℰ1 = 2

ℰ3 = 3
ℰ2 = 2

ℰ4 = 3
34
Geometric Representation Of Signal Waveforms :

35
Geometric Representation Of Signal Waveforms :
−∞

𝒄𝟐𝟏 = න 𝒔𝟐 𝒕 𝝍𝟏 𝒕 𝒅𝒕 = 𝟎
−∞

 𝒅𝟐 𝒕 = 𝒔𝟐 (𝒕)

36
Geometric Representation Of Signal Waveforms :

37
Geometric Representation Of Signal Waveforms :
Next we compute 𝑐31 (𝑡) :
−∞

𝒄𝟑𝟏 = න 𝒔𝟑 𝒕 𝝍𝟏 𝒕 𝒅𝒕 = 𝟎
−∞

38
Geometric Representation Of Signal Waveforms :
Next we compute 𝑐32 (𝑡) :
−∞

𝒄𝟑𝟐 = න 𝒔𝟑 𝒕 𝝍𝟐 𝒕 𝒅𝒕
−∞
𝟏 𝟏 𝟐
=− − =− =− 𝟐
𝟐 𝟐 𝟐

39
Geometric Representation Of Signal Waveforms :
First construct
sum signal, then
compute its
energy.

40
Geometric Representation Of Signal Waveforms :

First construct sum signal 𝑑3 (𝑡), then compute its energy.

41
Geometric Representation Of Signal Waveforms :

42
Geometric Representation Of Signal Waveforms :
Next we compute 𝑐41 (𝑡) :
−∞

𝒄𝟒𝟏 = න 𝒔𝟒 𝒕 𝝍𝟏 𝒕 𝒅𝒕 = 𝟐
−∞

43
Geometric Representation Of Signal Waveforms :
Next we compute 𝑐42 (𝑡) :
−∞

𝒄𝟒𝟐 = න 𝒔𝟒 𝒕 𝝍𝟐 𝒕 𝒅𝒕 = 𝟎
−∞

44
Geometric Representation Of Signal Waveforms :
Next we compute 𝑐43 (𝑡) :
−∞

𝒄𝟒𝟑 = න 𝒔𝟒 𝒕 𝝍𝟑 𝒕 𝒅𝒕 = 𝟏
−∞

45
Geometric Representation Of Signal Waveforms :

 𝒔𝟒 𝒕 = 𝟐 𝝍𝟏 𝒕
+ 𝝍𝟑 𝒕

46
Geometric Representation Of Signal Waveforms :

47
Geometric Representation Of Signal Waveforms :

48
Geometric Representation Of Signal Waveforms :

The 𝑁 orthonormal waveforms

{𝝍𝒏 𝒕 } form an orthonormal basis
in the 𝑁-dimensional signal space.
49
Geometric Representation Of Signal Waveforms :

50
Geometric Representation Of Signal Waveforms :

NB: Order of integration and summation is interchangeable.

51
Geometric Representation Of Signal Waveforms :

52
Geometric Representation Of Signal Waveforms :

53
Geometric Representation Of Signal Waveforms :
• It should also be noted that since selection of the
orthonormal basis is NOT UNIQUE, the vector
representation of signal waveforms is not unique
either.
• However, the dimensionality of the signal
space N and the length of the vectors
representing the signals are independent of the
selected orthonormal basis.
54
Geometric Representation Of Signal Waveforms :

See also:
Appendix C
“GRAM-SCHMIDT ORTHOGONALIZATION OF A VECTOR SET ”
of [3].

55
EX. 8.1.2

56
Geometric Representation Of Signal Waveforms :

57
Geometric Representation Of Signal Waveforms :

58
Geometric Representation Of Signal Waveforms :

59
Geometric Representation Of Signal Waveforms :
−∞

𝑠𝑚𝑛 = න 𝑠𝑚 (𝑡) 𝜓𝑛 𝑡 𝑑𝑡 ⋯ ⋯ ⋯ (8.1.11)

−∞

Weighting Coeficients:
−∞

𝑠11 = න 𝑠1 (𝑡) 𝜓1 𝑡 𝑑𝑡 = 2
−∞
−∞

𝑠12 = න 𝑠1 (𝑡) 𝜓2 𝑡 𝑑𝑡 = 0
−∞
−∞

𝑠13 = න 𝑠1 (𝑡) 𝜓3 𝑡 𝑑𝑡 = 0
−∞ 60
Geometric Representation Of Signal Waveforms :
−∞

𝑠𝑚𝑛 = න 𝑠𝑚 (𝑡) 𝜓𝑛 𝑡 𝑑𝑡 ⋯ ⋯ ⋯ (8.1.11)

−∞
−∞

𝑠21 = න 𝑠2 (𝑡) 𝜓1 𝑡 𝑑𝑡 = 0
−∞
−∞

𝑠22 = න 𝑠2 (𝑡) 𝜓2 𝑡 𝑑𝑡 = 2
−∞
−∞

𝑠23 = න 𝑠2 (𝑡) 𝜓3 𝑡 𝑑𝑡 = 0
−∞
61
Geometric Representation Of Signal Waveforms :
−∞

𝑠𝑚𝑛 = න 𝑠𝑚 (𝑡) 𝜓𝑛 𝑡 𝑑𝑡 ⋯ ⋯ ⋯ (8.1.11)

−∞
−∞

𝑠31 = න 𝑠3 (𝑡) 𝜓1 𝑡 𝑑𝑡 = 0
−∞
−∞

𝑠32 = න 𝑠3 (𝑡) 𝜓2 𝑡 𝑑𝑡 = − 2
−∞
−∞

𝑠33 = න 𝑠3 (𝑡) 𝜓3 𝑡 𝑑𝑡 = 1
−∞
62
Geometric Representation Of Signal Waveforms :
−∞

𝑠𝑚𝑛 = න 𝑠𝑚 (𝑡) 𝜓𝑛 𝑡 𝑑𝑡 ⋯ ⋯ ⋯ (8.1.11)

−∞
−∞

𝑠41 = න 𝑠4 (𝑡) 𝜓1 𝑡 𝑑𝑡 = 2
−∞
−∞

𝑠42 = න 𝑠4 (𝑡) 𝜓2 𝑡 𝑑𝑡 = 0
−∞
−∞

𝑠43 = න 𝑠4 (𝑡) 𝜓3 𝑡 𝑑𝑡 = 1
−∞
63
Geometric Representation Of Signal Waveforms :

64
Geometric Representation Of Signal Waveforms :
ℰ1 = 2 ℰ3 = 3

−∞

𝑠1 𝑡 , 𝑠3 𝑡 = න 𝑠1 𝑡 𝑠3 𝑡 𝑑𝑡 = 0
−∞
𝒔𝟏 . 𝒔𝟑 = 2, 0,0 . 0, − 2, 1
=0+0+0=0
65
Geometric Representation Of Signal Waveforms :

66
Geometric Representation Of Signal Waveforms :

67
Geometric Representation Of Signal Waveforms :

68
PR. 8.4

69
Geometric Representation Of Signal Waveforms :
Determine a set
of orthonormal
functions for the
four signals
shown in Figure
P-8.4. You may
determine these
functions by
inspection.

70
Geometric Representation Of Signal Waveforms :

71
Geometric Representation Of Signal Waveforms :

72
 QUIZ-1
NEXT WEEK
73
DISCLAIMER
74
These power point slides are NOT
SUBSTITUTE of TEXT BOOK(s).
You’re ALWAYS DIRECTED to CAREFULLY
READ the relevant book chapter and
SOLVE ALL examples and end problems.

75
REFERENCES :
[1] FCS 2nd Ed. [Proakis-2014]

[2] [Hsu-2003] A&D Communications

[3] [Lathi-2009] MDnACS

76