Chapter 5: Signal Space Analysis

Chapter 5: Signal Space Analysis
CHAPTER 5
SIGNAL SPACE ANALYSIS
Digital Communication Systems 2012

1/45
R.Sokullu
Outline
5.1 Introduction
5.2 Geometric Representation of Signals
Gram-Schmidt Orthogonalization Procedure
5.3 Conversion of the AWGN into a Vector Channel
5.4 Maximum Likelihood Decoding
5.5 Correlation Receiver
5.6 Probability of Error
Digital Communication Systems 2012 R.Sokullu 2/45

Introduction the Model

We consider the following model of a generic
transmission system (digital source):
A message source transmits 1 symbol every T sec
Symbols belong to an alphabet M (m1, m2, mM)
Binary symbols are 0s and 1s
Quaternary PCM symbols are 00, 01, 10, 11

Transmitter Side
Symbol generation (message) is probabilistic, with
a priori probabilities p1, p2, .. pM. or
Symbols are equally likely
So, probability that symbol mi will be emitted:
i P (mi )
1
= for i=1,2,....,M (5.1)
M

Transmitter takes the symbol (data) mi (digital

message source output) and encodes it into a
distinct signal si(t).
The signal si(t) occupies the whole slot T allotted
to symbol mi.
si(t) is a real valued energy signal (???)
T
E i si2 (t )dt , i=1,2,....,M (5.2)
0

Transmitter takes the symbol (data) mi (digital

message source output) and encodes it into a
distinct signal si(t).
The signal si(t) occupies the whole slot T allotted
to symbol mi.
si(t) is a real valued energy signal (signal with
finite energy)
T
E i si2 (t )dt , i=1,2,....,M (5.2)
0

Channel Assumptions:
Linear, wide enough to accommodate the signal si(t)
with no or negligible distortion
Channel noise is w(t) is a zero-mean white Gaussian
noise process AWGN
additive noise
received signal may be expressed as:
0tT
x(t ) si (t ) w(t ), (5.3)
i=1,2,....,M

Receiver Side
Observes the received signal x(t) for a duration of time T sec
Makes an estimate of the transmitted signal si(t) (eq. symbol mi).
Process is statistical
presence of noise
errors
So, receiver has to be designed for minimizing the average
probability of error (Pe)
What is this?
M
Pe = p P(m m
i 1
i i / mi ) (5.4)
Symbol sent
cond. error probability
given ith symbol was
sent Digital Communication Systems 2012 R.Sokullu 8/45
Outline
5.1 Introduction

5.2. Geometric Representation of

Signals
Objective: To represent any set of M energy signals
{si(t)} as linear combinations of N orthogonal
basis functions, where N M
Real value energy signals s1(t), s2(t),..sM(t), each of
duration T sec Orthogonal basis
function
N
0tT
si (t ) sij j (t ), (5.5)
j 1 i==1,2,....,M
coefficient
Energy signal
Coefficients:
T i=1,2,....,M
sij si (t ) j (t )dt , (5.6)
0
j=1,2,....,M
Real-valued basis functions:
T
1 if i j
0 i (t ) j (t )dt ij 0 if i j (5.7)

The set of coefficients can be viewed as a N-

dimensional vector, denoted by si
Bears a one-to-one relationship with the
transmitted signal si(t)

Figure 5.3
(a) Synthesizer for generating the signal si(t). (b) Analyzer
for generating the set of signal vectors si .

So,
Each signal in the set si(t) is completely
determined by the vector of its coefficients
si1
s
i 2
.
si , i 1,2,....,M (5.8)
.
.

siN

Finally,
The signal vector si concept can be extended to 2D, 3D etc. N-
dimensional Euclidian space
Provides mathematical basis for the geometric representation of
energy signals that is used in noise analysis
Allows definition of
Length of vectors (absolute value)
Angles between vectors
Squared value (inner product of s i with itself)
2 Matrix
si siT s i Transposition
N
= sij2 , i 1,2,....,M (5.9)
j 1

Figure 5.4
Illustrating the geometric
representation of signals
for the case when N 2
and M 3.
(two dimensional space,
three signals)

Also,
What is the relation between the vector representation
of a signal and its energy value?
start with the definition T
of average energy in a E i si2 (t )dt (5.10)

0
signal(5.10)
N
Where si(t) is as in (5.5): si (t ) sij j (t ), (5.5)
j 1

T
N N

After substitution: E i sij j (t ) s (t )
ik k dt
0 j 1 k 1
N N T
After regrouping: Ei s s (t ) (t )dt

ij ik j k (5.11)
j 1 k 1 0
j(t) is orthogonal, so N
Ei s 2 2
= si (5.12)
finally we have: j 1
ij
The energy of a signal

is equal to the squared
length of its vector
Formulas for two signals

Assume we have a pair of signals: si(t) and sj(t), each
represented by its vector,
Then:
T
sij si (t ) sk (t )dt s s T
i k (5.13)
0
Inner product is invariant

to the selection of basis
Inner product of the signals functions
is equal to the inner product
of their vector
representations [0,T]
Euclidian Distance
The Euclidean distance between two points
represented by vectors (signal vectors) is equal to
||si-sk|| and the squared value is given by:
N
= (sij -skj ) 2
2
si s k (5.14)
j 1
T
= ( si (t ) sk (t )) 2 dt
0

Angle between two signals

The cosine of the angle ik between two signal vectors si and
sk is equal to the inner product of these two vectors, divided
by the product of their norms:
T
s s
cosik i k
(5.15)
si sk
So the two signal vectors are orthogonal if their inner product

siTsk is zero (cos ik = 0)

Schwartz Inequality
Defined as:

2
s1 (t )s2 (t )dt 2
s (t )dt s22 (t )dt (5.16)
1
accept without proof

Outline
5.1 Introduction

Gram-Schmidt Orthogonalization
Procedure
Assume a set of M energy signals denoted by s1(t), s2(t), .. , sM(t).
1. Define the first basis function s1 (t )

starting with s1 as: (where E is the 1 (t ) (5.19)
energy of the signal) (based on E1
5.12)
2. Then express s1(t) using the basis
function and an energy related s1 (t ) E11 (t ) = s111 (t ) (5.20)
coefficient s11 as:
3. Later using s2 define the T

coefficient s21 as: s21 s2 (t )1 (t )dt (5.21)
0

4. If we introduce the g 2 (t ) s2 (t ) s211 (t ) (5.22)

intermediate function g2 as:
Orthogonal to 1(t)
g 2 (t )
5. We can define the second 2 (t ) (5.23)
T
basis function 2(t) as:
0
g 22 (t )dt
6. Which after substitution of s2 (t ) s211 (t )

2 (t ) (5.24)
g2(t) using s1(t) and s2(t) it E2 s212
becomes:
T
Note that 1(t) and 2(t) are 0
22 (t )dt 1 (Look at 5.23)
orthogonal that means:
T
0
1 (t )2 (t )dt 0
And so on for N dimensional space,

In general a basis function can be defined using the
following formula:
i 1
g i (t ) si (t ) sij - j (t) (5.25)
j 1
where the coefficients can be defined using:

T
sij si (t ) j (t )dt , j 1, 2,....., i 1 (5.26)
0

Special case:
For the special case of i = 1 gi(t) reduces to si(t).
General case:
Given a function gi(t) we can define a set of basis

functions, which form an orthogonal set, as:
gi (t )
i (t ) , i 1, 2,....., N (5.27)
T

0
gi2 (t )dt

Outline
5.1 Introduction

Conversion of the Continuous AWGN

Channel into a Vector Channel
Suppose that the si(t) is
x(t ) si (t ) w(t ),
not any signal, but
specifically the signal at 0tT
(5.28)
the receiver side, defined i=1,2,....,M
in accordance with an
AWGN channel: T
x i x(t ) j (t ) dt
So the output of the 0
correlator (Fig. 5.3b) can =sij wi ,

be defined as: j 1, 2,....., N (5.29)
T T
sij si (t )i (t )dt (5.30) wi w(t )i (t )dt (5.31)

0 0
deterministic quantity random quantity
contributed by the sample value of the

transmitted signal si(t) variable Wi due to noise

Now,
Consider a random
N
process X1(t), with x1(t), a
x(t ) x(t ) x ji (t ) (5.32)
sample function which is j 1
related to the received N
signal x(t) as follows: x(t ) x(t ) ( sij w j ) j (t )
Using 5.28, 5.29 and 5.30 j 1
and the expansion 5.5 we N
get: =w(t ) w j j (t )
j 1
=w(t ) (5.33)
which means that the sample function x1(t) depends only on
the channel noise!
The received signal can

be expressed as:
N
x(t ) x ji (t ) x(t )
j 1
N
x ji (t ) w(t ) (5.34)
j 1
NOTE: This is an expansion similar to the one

in 5.5 but it is random, due to the additive
noise.

Statistical Characterization
The received signal (output of the correlator of
Fig.5.3b) is a random signal. To describe it we need
to use statistical methods mean and variance.
The assumptions are:
X(t) denotes a random process, a sample function of which
is represented by the received signal x(t).
Xj(t) denotes a random variable whose sample value is
represented by the correlator output xj(t), j = 1, 2, N.
We have assumed AWGN, so the noise is Gaussian, so X(t)
is a Gaussian process and being a Gaussian RV, X j is
described fully by its mean value and variance.

Mean Value
Let Wj, denote a random variable, represented by its
sample value wj, produced by the jth correlator in
response to the Gaussian noise component w(t).
So it has zero mean (by definition of the AWGN
model)
x j E X j
then the mean of
=E sij W j
Xj depends only on
=sij E[W j ]
sij:
x j = sij (5.35)

Variance
x2i var[ X j ]
Starting from the definition,
we substitute using 5.29 and =E ( X j sij ) 2
5.31
T =E W j2 (5.36)
wi w(t )i (t )dt (5.31)
0
T T

=E W (t ) j (t )dt W (u ) j (u )du
2
xi
0 0
T T
T T =E j (t )i (u )W (t )W (u )dtdu (5.37)
x2i = o 0
o
i (t ) j (u) E[W (t )W (u)]dtdu
0
T T

=E

0 j (t )i (u) Rw (t , u )dtdu (5.38) Autocorrelation function of
o the noise process
It can be expressed as:
(because the noise is N0
stationary and with a R w (t , u ) (t u ) (5.39)
2
constant power spectral
density) N T T
After substitution for
x2i = 0
2 (t ) (u) (t u)dtdu
o 0
i j
the variance we get: N0 T 2

=
2 0 j (t )dt (5.40)
And since j(t) has unit N0
energy for the variance x = 2
2
i
for all j (5.41)
we finally have:
Correlator outputs, denoted by Xj have variance
equal to the power spectral density N0/2 of the noise
process W(t). Digital Communication Systems 2012 R.Sokullu 36/45
Properties (without proof)

Xj are mutually uncorrelated
Xj are statistically independent (follows from above
because Xj are Gaussian)
and for a memoryless channel the following equation
is true:
N
f x ( x / mi ) f x j ( x j / mi ), i=1,2,....,M (5.44)
j 1

Define (construct) a vector X of N random variables, X1, X2,

XN, whose elements are independent Gaussian RV with
mean values sij, (output of the correlator, deterministic part of
the signal defined by the signal transmitted) and variance
equal to N0/2 (output of the correlator, random part,
calculated noise added by the channel).
then the X1, X2, XN , elements of X are statistically
independent.
So, we can express the conditional probability of X, given si(t)
(correspondingly symbol mi) as a product of the conditional
density functions (fx) of its individual elements fxj.
NOTE: This is equal to finding an expression of the probability
of a received symbol given a specific symbol was sent,
assuming a memoryless channel

that is:
N
f x ( x / mi ) f x j ( x j / mi ), i=1,2,....,M (5.44)
j 1
where, the vector x and the scalar xj, are sample

values of the random vector X and the random
variable Xj.

N
f x ( x / mi ) f x j ( x j / mi ), i=1,2,....,M (5.44)
j 1
Vector x is called
observation vector
Vector x and scalar xj Scalar xj is called
are sample values of observable element
the random vector X
and the random
variable Xj

Since, each Xj is Gaussian with mean sj and variance

N0/2

1 2 j=1,2,....,N
f x j ( x / mi ) ( N 0 ) N /2
exp ( x j sij ) , (5.45)
N0 i=1,2,....,M
we can substitute in 5.44 to get 5.46:
1 N 2
f x ( x / mi ) ( N 0 ) exp ( x j sij ) ,
N /2
N 0 j 1 i=1,2,....,M (5.46)

If we go back to the formulation of the received

signal through a AWGN channel 5.34
N
x(t ) x ji (t ) x(t )
j 1
N
x ji (t ) w(t ) (5.34)
j 1
Only projections of the noise onto

the basis functions of the signal set
The vector that we
{si(t)Mi=1 affect the significant
have constructed fully statistics of the detection problem
defines this part

Finally,
The AWGN channel, is equivalent to an N-
dimensional vector channel, described by the
observation vector
x si w, i 1, 2,....., M (5.48)

Outline
5.1 Introduction

Maximum Likelihood Decoding

to be continued.

Chapter 5: Signal Space Analysis

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 5: Signal Space Analysis

Uploaded by

Copyright:

Available Formats

Chapter 5: Signal Space Analysis

SIGNAL SPACE ANALYSIS

Digital Communication Systems 2012

Digital Communication Systems 2012 R.Sokullu 2/45

Introduction the Model

Digital Communication Systems 2012 R.Sokullu 3/45

Digital Communication Systems 2012 R.Sokullu 4/45

Transmitter takes the symbol (data) mi (digital

Digital Communication Systems 2012 R.Sokullu 5/45

Transmitter takes the symbol (data) mi (digital

Digital Communication Systems 2012 R.Sokullu 6/45

Digital Communication Systems 2012 R.Sokullu 7/45

Digital Communication Systems 2012 R.Sokullu 9/45

5.2. Geometric Representation of

Digital Communication Systems 2012 R.Sokullu 11/45

The set of coefficients can be viewed as a N-

Digital Communication Systems 2012 R.Sokullu 12/45

Digital Communication Systems 2012 R.Sokullu 13/45

Digital Communication Systems 2012 R.Sokullu 14/45

Digital Communication Systems 2012 R.Sokullu 15/45

Digital Communication Systems 2012 R.Sokullu 16/45

start with the definition T

of average energy in a E i si2 (t )dt (5.10)

Digital Communication Systems 2012 R.Sokullu 17/45

After regrouping: Ei s s (t ) (t )dt

The energy of a signal

Formulas for two signals

Inner product is invariant

Digital Communication Systems 2012 R.Sokullu 20/45

Angle between two signals

So the two signal vectors are orthogonal if their inner product

Digital Communication Systems 2012 R.Sokullu 21/45

accept without proof

Digital Communication Systems 2012 R.Sokullu 22/45

Digital Communication Systems 2012 R.Sokullu 23/45

1. Define the first basis function s1 (t )

3. Later using s2 define the T

Digital Communication Systems 2012 R.Sokullu 24/45

4. If we introduce the g 2 (t ) s2 (t ) s211 (t ) (5.22)

6. Which after substitution of s2 (t ) s211 (t )

And so on for N dimensional space,

where the coefficients can be defined using:

Digital Communication Systems 2012 R.Sokullu 26/45

Given a function gi(t) we can define a set of basis

Digital Communication Systems 2012 R.Sokullu 27/45

Digital Communication Systems 2012 R.Sokullu 28/45

Conversion of the Continuous AWGN

correlator (Fig. 5.3b) can =sij wi ,

sij si (t )i (t )dt (5.30) wi w(t )i (t )dt (5.31)

deterministic quantity random quantity

contributed by the sample value of the

Digital Communication Systems 2012 R.Sokullu 30/45

and the expansion 5.5 we N

The received signal can

NOTE: This is an expansion similar to the one

Digital Communication Systems 2012 R.Sokullu 32/45

Digital Communication Systems 2012 R.Sokullu 33/45

Digital Communication Systems 2012 R.Sokullu 34/45

the variance we get: N0 T 2

Properties (without proof)

Digital Communication Systems 2012 R.Sokullu 37/45

Define (construct) a vector X of N random variables, X1, X2,