Scribd
Upload
122 views

Digital Communication Systems in Bandlimited Channel: MET 1413 Faculty of Electrical Enginerring 81310 UTM Skudai

1) The document discusses digital communication systems over bandlimited channels and optimal filtering techniques. It aims to maximize the signal-to-noise ratio and minimize the probability of error. 2) The matched filter is introduced, which maximizes the signal amplitude to noise ratio at the detection time and is the optimal filter for white noise channels. 3) It is shown that the matched filter transfer function is the time-reversed version of the signal waveform, which results in the maximum correlation between the signal and filter output at the detection time.

Uploaded by

elhenshire_h
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
122 views

Digital Communication Systems in Bandlimited Channel: MET 1413 Faculty of Electrical Enginerring 81310 UTM Skudai

1) The document discusses digital communication systems over bandlimited channels and optimal filtering techniques. It aims to maximize the signal-to-noise ratio and minimize the probability of error. 2) The matched filter is introduced, which maximizes the signal amplitude to noise ratio at the detection time and is the optimal filter for white noise channels. 3) It is shown that the matched filter transfer function is the time-reversed version of the signal waveform, which results in the maximum correlation between the signal and filter output at the detection time.

Uploaded by

elhenshire_h
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

1

Digital Communication Systems in


Bandlimited channel II
MET 1413
Faculty of Electrical Enginerring
81310 UTM Skudai
Optimum Performance
To minimize P
E
, must maximize since Q
decreases monotonically as increases
Goal: filter signal to enhance signal and
reduce noise power Matched Filter
Provide maximum signal-to-noise power ratio
at its output for a given transmitted symbol
waveform
( )

Q
a a
Q P
E
=
|
|

\
|
=
0
2 1
2
2
Signal Amplitude
The goal is maximize
(same as maximizing )
h(t)
H(f )
s
i
(t)+n(t)
a
i
(t)+n
0
(t)
Threshold
Detector
t=t
b
s(t)+n(t)
s(t)
T
a(t)+n(t)
a(t)
T
2
0
2

i
T
a
N
S
=
|

\
|
(#)
Matched Filter
We wish to find the filter transfer function
H
0
(f) that maximises equation (#).
This filter is called the Matched Filter.
For white channel noise, this filter
maximizes the signal amplitude to rms
noise ratio at T.
This is equivalent to minimizing the
probability of error
3
Signal and Noise Calculation
Signal output:
Output noise power or variance
( ) { }
df f S f H t a
f S f H F t a
ft j
i
i
2
1
exp ) ( ) ( ) (
) ( ) (

=
=
( )
df f H
N
df f H f S n
n o


=
= =
2
0
2
2
2
0
) (
2
) (
Ratio Calculation
Putting the pieces together gives:
Simplify Using Schwartz Inequality (Sklars)
Equality occurs only if f
1
(x) = k f
2
*
(x)


= =
|

\
|
df f H
N
df f S f H
a
N
S
fT j
i
T
2
0
2
2
2
0
2
) (
2
exp ) ( ) (

dx x f dx x f dx x f x f


2
2
2
1
2
2 1
) ( ) ( ) ( ) (
4
Simplifying the Equation
Apply the Schwartz Inequality:
Then:
H(f ) f
1
(x) S(f )e
j2fT
f
2
(x)
df f S df f H df f S f H
fT j


2 2
2
2
) ( ) ( exp ) ( ) (

Simplifying the Equation


=
=
|

\
|

\
|
df f S E
N
E
N
S
df f S
N N
S
T
T
2
0
2
0
) (
where
2
max
or
) (
2
(##)
5
Equality in (##) occurs if:
Simplifying the Equation
fT j
f kS f H f H
2 *
0
exp ) ( ) ( ) (

= =
or
( ) { }
fT j
f kS F t h
2 * 1
exp ) (

=
Since s(t) is a real-valued signal, we can write
( )
( )


=
elsewhere 0
0 T t t T s k
t h
Comparison of Convolution and Correlation
The output z(t) of a causal filter can be
described as
From matched filter, h(t)=s(T-t)
( ) ( ) ( ) ( ) ( ) d t h r t h t r t z
t
= =

0
*
6
( ) ( ) ( ) [ ]
( ) [ ]

d t T s r
d t T s r t z
t
t
+ =
=

0
0
When t=T,
( ) ( ) ( ) d s r T z
T

=
0
The equation above is known as the correlation of
r(t) with s(t)
Comparison of Convolution and Correlation
s
i
(T-t)
r(t)=s
i
(t)+n(t)
z(T)
Matched to s
i
(t)
Matched
Filter
r(t)=s
i
(t)+n(t)
z(T)
Matched to s
i
(t)
Correlator
X

T
0
s
i
(t)

You might also like