0% found this document useful (0 votes)
30 views

Outline: - Receivers (Chapter 5) (Week 3 and 4)

The document outlines a digital communication system and discusses various components in it over 12 weeks. It covers topics like transmitters, receivers, channel capacity, error correction codes, equalization techniques, and spread spectrum. It then focuses on receivers and discusses optimal receivers, probability of error, correlation demodulators, matched filters, and the optimal detector for minimizing error between the received and transmitted signals. Different modulation schemes like PAM and QAM are also covered.

Uploaded by

Harsha
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 PPT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
30 views

Outline: - Receivers (Chapter 5) (Week 3 and 4)

The document outlines a digital communication system and discusses various components in it over 12 weeks. It covers topics like transmitters, receivers, channel capacity, error correction codes, equalization techniques, and spread spectrum. It then focuses on receivers and discusses optimal receivers, probability of error, correlation demodulators, matched filters, and the optimal detector for minimizing error between the received and transmitted signals. Different modulation schemes like PAM and QAM are also covered.

Uploaded by

Harsha
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 PPT, PDF, TXT or read online on Scribd
You are on page 1/ 31

Outline

• Transmitters (Chapters 3 and 4, Source Coding and


Modulation) (week 1 and 2)
• Receivers (Chapter 5) (week 3 and 4)
• Received Signal Synchronization (Chapter 6) (week 5)
• Channel Capacity (Chapter 7) (week 6)
• Error Correction Codes (Chapter 8) (week 7 and 8)
• Equalization (Bandwidth Constrained Channels) (Chapter
10) (week 9)
• Adaptive Equalization (Chapter 11) (week 10 and 11)
• Spread Spectrum (Chapter 13) (week 12)
• Fading and multi path (Chapter 14) (week 12)
Digital Communication System:

Transmitter
Receiver
Receivers (Chapter 5) (week 3 and
4)
• Optimal Receivers
• Probability of Error
Optimal Receivers
• Demodulators
• Optimum Detection
Demodulators
• Correlation Demodulator
• Matched filter
Correlation Demodulator
• Decomposes the
signal into
orthonormal
basis vector
correlation terms
• These are
strongly
correlated to the
signal vector
coefficients sm
Correlation Demodulator
• Received Signal model
– Additive White Gaussian Noise (AWGN)

r (t )  sm (t )  n(t )

– Distortion
• Pattern dependant noise
– Attenuation
• Inter symbol Interference
– Crosstalk
– Feedback
Additive White Gaussian Noise
(AWGN)

r (t )  sm (t )  n(t )
1
 rr ( f )   ss ( f )  N 0
2

1
 nn ( f )  N 0
2
i.e., the noise is flat in Frequency domain
Correlation Demodulator
• Consider each
demodulator
output
T
rk   r (t ) f k (t )dt
0
T
  sm (t ) f k (t )dt
0
T
  n(t ) f k (t )dt
0

 smk  nk
Correlation Demodulator
• Noise components
T T
E (nk nm )    E[n(t )n( )] f k (t ) f m ( )dtd
0 0

1 T
 N 0  f k (t ) f m ( )dt
2 0

1
 N 0 m  k {nk} are uncorrelated
 2
0 m  k Gaussian random
variables
Correlation Demodulator
• Correlator outputs
E (rk )  E ( smk  nk )  smk Have mean = signal

1  N (rk  smk ) 2 
p(r | s m )  exp   
(N 0 ) N /2
 k 1 N0 

m  1,2,  , M For each of the M codes

N  Number of basis functions (=2 for QAM)


Matched filter Demodulator
• Use filters whose
impulse response is
the orthonormal
basis of signal
• Can show this is
exactly equivalent to
the correlation
demodulator
Matched filter Demodulator
• We find that this
Demodulator
Maximizes the SNR
• Essentially show that
any other function
than f1() decreases
SNR as is not as well
correlated to
components of r(t)
The optimal Detector
• Maximum Likelihood (ML):

 1  N (rk  smk ) 2  
max p (r | s m )  max  exp   
 (N 0 )
N /2
 k 1 N0 
1 N
(rk  smk ) 2 
 max  N ln N 0   
2 k 1 N0 
N
 min  (rk  smk ) 2
k 1

 2
 min r  2r  s m  s m
2

The optimal Detector
• Maximum Likelihood (ML):

 2
min r  2r  s m  s m
2
  max2r  s m  sm
2


 max r  s m 
 m
2 
 

m  1,2,  M
Optimal Detector
• Can show that
N N T T
r  s m   rn smn    r (t ) f n (t )dt  sm (t ) f n (t )dt
n 1 n 1 0 0
T
  r (t ) sm (t )dt
so 0


max r  s m 
 m
 T
 max   r (t ) sm (t )dt 
 m

 
 2  0 2 
Optimal Detector
• Thus get new type of correlation demodulator
using symbols not the basis functions:
Alternate Optimal rectangular QAM
Detector
• M level QAM = 2 x M level PAM signals
• PAM = Pulse Amplitude Modulation
sm (t )  Am g (t ) cos 2f c t
 sm f (t )
(e)
d 2
g
d min
1 g

sm  Am
2
2 1
 g ( 2m  1  M ) d
f (t )  g (t ) cos 2f c t sm 
 g
2
m  1,2,  , M
The optimal PAM Detector

  g A2 
   
max r  s m  m   max r  s m  2
m

 2   2 
 
  g d 2 ( 2m  1  M ) 2 
 
 max r  s m  2 
 2 
 

 g
d 2 g For PAM
r  sm  d 
2 2
The optimal PAM Detector
d 2 g (e)
d min
r  sm  
2 2

sm

r  si r  si 1

(e)
d min
2
Optimal rectangular QAM Demodulator
• d = spacing of rectangular grid
s si 
1
 g (2i  1  M )d
M 2

f1 (t ) 
2
g (t ) cos 2f c t  Select si
 g for which

s d 2 g sm1  si
1
 
T

0
() dt

2


2
f 2 (t )  g (t ) sin 2f ct
 g

 T  Select si
 0
()dt for which

s d 2 g sm 2  si
1 
2


Probability of Error for rectangular
M-ary QAM
• Related to error probability of M PAM

M 1  d 2 g 
PM  P  r  sm  
M  2 

Accounts for ends


sm

r
Probability of Error for rec. QAM
• Assume Gaussian noise
d 2 g
0 r  sm

d 2 g
2

  2 

 x2 / N0
P  r  sm   e dx
2 N 0 d  g / 2
2
 
 d 2 
 erfc  
g

 2 N 0 
 d 2 
 2Q  g

 N 0 
Probability of Error for rectangular
M-ary QAM
• Error probability of M PAM

M 1  d g 
2

PM  2 Q 
M  2 N 0 
SNR for M-ary QAM
• Related to M PAM
• For M PAM find average energy in equally
probable signals

 
M
1
av  m
M m 1

d 2 g M

2 M m 1
 ( 2m  1  M) 2

 ( M  1)d  g
1 2

6
SNR for M-ary QAM
• Related to M PAM

Find average Power

Pav 
 av
T
1 d g
2

 ( M  1)
6 T
SNR for M-ary QAM
• Related to M PAM
SNR 
 av (ratio of powers)
Find SNR
N0
Tb  av
Then SNR per bit SNRb  
T N0


 M
av

N 0 log 2
1 d 2 g
 ( M  1)
6 N 0 log 2  M
SNR for M-ary QAM
• Related to M PAM

d 
2
g 
6 N 0 SNRb log 2  M
( M  1)

PM 
M 1  6 log M SNR
2 Q 2  b


M  ( M  1) 
SNR for M-ary QAM
• Related to PAM
M
• Now need to get
M-ary QAM from PAM

M½=16

M½=8

M½=4

M½=2
SNR for M-ary QAM
• Related to M PAM
PM  1  (1  P M ) 2 (1- probability of no QAM error)

 
2
 M 1  3 log M SNR 
 1  1  2 Q 2 b

 M  ( M  1)  
 

SNR b QAM
SNR b PAM 
2 (Assume ½ power in each PAM)
SNR for M-ary QAM
Probability of Symbol Error for QAM
• Related to M PAM
1.E-01

1.E-02

 

Probabilty of symbol Error PM


2
 M  1  3 log 2 M SNRb  
PM  1  1  2 Q  M=
 M  ( M  1)   1.E-03
 256
64
16
1.E-04 4

1.E-05

1.E-06
-6
-4
-2
0
2
4
6
8
10
12
14
16
18
20
22
24
SNR per bit (dB)

You might also like