Modulation, Demodulation and

Coding Course

Period 3 - 2005
Sorour Falahati
Lecture 7
 Last time we talked about:

 Some bandpass modulation schemes

 M-PAM, M-PSK, M-FSK, M-QAM

 How to perform coherent detection

2005-02-07 Lecture 7 2
 Example of two dim. modulation
 2 (t )
16QAM 8PSK s3 “011”
“010” “001”
 2 (t ) s4 s2
“0000” “0001” “0011” “0010”
s1 s2 s3 s4 Es
3
“110” s“000”
1
“1000” “1001” “1011” “1010” s5  1 (t )
s5 s6 s7 s8
1 “111” “100”
-3 -1 1 3
 1 (t )
s6 s8
s9 s10 -1
s11 s
12 “101” s7
 2 (t )
“1100” “1101” “1111” “1110”
QPSK “00”
s 2“01” s1
s13 s14 -3
s15 s
16
“0100” “0101” “0111” “0110” Es

 1 (t )

2005-02-07 Lecture 7
s3 “11” “10”
s4 3
 Today, we are going to talk about:
 How to perform non-coherent detection

 How to calculate the average probability

of symbol error for different modulation
schemes that we studied?

 How to compare different modulation

schemes based on their error
performances?

2005-02-07 Lecture 7 4
 Coherent detections
 Coherent detection
 requires carrier phase recovery at the
receiver and hence, circuits to perform
phase estimation.
 Source of carrier-phase mismatch at the
receiver:
 Propagation delay causes carrier-phase offset in
the received signal.
 The oscillators at the receiver which generate

the carrier signal, are not usually phased locked

to the transmitted carrier.

2005-02-07 Lecture 7 5
 Coherent detection
 Circuits such as Phase-Locked-Loop (PLL)
are implemented at the receiver for
carrier phase estimation (   ˆ ).

I branch
2 Ei 2
r (t )  gT (t ) cos i t  i (t )     n(t ) cos c t  ˆ 
T T
PLL
Used by
Oscillator 90 deg. correlators
2
sin  c t  ˆ 
T
Q branch

2005-02-07 Lecture 7 6
 Non-coherent detection
 Non-coherent detection
 does not require carrier phase recovery
(uses differentially encoded mod. or energy
detectors) and hence, has less complexity at
the price of higher error rate.
 No need in a reference in phase with the
received carrier

2005-02-07 Lecture 7 7
 Types of non-coherent detection

 Differentially coherent detection

 Differential PSK (DPSK)
 The information bits and previous symbol,
determine the phase of the current symbol.
 Energy detection
 Non-coherent detection for orthogonal signals
(e.g. M-FSK)
Carrier-phase offset causes partial correlation
between I and Q braches for each candidate signal.
 The received energy corresponding to each
candidate signal is used for detection.

2005-02-07 Lecture 7 8
 Differentially coherent detection
 Differentially encoding binary PSK
(DPSK)
 The symbol phase changes if the current bit
is different from the previous bit.
 Non-coherent detection
 assumes slow variation in carrier-phase
mismatch during two symbol intervals.
 uses the phase difference between two
successive symbols for detection.

2005-02-07 Lecture 7 9
 Energy detection
 Non-coherent detection of BFSK
2 / T cos(1t )

T z11
 2
0 2 2
z11  z12
2 / T sin(1t )
T z12
r (t )
0
 2 + z (T )
Decision stage:
m̂
2 / T cos( 2t ) if z (T )  0, mˆ  1
z 21 if z (T )  0, mˆ  0
T
 2 -
0

2 2
2 / T sin(2t ) z 21  z 22
T z 22
0
 2

2005-02-07 Lecture 7 10
 Error probability of bandpass modulation

 Before evaluating the error probability, it is

important to remember that:
 Type of modulation and detection ( coherent or non-
coherent), determines the structure of the decision circuits
and hence the decision variable, denoted by z.
 The decision variable, z, is compared with M-1 thresholds,
corresponding to M decision regions for detection purposes.

 1 (t )
T r1

0
 r1  Decision
r (t )   r
r Circuits
m̂
 N (t )   Compare z
T rN  with threshold.

0 rN

2005-02-07 Lecture 7 11
 Error probability …
 The matched filters output (observation vector= r ) is
the detector input and the decision variable is a z  f (r )
function of r , i.e.
 For MPAM, MQAM and MFSK with coherent detection z  r

 For MPSK with coherent detection z  r

 For non-coherent detection (M-FSK and DPSK), z | r |

 We know that for calculating the average probability of

symbol error, we need to determine

Pr(r lies inside Zi | s i sent)  Pr(z satisfies condition Ci | s i sent)

 Hence, we need to know the statistics of z,
which depends on the modulation scheme
and the detection type.
2005-02-07 Lecture 7 12
 Error probability …
 AWGN channel model: r  s i  n
 Signal vector s i  (ai1 , ai 2 ,..., aiN ) is deterministic.
 Elements of noise vector n  (n1 , n2 ,..., nN ) are i.i.d
Gaussian random variables with zero-mean and
variance N 0 / 2 . The noise vector pdf is
1  n
2

pn (n)  exp  
 N 0  N /2
 N0 
 
 The elements of observed vector r  (r1 , r2 ,..., rN ) are
independent Gaussian random variables. Its pdf is
1  r  s
2

pr (r | s i )  exp i 
 N 0  N / 2  N 0 

2005-02-07 Lecture 7 13
 Error probability …

 BPSK and BFSK with coherent detection:

 s1  s 2 / 2 
PB  Q 
 N /2 
 2 (t )  0   1 (t )
“0” “1”
BPSK s1 s2 BFSK
“0”
s1 Eb
s1  s 2  2 Eb

 Eb Eb  1 (t )
s 2“1”
 2 (t )
s1  s 2  2 Eb Eb

 2 Eb   Eb 
PB  Q 
 PB  Q 

 N0   N0 

2005-02-07 Lecture 7 14
 Error probability …
 Non-coherent detection of BFSK Decision variable:
2 / T cos(1t )
Difference of envelopes
T r11
 2 z  z1  z 2

0 2 2
z1  r11  r12
2 / T sin(1t )
T r12
r (t )

0
 2 + z
Decision rule:
m̂
2 / T cos( 2t ) if z (T )  0, mˆ  1
r21 if z (T )  0, mˆ  0
T
 2 -

0

2 2
2 / T sin(2t ) z 2  r21  r22
T r22

0
 2

2005-02-07 Lecture 7 15
 Error probability – cont’d
 Non-coherent detection of BFSK …
1 1
PB  Pr( z1  z2 | s 2 )  Pr( z 2  z1 | s1 )
2 2
 Pr( z1  z 2 | s 2 )  E  Pr( z1  z2 | s 2 , z2 )

  Pr( z1  z2 | s 2 , z 2 ) p( z2 | s 2 )dz2  

  p( z | s )dz  p( z | s )dz
0 0  z2 1 2 1
 2 2 2

1  Eb 
PB  exp   Rayleigh pdf Rician pdf
2  2N0 

 Similarly, non-coherent detection of DBPSK

1  Eb 
PB  exp  
2  N0 
2005-02-07 Lecture 7 16
 Error probability ….
 Coherent detection of M-PAM
 Decision variable:

z  r1
“00” “01” “11” “10”
s1 s2 s3 s4
4-PAM  1 (t )
 3 Eg  Eg 0 Eg 3 Eg

 1 (t )
T r1

ML detector
r (t ) (Compare with M-1 thresholds) m̂
0

2005-02-07 Lecture 7 17
 Error probability ….
 Coherent detection of M-PAM ….
 Error happens if the noise, n1  r1  s m , exceeds in amplitude
one-half of the distance between adjacent symbols. For symbols
on the border, error can happen only in one direction. Hence:

Pe (s m )  Pr | n1 || r1  s m | E g  for 1  m  M ;

Pe (s1 )  Pr n1  r1  s1  E g  
and Pe (s M )  Pr n1  r1  s M   E g 
M 2
     
M
1 1 1
PE ( M ) 
M
 P (s
m 1
e m )
M
Pr | n1 | E g  Pr n1  E g  Pr n1   E g
M M
2( M  1)  2 E g 

2( M  1)
M
Pr n1  E g  
2( M  1) 
M  E
pn1 (n)dn 
M
Q
 N0


 
g

( M 2  1)
Es  (log 2 M ) Eb  Eg
3
Gaussian pdf with
2( M  1)  6 log 2 M Eb  zero mean and variance N0 / 2
PE ( M )  Q 
 M 1 N0 
2
M
2005-02-07 Lecture 7 18
 Error probability …
 Coherent detection  2 (t )

of M-QAM
“0000” “0001”
s1 s2 s 3“0011”s 4 “0010”

“1000”
s “1001”
s s 7“1011”s8 “1010”
5 6

16-QAM  1 (t )
s9 s10 s11 s12
“1100” “1101” “1111” “1110”

 1 (t ) s13 s14 s15 s16

T r1 ML detector “0100” “0101” “0111” “0110”

0 (Compare with M  1 thresholds)

r (t ) Parallel-to-serial m̂
converter
 2 (t )
T r2 ML detector

0 (Compare with M  1 thresholds)

2005-02-07 Lecture 7 19
 Error probability …
 Coherent detection of M-QAM …
 M-QAM can be viewed as the combination of two M  PAM
modulations on I and Q branches, respectively.
 No error occurs if no error is detected on either I and Q
branches. Hence:
 Considering the symmetry of the signal space and orthogonality
of I and Q branches:

PE ( M )  1  PC ( M )  1  Pr(no error detected on I and Q branches)

Pr( no error detected on I and Q branches)  Pr(no error on I)Pr(no error on Q)


 Pr(no error on I) 2  1  PE  M  2

 1   3 log 2 M Eb 
PE ( M )  41  Q Average probability of
 M   M  1 N 0  symbol error for M  PAM
2005-02-07 Lecture 7 20
 Error probability …
 Coherent detection
of MPSK
 2 (t )
s 3 “011”
“010”
s4 s“001”
2
Es

“110”
s“000”
1
8-PSK s5  1 (t )

“111”
 1 (t ) s 6 s8“100”
T r1 “101”s 7


0
r (t ) r1 ˆ Compute Choose m̂
arctan
 2 (t ) r2 | i  ˆ | smallest
T

0
r2 Decision variable
z  ˆ  r
2005-02-07 Lecture 7 21
 Error probability …
 Coherent detection of MPSK …
 The detector compares the phase of observation vector to M-1
thresholds.
 Due to the circular symmetry of the signal space, we have:
M
1  /M
PE ( M )  1  PC ( M )  1 
M
 P (s
m 1
c m )  1  Pc (s1 )  1  
 / M
pˆ ( )d

where
2 Es  E  
pˆ ( )  cos( ) exp  s sin 2  ; |  |
 N0  N0  2
 It can be shown that

 2 Es     2 log 2 M  Eb    
PE ( M )  2Q sin    or PE ( M )  2Q sin   
 N0  M 
 N0  M 

2005-02-07 Lecture 7 22
 Error probability …
 Coherent detection of M-FSK

 1 (t )
T r1

0
 r1  ML detector:
r (t )   r
r Choose
the largest element m̂
 M (t )   in the observed vector
T rM 

0 rM

2005-02-07 Lecture 7 23
 Error probability …
 Coherent detection of M-FSK …
 The dimensionality of signal space is M. An upper
bound for average symbol error probability can be
obtained by using union bound. Hence
 Es 
PE ( M )   M  1 Q 

 N0 

or, equivalently


PE ( M )   M  1 Q
 log 2 M  Eb 
N0 
 

2005-02-07 Lecture 7 24
 Bit error probability versus symbol error
probability
 Number of bits per symbol k  log 2 M
 For orthogonal M-ary signaling (M-FSK)
PB 2 k 1 M /2
 k 
PE 2  1 M  1
PB 1
lim 
k  P 2
E

 For M-PSK, M-PAM and M-QAM

PE
PB  for PE  1
k

2005-02-07 Lecture 7 25
Probability of symbol error for binary
modulation

Note!
• “The same average symbol
PE energy for different sizes of
signal space”

Eb / N 0 dB
2005-02-07 Lecture 7 26
Probability of symbol error for M-PSK

Note!
• “The same average symbol
energy for different sizes of
PE signal space”

Eb / N 0 dB
2005-02-07 Lecture 7 27
Probability of symbol error for M-FSK

Note!
• “The same average symbol
energy for different sizes of
PE signal space”

Eb / N 0 dB
2005-02-07 Lecture 7 28
 Probability of symbol error for M-PAM

Note!
• “The same average symbol
energy for different sizes of
PE signal space”

Eb / N 0 dB
2005-02-07 Lecture 7 29
 Probability of symbol error for M-
QAM

Note!
• “The same average symbol
energy for different sizes of
PE signal space”

Eb / N 0 dB
2005-02-07 Lecture 7 30
Example of samples of matched filter output
for some bandpass modulation schemes

2005-02-07 Lecture 7 31