Sei sulla pagina 1di 46

# Chapter 8: M -ary Signaling Techniques

## A First Course in Digital Communications

Ha H. Nguyen and E. Shwedyk

February 2009

1/46

## Chapter 8: M -ary Signaling Techniques

Introduction
There are benefits to be gained when M -ary (M = 4)
signaling methods are used rather than straightforward binary
signaling.
In general, M -ary communication is used when one needs to
design a communication system that is bandwidth efficient.
Unlike QPSK and its variations, the gain in bandwidth is
accomplished at the expense of error performance.
To use M -ary modulation, the bit stream is blocked into
groups of bits the number of bit patterns is M = 2 .
The symbol transmission rate is rs = 1/Ts = 1/(Tb ) = rb /
symbols/sec there is a bandwidth saving of 1/ compared
to binary modulation.
modulation) and FSK.
A First Course in Digital Communications

2/46



mi



 



r (t )

si (t )





m i

w (t )

## w(t) is zero-mean white Gaussian noise with power spectral

density of N20 (watts/Hz).
Receiver needs to make the decision on the transmitted signal
based on the received signal r(t) = si (t) + w(t).
The determination of the optimum receiver (with minimum
error) proceeds in a manner analogous to that for the binary
case.
A First Course in Digital Communications

3/46

## Represent M signals by an orthonormal basis set, {n (t)}N

n=1 ,
N M:
si (t) = si1 1 (t) + si2 2 (t) + + siN N (t),
Z Ts
sik =
si (t)k (t)dt.
0

## Expand the received signal r(t) into the series

r(t) = si (t) + w(t)
= r1 1 (t) + r2 2 (t) + + rN N (t) + rN +1 N +1 (t) +
For k > N , the coefficients rk can be discarded.
Need to partition the N -dimensional space formed by
~r = (r1 , r2 , . . . , rN ) into M regions so that the message error
probability is minimized.
A First Course in Digital Communications

4/46

## N dimensiona l observatio n space

r = (r1 , r2 , , rM )

1
Choose s1 (t ) or m1

M
Choose s M (t ) or mM

2
Choose s2 (t ) or m2

Choose mi if
PN
2
2
(r

s
ik ) <
k=1 k
k=1 (rk sjk ) ;
j = 1, 2, . . . , M ; j 6= i.

PN

5/46

## M-ary Coherent Amplitude-Shift Keying (M-ASK)

r

2
cos(2fc t), 0 t Ts
Ts
r
2
= [(i 1)]1 (t), 1 (t) =
cos(2fc t), 0 t Ts ,
Ts
i = 1, 2, . . . , M.

si (t) = Vi

s1 (t )

s2 ( t )

s3 (t )

si (t )

r (t )

sk (t )

 sM 1 (t )

(k 1)

kTs

(M 2)

t = kTs

( )dt

( k 1) Ts

r1

sM (t )

1 (t )

(M 1)

 !
"

m i

w (t )
1 (t )
N0
WGN, strength
watts/Hz
2
A First Course in Digital Communications

6/46




## sk (t), if k 32 < r1 < k 12 , k = 2, 3, . . . , M 1

.
Choose
s (t), if r1 <
2

1
sM (t), if r1 > M 32
f (r1 sk (t ) )

r1
0

(k 1)

Choose s1 (t )

# Choose s (t )
M

Choose sk (t )

7/46

f (r1 sk (t ) )

r1

P [error] =

M
X

## P [si (t)]P [error|si (t)].

i=1

 p

P [error|si (t)] = 2Q / 2N0 , i = 2, 3, . . . , M 1.
 p

P [error|si (t)] = Q / 2N0 , i = 1, M.

2(M 1)  p
P [error] =
Q / 2N0 .
M

8/46

## Chapter 8: M -ary Signaling Techniques

The maximum and average transmitted energies can be reduced, without
any sacrifice in error probability, by changing the signal set to one which
includes the negative version of each signal.
r

2
si (t) = (2i 1 M )
cos(2fc t), 0 t Ts , i = 1, 2, . . . , M.
2} Ts
|
{z
Vi

'()

\$
'*)

&

3
2

PM

1 (t )

3
2

\$
2

1 (t )

M
(M 2 1)2
2 X
(2i 1 M )2 =
.
M
4M i=1
12
r
Es
(M 2 1)2
(12 log2 M )Eb
Eb =
=
=
log2 M
12 log2 M
M2 1

Es

i=1

Ei

9/46

## Probability of Symbol Error for M-ASK

s

2(M 1)
P [error] =
Q
M

6Es
(M 2 1)N0

2(M 1)
=
Q
M
s

2(M 1)
1
P [bit error] = P [symbol error] =
Q

M log2 M

6 log2 M Eb
M 2 1 N0

6 log2 M Eb
M 2 1 N0
!

## (with Gray mapping)

10

10

M=16
(W=1/4Tb)

P[symbol error]

10

M=8
(W=1/3T )
b

10

M=4
(W=1/2T )
b

10

M=2
(W=1/Tb)

10

10

10

15

20

25

E /N (dB)
b

W is obtained by using the W Ts = 1 rule-of-thumb. Here 1/Tb is the bit rate (bits/s).
A First Course in Digital Communications

10/46

## Chapter 8: M -ary Signaling Techniques

Baseband information signal
1
0
1
0

Tb

2Tb

3Tb

4Tb

5Tb

6Tb

7Tb

8Tb

9Tb

10Tb

7Tb

8Tb

9Tb

10Tb

BPSK Signalling
2
0
2
0

Tb

2Tb

3Tb

4Tb

5Tb

6Tb

2
0
2
0

2Tb

4Tb

6Tb

8Tb

10Tb

11/46

## M-ary Phase-Shift Keying (M-PSK)



(i 1)2
si (t) = V cos 2fc t
,
M

0 t Ts ,





(i 1)2
(i 1)2
si (t) = V cos
cos(2fc t)+V sin
sin(2fc t).
M
M
V cos(2fc t)
V sin(2fc t)

, 2 (t) =
.
Es
Es




p
p
(i 1)2
(i 1)2
si1 = Es cos
, si2 = Es sin
.
M
M

1 (t) =

12/46

## Signal Space Plot of 8-PSK



(i 1)2
,
si (t) = V cos 2fc t
M

0 t Ts ,

## i = 1, 2, . . . , M ; fc = k/Ts , k integer; Es = V 2 Ts /2 joules

2 (t )

s3 (t ) 011

010 s4 (t )

s2 (t ) 001

Es

110 s5 (t )
0

s1 ( t ) 000

1 (t )

s8 (t ) 100

111 s6 (t )
s7 (t ) 101
A First Course in Digital Communications

13/46

## Signal Space Plot of General M-PSK



(i 1)2
,
si (t) = V cos 2fc t
M

0 t Ts ,

2 (t )

s 2 (t )

Es

2 M
0

s1 (t )

2 M

1 (t )

s M (t )

14/46

,

Ts

t = Ts

r1

()dt

Compute

2
(r1 si1 )2 + (r+
2 si 2 )

r (t )

1 (t )

Ts

for i = 1, 2,

t = Ts

and choose

r2

( )dt

m i

,M

the smallest

2 (t )

P [error]

r2

=
=

Region 2
Choose s2 (t )

P [error|s1 (t)]
ZZ
1

r1 ,r2 Region 1

s2 ( t )
Es

M
0

s1 (t )

r1

Region 1
Choose s1 (t )

15/46

r2

Region 2
Choose s2 (t )
r2

s2 ( t )
Es

M
0

s2 ( t )

Es sin ( M )
s1 (t )

r1

Region 1
Choose s1 (t )

M
0

s1 (t )

Es ,0

r1

## P [error|s1 (t)] > P [r1 , r2 fall in 1 |s1 (t)], or

n  p
o
P [error|s1 (t)] > Q sin
2Es /N0 .
M
A First Course in Digital Communications

16/46

## Upper Bound of P [error] of M-PSK

r2

s2 ( t )

Es sin ( M )

r2

s1 (t )

Region 2
Choose s2 (t )

Es ,0

Es ,0

r1

s2 ( t )
Es

M
0

r2

s1 (t )

r1

Region 1
Choose s1 (t )

r1
s1 (t )

Es sin ( M )

s M (t )

P [error] < P [r1 , r2 fall in 1 |s1 (t)] + P [r1 , r2 fall in 2 |s1 (t)], or
  p

P [error] < 2Q sin
2Es /N0 ,
M
A First Course in Digital Communications

17/46

10

M=32
10

P[symbol error]

M=16
10

10

10

M=8

Exact

M=4
5

M=2
10

10

Lower bound
Upper bound

10

15

20

25

E /N (dB)
b

## With a Gray mapping, the bitq

error probability is approximated as:

1
2Eb
P [bit error]M -PSK log M Q
sin2 M
N0 .
2

18/46

## Comparison of BPSK and M-PSK

P [error]M -PSK Q

  2E
b
sin2
M N0

where Es = Eb .

p
P [error]BPSK = Q( 2Eb /N0 ).

M -ary BW/Binary BW

sin2 (/M )

3
4
5
6

8
16
32
64

1/3
1/4
1/5
1/6

0.44
0.15
0.05
0.0144

3.6 dB
8.2 dB
13.0 dB
17.0 dB

19/46

## M-ary Quadrature Amplitude Modulation (M-QAM)

M -QAM constellations are two-dimensional and they involve
inphase (I) and quadrature (Q) carriers:
r
2
I (t) =
cos(2fc t), 0 t Ts ,
Ts
r
2
Q (t) =
sin(2fc t), 0 t Ts ,
Ts
The ith transmitted M -QAM signal is:
r
r
2
2
0 t Ts
si (t) = VI,i
cos(2fc t) + VQ,i
sin(2fc t),
i = 1, 2, . . . , M
Ts
Ts
r
p
2
=
Ei
cos(2fc t i )
Ts

## VI,i and VQ,i are the information-bearing discrete amplitudes of the

2
2
two quadrature carriers, Ei = VI,i
+ VQ,i
and
1
i = tan (VQ,i /VI,i ).
A First Course in Digital Communications

20/46

(1,7)

(4,4)

Rectangle

Triangle

Rectangle

M =8

M =4

(1,3)

21/46

R2

(4,12)

(8,8)
Triangle

R2
R1

(1,5,10)

M = 16

R1

R1

R2

Hexagon
Rectangle

22/46

## A Simple Comparison of M-QAM Constellations

With the same minimum distance of all the constellations, a more
efficient signal constellation is the one that has smaller average
transmitted energy.
3456789: 4

./012

<=>?@AB
;

CDEDF

M =8

## Es for the rectangular, triangular, (1,7) and (4,4) constellations are

found to be 1.502 , 1.1252 , 1.1622 and 1.1832 , respectively.

23/46

## Chapter 8: M -ary Signaling Techniques

Rectangular M-QAM
Q (t )
M = 64

M = 16

M = 32

M =8

I (t )

M =4

The signal components take value from the set of discrete values
{(2i 1 M )/2}, i = 1, 2, . . . , M
2 .

24/46

## Modulation of Rectangular M-QAM

Each group of = log2 M bits can be divided into I inphase bits
and Q quadrature bits, where I + Q = .
Inphase bits and quadrature bits modulate the inphase and
Inphase bits
Select VI ,i

Infor. bits

GHI
JKLMNOL HP HQ

2
cos(2f c t )
Ts

si (t )

XYZ WN[VSL

Select VQ,i

## RST UQSVWJ NMMHQ

A First Course in Digital Communications

2
sin( 2f c t )
Ts
25/46

## Demodulation of Rectangular M-QAM

Due to the orthogonality of the inphase and quadrature signals, inphase
t = Ts
Inphase
decision

Ts

( ) dt
0

r (t ) = si (t ) + w (t )

I ( t ) =

2
cos(2 f c t )
Ts
Ts

( ) dt
0

Q (t ) =

2
sin(2 f ct )
Ts

Decision
Multiplexer
t = Ts

decision

## The most practical rectangular QAM constellation is one which

I = Q = /2, i.e., M is a perfect square and the rectangle is a square.
A First Course in Digital Communications

26/46

## Symbol Error Probability of M-QAM

For square constellations:

2
P [error] = 1 P [correct] = 1 1 PM [error] ,
s
!


1
3Es

P M [error] = 2 1
Q
,
(M 1)N0
M
where Es /N0 is the average SNR per symbol.
For general rectangular constellations:
s
"
!#2
3Es
P [error] 1 1 2Q
(M 1)N0
s
!
3Eb
4Q
(M 1)N0
where Eb /N0 is the average SNR per bit.
A First Course in Digital Communications

27/46

## Chapter 8: M -ary Signaling Techniques

10

M=256
10

P[symbol error]

M=64
10

M=16
10

10

M=4
5

M=2
10

10

Exact performance
Upper bound
5

10

15

20

25

Eb/N0 (dB)
A First Course in Digital Communications

28/46

q

2Es

## For M -PSK, approximate P [error] Q

sin
M .
q N0

3Eb
For M -QAM, use the upper bound 4Q
(M 1)N0 .
Comparing the arguments of Q() for the two modulations
gives:
3/(M 1)
M =
.
2 sin2 (/M )

10 log10 M

8
16
32
64
256
1024

1.65
4.20
7.02
9.95
15.92
21.93

dB
dB
dB
dB
dB
dB
29/46

1

10

P[symbol error]

10

10

4QAM or QPSK
4

10

M=4, 8, 16, 32
5

10

10

MPSK
MQAM

10
Eb/N0 (dB)

15

20
30/46

## M-ary Coherent Frequency-Shift Keying (M-FSK)

si (t) =

V cos(2fi t),
0,

0 t Ts
, i = 1, 2, . . . , M,
elsewhere

## where fi are chosen to have orthogonal signals over [0, Ts ].

 
(k i) 1 , (coherently orthogonal)
 2Ts
fi =
, i = 0, 1, 2, . . .
(k i) 1 , (noncoherently orthogonal)
Ts
2 (t )

s 2 (t )

Es
0

s3 ( t )

Es

s1 (t )

1 (t )

Es

3 (t )
A First Course in Digital Communications

31/46

## Chapter 8: M -ary Signaling Techniques

Choose mi if
M
M
X
X
2
(rk sik ) <
(rk sjk )2
k=1

k=1

j = 1, 2, . . . , M ; j 6= i,

Ts

ri > rj ,

Choose mi if
j = 1, 2, . . . , M ; j 6= i.

t = Ts

( )dt

r1

r (t )

s (t )
1 (t ) = 1
Es

Ts

\]^^_`
a]`
bcde` _a

Decision

t = Ts

()dt

rM

M ( t ) =

s M (t )
Es

32/46

## Symbol Error Probability of M-FSK

P [error] = P [error|s1 (t)] = 1 P [correct|s1 (t)].
P [correct|s1 (t)] = P [(r2 < r1 ) and and (rM < r1 )|s1 (t) sent].
Z
=
P [(r2 < r1 ) and and (rM < r1 )|{r1 = r1 , s1 (t)}]f (r1 |s1 (t))dr.
r1 =

## P [(r2 < r1 ) and and(rM < r1 )|{r1 = r1 , s1 (t)}] =

P [rj < r1 |{r1 = r1 , s1 (t)}] =
P [correct] =

r1

M
Y

j=2



1
2

exp
d.
N0
N0


 M1
1
2

exp
d

N0
N0
r1 =
=



1
(r1 Es )2

exp
dr1 .
N0
N0

Z

r1

33/46

10

10

M=64

P[symbol error]

10

M=4

M=32

M=2

10

M=16
5

10

M=8
6

10

10

1
P [error] = 1
2

4
y

x2 /2

6
8
Eb/N0 (dB)

dx

M 1

10

1
exp
2

12

14

16

2 log2 M Eb
N0

!2
dy.

34/46

## Bit Error Probability of M-FSK

Due to the symmetry of M -FSK constellation, all mappings
from sequences of bits to signal points yield the same bit
error probability.
For equally likely signals, all the conditional error events are
equiprobable and occur with probability
Pr[symbol error]/(M
1) = Pr[symbol error]/(2 1).

There are k ways in which k bits out of may be in error
The average number of bit errors per -bit symbol is
 

X
Pr[symbol error]
21
=

Pr[symbol error].
k
k
2 1
2 1
k=1

divided by :
Pr[bit error] =

## A First Course in Digital Communications

21
Pr[symbol error].
2 1

35/46

## Union Bound on the Symbol Error Probability of M-FSK

P [error] = P [(r1 < r2 ) or (r1 < r3 ) or, , or (r1 < rM )|s1 (t)].
Since the events are not mutually exclusive, the error
probability is bounded by:
P [error] < P [(r1 < r2 )|s1 (t)]+
P [(r1 < r3 )|s1 (t)] + + P [(r1 < rM )|s1 (t)].
p

But P [(r1 < rj )|s1 (t)] = Q
Es /N0 , j = 3, 4, . . . , M .
Then

p

p
Es /N0 < M Q
Es /N0 < M eEs /(2N0 ) .
P [error] < (M 1)Q
n 2o
where the bound Q(x) < exp x2 has been used.

36/46

Q(x) =

10

 2
 2
1

x
exp
d < exp
2
2
2
2

exp(x /2)
5

10

Q(x)
10

10

15

10

4
x

8
37/46

1

## Let M = 2 = e ln 2 and Es = Eb . Then

P [error] < e ln 2 eEb /(2N0 ) = e(Eb /N0 2 ln 2)/2 .
As , or equivalently, as M , the probability of
error approaches zero exponentially, provided that
Eb
> 2 ln 2 = 1.39 = 1.42 dB.
N0

## Since Es = Eb = V 2 Ts /2, then

P [error] < e ln 2 eV

2 T /(4N )
s
0

2 /(4N

0 )]

## If rb ln 2 + V 2 /(4N0 ) > 0, or rb < 4NV0 ln 2 the probability or

error tends to zero as Ts or M becomes larger and larger.
A First Course in Digital Communications

38/46

## Comparison of M-ary Signaling Techniques

A compact and meaningful comparison is based on the bit rate-to
bandwidth ratio, rb /W (bandwidth efficiency) versus the SNR per
bit, Eb /N0 (power efficiency) required to achieve a given P [error].
M -ASK with single-sideband (SSB) transmission, W = 1/(2Ts ) and
r 
b
= 2 log2 M (bits/s/Hz).
M -PSK (M > 2) must have double sidebands, W = 1/Ts and
r 
b
= log2 M, (bits/s/Hz),
W PSK

(Rectangular) QAM has twice the rate of ASK, but must have
double sidebands QAM and SSB-ASK have the same bandwidth
efficiency.
For M -FSK with the minimum frequency separation of 1/(2Ts ),
M
M
M
W = 2T
= 2(/r
= 2 log
rb , and
s
b)
2M
r 
2 log2 M
b
=
.
W FSK
M

39/46

## USSB Transmission of BPSK Signal

cos(2 f ct )

m(t )

sUSSB (t )

+
LTI Filter
h (t )
H( f )
H ( f ) = j sgn( f )

m (t )

sin(2 f ct )



Z
1
1 m(t )
m(t)

=
d
t

40/46

## Example of USSB-BPSK Transmitted Signal

(a) BPSK signal

V
0

Tb

2Tb

3Tb

4Tb

5Tb

6Tb

t
(b) USSBBPSK signal

V
0
V
0

Tb

2Tb

3Tb

4Tb

5Tb

6Tb

7Tb

8Tb

t
A First Course in Digital Communications

41/46

10
M=64
M=16

M=32
3
M=4

M=8
Bandwidthlimited
region: r /W>1

M=2

1
M=8

Powerlimited
region: r /W<1

r /W (bits/s/Hz)

M=64

M=16

M=16

0.5

M=32

0.3
0.2
0.1
10

M=64

5 1.6 0

## A First Course in Digital Communications

PSK
FSK

5
10
15
SNR per bit, Eb/N0 (dB)

20

25

30

42/46

## Chapter 8: M -ary Signaling Techniques

Two Statements
Consider information transmission over an additive white Gaussian
noise (AWGN) channel. The average transmitted signal power is
Pav , the noise power spectral density is N0 /2 and the bandwidth is
W . Two statements are:
1

## For each transmission rate rb , there is a corresponding limit

on the probability of bit error one can achieve.

## For some appropriate signalling rate rb , there is no limit on

the probability of bit error one can achieve, i.e., one can
achieve error-free transmission.
Which statement sounds reasonable to you?

43/46

## Shannons Channel Capacity



Pav
C = W log2 1 +
,
W N0
where W is bandwidth in Hz, Pav is the average power and N0 /2
is the two-sided power spectral density of the noise.
Shannon proved that it is theoretically possible to transmit
information at any rate rb , where rb C, with an arbitrarily
small error probability by using a sufficiently complicated
modulation scheme. For rb > C, it is not possible to achieve
an arbitrarily small error probability.
Shannons work showed that the values of Pav , N0 and W set
a limit on transmission rate, not on error probability!
The normalized channel capacity C/W (bits/s/Hz) is:




C
Pav
C Eb
= log2 1 +
= log2 1 +
.
W
W N0
W N0
A First Course in Digital Communications

44/46

## Shannons Capacity Curve

10

Eb
N0

2C/W 1
C/W

Channel capacity
limit, C/W

M=64
M=16

M=32
M=16
M=4

M=8

Shannon limit

Bandwidthlimited
region: r /W>1

M=2

1
M=8

Powerlimited
region: rb/W<1

r /W (bits/s/Hz)

3
2

M=64

M=16

0.5

M=32

0.3
0.2
0.1
10

M=64

5 1.6 0

PSK
FSK

5
10
15
SNR per bit, E /N (dB)
b

20

25

30

45/46

## Spectrum Efficiency of DVB-S2 Standard

5

Example 2:

Spectrum Efficiency

4. 5

4
3. 5
3
2. 5
2
1. 5
1

Example 1:

0. 5

0
-4

-2

SNR (dB)

10

12

14

16