Wireless communication

Wireline/Wireless transmission
over block-fading channel
Editor: 10DTLT - Group 3B
Email: [email protected]
Contents
Introduction
Wireless channel models
Digital modulation
Channel estimation
Simulation
Conclusion
[email protected]
Wireless Communication 2
Requirements
Simulate Wireline & Wireless transmission over block-fading
channel
Using M-ary QAM (expanded M=2048)
Estimating imperfect channel
[email protected]
Wireless Communication Page 3
Wireless channel models
Characterized by Variations of the channel strength over time
and frequency
Large-scale fading
Small-scale fading
[email protected]
Wireless Communication Page 4
Wireless channel models
Path loss is caused by dissipation of the power radiated by the
transmiiter as well as effects of the propagation channel.
Shadowing is caused by obstacles between the transmitter and
receiver that absorb power.
[email protected]
Wireless Communication Page 5
Path loss models
The path loss is defined by


Several useful empirical models for macrocellular systems:
Hatas model
Lees model
[email protected]
Wireless Communication Page 6
( )
) 4 (
2
sin
4
4 log 10 log 10
2
2
10 10

|
|
.
|

\
|
|
.
|

\
|
=
|
|
.
|

\
|
=
d
h h
d P
G G P
L
c
m b c
r
r t t
dB p

t
t

Multipath fading channels

=

( ) ( ) | |

=
i
i m l
W m W l c W m h
1 ,
sin t o
Advantages over analog modulation
Cheaper, faster and more power-efficient than analog
transceiver.
More importantly, digital modulation offers:
Higher data rates,
Powerful error correction techniques,
Resistance to channel imparments,
More efficient multiple access stratergies, and
Better security and privacy.
[email protected]
Wireless Communication 8
Signal and system model
Every T seconds, the system sends bits of
information through the channel for a data rate of
bps.
There are possible sequences of K bits and each bit
sequence of length K comprises a message ,
where is the set of all such messages
[email protected]
Wireless Communication 9
M K
2
log =
T K R =
K
M 2 =
{ } M b b m
K i
e e , ,
1

{ }
M
m m M , ,
1
=
Signal space representation
By representing signals as projection onto a set of basis
functions, we obtain a one-to-one correspondence between the
set of transmiteed signals and their vector representations.
Thus, we can analyze signals in finite-dimensional vector
space instead of infinite-dimensional function space, using
classical notions of distance for vector spaces.
[email protected]
Wireless Communication 10
Figure 3: An example of two-demensional signal space
Signal space representation (cont.)
In order to analyze signals via a signal space representation,
we need to use some definitions for the vector characterization
in vector spae . In particular, the length of a vector in is
defined as



The distance between two signal constellation points s
i
and s
k

is thus
[email protected]
Wireless Communication 11
) 22 (
1
2
,
=
=
N
j
j i i
s s
( ) | | ) 23 ( ) ( ) (
0
2
1
2
, ,
}

= =
=
T
k i
N
j
j k j i k i
dt t s t s s s s s
Maximum likelihood detection
Once the signal space has been partitioned by decision regions,
for a received vector , the optimal receiver outputs the
message estimate . The receiver processing consists
of
Computing the received vector r from r(t)
Finding which decision region Z
i
contain r
And outputting the corresponding message m
i
.
[email protected]
Wireless Communication 12
i
Z r e
i
m m =
Error probability analysis
With , the error probability of the ML
receiver:
P
e
is the probability of a symbol (message) error where m
i

corresponds to a message with bits. However,
typically the bit error probability, also called the bit error rate
(BER), is interesting than in the symbol error probability, since
bit errors drive the performace of higher layer networking
protocols and end-to-end performace.
[email protected]
Wireless Communication 13
( ) M sent m p
i
/ 1 =
M
2
log
) 42 (
log
2
M
P
P
e
b
~
Quarature amplitude modulation MQAM
the modulation signal can be represent as

in terms of its in-phase and quarature componets as


The energy in is


The distance between any pair of symbols in the signal
constellation is


[email protected]
Wireless Communication 14
) (t s
i
( ) ( ) ) 43 ( ) ( 2 cos ) ( ) ( 2 cos ) ( ) (
0 0
| | t o | u t o + + = + + = t t f t t t f t t s
C C
( ) ( )
( ) ( ) ) 44 ( 2 sin ) ( 2 cos ) (
2 sin ) ( sin ) ( 2 cos ) ( cos ) ( ) (
t f t s t f t s
t f t t t f t t t s
c Q c I
c c
t t
t | o t | o
=
=
) 46 ( ) (
2
0
2
i
T
i S
A dt t s E
i
= =
}
( ) ( ) ) 47 (
2
2 , 2 ,
2
1 , 1 , , k i k i k i k i
s s s s s s d + = =
MQAM (cont.)
[email protected]
Wireless Communication 15
4QAM and 16QAM constellations and Decision regions for
MQAM with M=16
MQAM (cont.)
[email protected]
Wireless Communication 16
Table 1 BPSK encoding table
Input bit (b0) I-out Q-out
0 -1 0
1 1 0
Table 2 QPSK encoding table
Input bit
(b0)
I-out Input bit
(b1)
Q-out
0 -1 0 -1
1 1 1 1
Channel estimation in block-fading channel
To reduce the complexity of channel estimation with discrete-
time domain (Without OFDM), the number of resolvable path
L is equal to 1 (L=1) on block fading channel.
In once iteration, we utilize h
0
which is assumed the known
CIR of header. So the known received signal is

where y
h
and s
h
are symbols in header of each frame which is
transmitted. And w
h
is coressponding noise samples.
Using the known received signal, one could compute channel
estimation of all symbols in this frame


where H is the length of header.
[email protected]
Wireless Communication 17
h h h
w s h y + =
0
,
1

0
0
1
0
h
w
h
s
y
H
h
h
H
i
i
i
+ = =

=
Simulation model
The Matlab script performs the following
Setting system
Compute power average of noise and signal power
Generating symbol sequences to transmit
Digital modulation
Channel estimation h
0
Adding AWGN noise
Demodulation
Count the number of errors to compute BER
Repeating of multiple values of number of bursts
Repeating values of SNR.
Plot the simulation results
[email protected]
Wireless Communication 18
Simulation
Parameters:
M = 4
Number of frames = 100
Number of MQAM complex symbols = 1000
Header = 1% frame
[email protected]
Wireless Communication 19
M=4
[email protected]
Wireless Communication 20
0 5 10 15 20 25 30
10
-4
10
-3
10
-2
10
-1
10
0
SNR (dB)
B
i
t

e
r
r
o
r

r
a
t
e

(
B
E
R
)


Wireless transmission without CSI
Wireless transmission with perfect channel
Wireless transmission with imperfect channel
Wireline transmission, i.e., AWGN channels
M=2048
[email protected]
Wireless Communication 21
0 5 10 15 20 25 30
10
-2
10
-1
10
0
SNR (dB)
B
i
t

e
r
r
o
r

r
a
t
e

(
B
E
R
)


Wireless transmission without CSI
Wireless transmission with perfect channel
Wireless transmission with imperfect channel
Wireline transmission, i.e., AWGN channels
Conclusion
From these results, it can be seen that the BER in imperfect
channel which is always higher than BER in perfect channel. If
we increase the length of header, the two curves will tend
nearerly. The result also shown performance which dues to
number of constellation points M.

[email protected]
Wireless Communication 22
[email protected]