PhD Seminar (ENCS8051)

New Power Allocation Strategy Among MIMO

Spatial Subchannels - Beam-Nulling
Mabruk Gheryani
Supervisor: Dr. Yousef R. Shayan
November 10, 2007
Abstract
Since the discovery of multiple-input-multiple-output (MIMO) channel ca-
pacity, a lot of research eorts have been put into this eld. It has been
recognized that adaptive techniques proposed for single-input-single-output
(SISO) channel, can also be applied to improve MIMO channel capacity.
The original MIMO channel can be converted to uncorrelated spatial
subchannels via singular value decomposition. Strategies of power allocation
over these spatial subchannels for various channel state information scenarios
have been studied, such as equal power, water-lling, beamforming. A novel
scheme called beam-nulling has been proposed. Using the same feedback
bandwidth as beamforming, the new scheme discards the weakest spatial
subchannel but exploits the other subchannels. Hence, it can achieve signi-
cant high capacity, which is near the optimal water-lling solution at medium
signal-to-noise ratio.
Additionally, the capacities of equal power, beamforming and beam-
nulling are compared through theoretical analysis. Numerical results of the
three schemes and the optimal water-lling scheme are also compared. It is
shown that at low signal-to-noise ratio, beamforming nears the optimal, at
medium signal-to-noise ratio, beam-nulling nears the optimal, and at high
signal-to-noise ratio, equal power nears the optimal.
As future works, performance and application of beam-nulling and new
adaptation strategies will be further investigated. The remaining tasks have
been scheduled.
Contents
List of Tables 3
List of Tables 4
Notations and Abbreviations 5
1 Introduction 7
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2 Literature Survey . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3 Problem Statement and Objective . . . . . . . . . . . . . . . . 10
1.4 Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2 Preliminaries 12
2.1 Channel Model . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.2 Power Allocation Among Spatial Subchannels . . . . . . . . . 14
2.2.1 Equal Power . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2.2 Water-Filling . . . . . . . . . . . . . . . . . . . . . . . 16
2.2.3 Eigen-Beamforming . . . . . . . . . . . . . . . . . . . . 18
1
3 Recent Research Results 22
3.1 Beam-Nulling . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.2 Comparisons Among the Four Schemes . . . . . . . . . . . . . 26
3.2.1 At low SNR . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2.2 At medium SNR . . . . . . . . . . . . . . . . . . . . . 33
3.2.3 At high SNR . . . . . . . . . . . . . . . . . . . . . . . 34
4 Conclusions and Future Works 35
4.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.2 Future Works . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.2.1 Performance and Application of Beam-Nulling . . . . . 36
4.2.2 Extended Adaptive Frameworks . . . . . . . . . . . . . 36
4.3 List of Publications . . . . . . . . . . . . . . . . . . . . . . . . 37
4.4 Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2
List of Tables
4.1 Schedule for the remaining tasks. . . . . . . . . . . . . . . . . 38
3
List of Figures
2.1 MIMO channel model. . . . . . . . . . . . . . . . . . . . . . . 13
2.2 Capacity for equal power allocation. . . . . . . . . . . . . . . . 15
2.3 Water-lling scheme. . . . . . . . . . . . . . . . . . . . . . . . 16
2.4 Capacity for water-lling. . . . . . . . . . . . . . . . . . . . . 18
2.5 Eigen-beamforming scheme. . . . . . . . . . . . . . . . . . . . 19
2.6 Capacity for eigen-beamforming. . . . . . . . . . . . . . . . . . 21
3.1 Beam-Nulling scheme. . . . . . . . . . . . . . . . . . . . . . . 23
3.2 Capacity for beam-nulling. . . . . . . . . . . . . . . . . . . . . 26
3.3 2 2 Rayleigh fading channel. . . . . . . . . . . . . . . . . . . 29
3.4 3 3 Rayleigh fading channel. . . . . . . . . . . . . . . . . . . 30
3.5 4 4 Rayleigh fading channel. . . . . . . . . . . . . . . . . . . 31
3.6 5 5 Rayleigh fading channel. . . . . . . . . . . . . . . . . . . 32
4.1 Gantt chart for the remaining tasks. . . . . . . . . . . . . . . . 39
4
Notations and Abbreviations
X: upper bold letter for matrix
x: lower bold letter for column vector
X
H
: hermitian of X
X
T
: transpose of X
diag[x]: a diagonal matrix with x on its main diagonal
tr(X): trace of X
det(X): determinant of X
{x}: a set of x
E(x): expectation of x
AWGN: additive white Gaussian noise
BF: beamforming
BN: beam-nulling
CSI: channel state information
EQ: equal power
FDFR: Full Diversity Full Rate
LDC: linear dispersion code
MIMO: multiple-input-multiple-output
5
SISO: single-input-single-output
SNR: signal-to-noise ratio
ST: space-time
STBC: space-time block code
SVD: singular value decomposition
WF: water-lling
i.i.d.: independently identically distributed
6
Chapter 1
Introduction
1.1 Motivation
Since the discovery of multiple-input-multiple-output (MIMO) channel ca-
pacity, a lot of research eorts have been put into this eld [1] [2].
To combat channel quality variation and thus further improve system
performance such as power eciency, error rate and average data rate, it has
been recognized that adaptive techniques proposed for single-input-single-
output (SISO) channel [3] [4], can also be applied to improve MIMO chan-
nel capacity. In this method, a feedback channel is utilized to provide CSI
from the receiver to the transmitter. According to the feedback of CSI, the
transmitter will adjust transmission parameters, such as power allocation,
modulation, coding rate, etc. This is conditioned by the fact that the chan-
nel keeps relatively constant until the transmitter receives the CSI and then
transmits the next data block accordingly.
The ideal scenario is that the transmitter has full knowledge of channel
7
state information (CSI) fed back from the receiver and the CSI keeps con-
stant before the transmitter sends information to the receiver. With such a
perfect CSI feedback, the original MIMO channel can be converted to multi-
ple uncoupled SISO channels via singular value decomposition (SVD) at the
transmitter and the receiver [1]. In other words, the original MIMO chan-
nel can be decomposed into several orthogonal spatial subchannels with
various propagation gains. To achieve better performance, various strategies
to allocate constrained power to these subchannels can be implemented de-
pending on the availability of CSI at the transmitter [6]- [8]. In this study,
we propose to develop new scheme with limited CSI feedback.
1.2 Literature Survey
If the transmitter has full knowledge about channel matrix, i.e., full CSI,
the so-called water-lling (WF) principle is performed on each spatial sub-
channel to maximize the channel capacity. This scheme is optimal in this
case. Note that in practice, water-lling power allocation has to cooper-
ate with the other adaptive parameters to improve performance or/and data
rate, such as constellation and coding rate. For example, over time-invariant
MIMO channels, it is known that the optimal performance (ergodic capac-
ity) is attained by power water-lling across channel eigenvalues with the
total power constraint [1]. Also, for time-varying MIMO channels, the op-
timal performance is obtained through power water-lling over both space
and time domains with the average power constraint [11]. The space-time
WF-based scheme and the spatial WF-based scheme for MIMO fading chan-
8
nels were compared in [12]. The comparison shows that for Rayleigh channels
without shadowing, space-time WF-based scheme gains little in capacity over
spatial WF-based scheme. However, for Rayleigh channels with shadowing,
space-time WF-based scheme achieves higher spectral eciency per antenna
over spatial WF-based scheme. A WF-based scheme using imperfect CSI in
MIMO systems was studied in [13].
For the WF-based scheme, the feedback bandwidth for the full CSI grows
with respect to the number of transmit and receive antennas and the perfor-
mance is often very sensitive to channel estimation errors. To overcome these
disadvantages, various beamforming techniques for MIMO channels are also
investigated intensively. Beamforming is a linear signal processing technique
that control the complex weights of the transmit and receive antennas jointly
to optimize the signal-to-noise ratio (SNR) in one direction [10]. In other
words, beamforming can increase the sensitivity in the direction of wanted
signals but decrease the sensitivity in the direction of interfering signals.
In an adaptive beamforming scheme, complex weights of the transmit
antennas are fed back from the receiver. If only partial CSI is available at
the transmitter such as the eigenvector associated with the strongest spa-
tial subchannel, eigen-beamforming [7] is optimal in this case. The eigen-
beamforming scheme only allocates power to the strongest spatial subchan-
nel but can achieve full diversity and high signal-to-noise ratio (SNR). Also,
in practice, the eigen-beamforming scheme has to cooperate with the other
adaptive parameters to improve performance or/and data rate, such as con-
stellation and coding rate.
There are also other beamforming schemes based on various criteria.
9
For example, an optimal eigen-beamforming space-time block code (STBC)
scheme based on channel mean feedback was proposed in [7]. A MIMO sys-
tem based on transmit beamforming and adaptive modulation was proposed
in [8], where the transmit power, the signal constellation, the beamforming
direction, and the feedback strategy were considered jointly. The analysis of
MIMO beamforming systems with quantized CSI for uncorrelated Rayleigh
fading channels was provided in [9].
1.3 Problem Statement and Objective
Note that the conventional beamforming is optimal in terms of maximizing
the SNR at the receiver. However, it is sub-optimal from a MIMO capac-
ity point of view, since only one data stream, instead of parallel streams,
is transmitted through the MIMO channel [17]. Inspired by existing beam-
forming schemes, we will propose a new beamforming-like technique called
minimum eigenvector beam-nulling (BN). This scheme uses the same feed-
back bandwidth as beamforming. That is, only one eigenvector is fed back
to the transmitter. Unlike the eigen-beamforming scheme in which only the
best spatial subchannel is considered, in the beam-nulling scheme, only the
worst spatial subchannel is discarded. Hence, the loss of channel capacity as
compared to the optimal water-lling scheme can be reduced. In this scheme,
power is only allocated to the other good spatial subchannels. As compared
to the beamforming scheme, this scheme outperforms signicantly in terms
of channel capacity.
10
1.4 Organization
The rest of the seminar is organized as follows. In Chapter 2, preliminaries
are presented. In Chapter 3, current research results are provided. Finally
in Chapter 4, we will conclude the seminar and present the research schedule
for the remaining tasks.
11
Chapter 2
Preliminaries
2.1 Channel Model
In this study, the channel is assumed to be a Rayleigh at fading channel
with N
t
transmit and N
r
(N
r
N
t
) receive antennas. Lets denote the
complex gain from transmit antenna n to receiver antenna m by h
mn
and
collect them to form an N
r
N
t
channel matrix H = [h
mn
]. The channel
is known perfectly at the receiver but partially informed to the transmitter.
The entries in H are assumed to be independently identically distributed
(i.i.d.) symmetrical complex Gaussian random variables with zero mean and
unit variance. The MIMO channel is shown in Figure 2.1.
The symbol vector at the N
t
transmit antennas is denoted by
x = [x
1
, x
2
, . . . , x
Nt
]
T
. During the duration of x, H keeps constant but vary
for next x. According to information theory [5], the optimal distribution
of the transmitted symbols is Gaussian. Thus, the elements {x
i
} of x are
assumed to be i.i.d. Gaussian variables with zero mean and unit variance,
12
h
mn
x
n
x
1
y
m
y
1
y
Nr
x
Nt
Figure 2.1: MIMO channel model.
i.e., E(x
i
) = 0 and E|x
i
|
2
= 1. By using the linear model, the received
signals can be written as
y = Hx +z (2.1)
z is the additive white Gaussian noise (AWGN) vector with i.i.d. symmetrical
complex Gaussian elements of zero mean and variance
2
z
.
The singular-value decomposition of H can be written as
H = UV
H
(2.2)
where U is an N
r
N
r
unitary matrix, is an N
r
N
t
matrix with singular
values {
i
} on the diagonal and zeros o the diagonal, and V is an N
t

N
t
unitary matrix. For convenience, we assume
1

2
. . .
N
t
, U =
[u
1
u
2
. . . u
Nr
] and V = [v
1
v
2
. . . v
Nt
]. {u
i
} and {v
i
} are column vectors. We
13
assume that the rank of H is r (r N
t
). That is, the number of non-zero
singular values is r.
From (2.2), the original channel can be considered as consisting of r
uncoupled parallel subchannels. Each subchannel corresponds to a singular
value of H. In the following context, the subchannel is also referred to as
spatial subchannel. For instance, one spatial subchannel corresponds to

i
, u
i
and v
i
.
2.2 Power Allocation Among Spatial Subchan-
nels
We assume that the total transmitted power is constrained to P. Given the
power constraint, dierent power allocation among spatial subchannels can
aect the channel capacity tremendously. In the following context, depending
on the power allocation among spatial subchannels, several popular schemes
are presented.
2.2.1 Equal Power
If the transmitter has no knowledge about the channel, the most judicious
strategy is to allocate the power to each transmit antenna equally. In this
case, the received signals can be written as
y =

P
N
t
Hx +z (2.3)
14
The associated instantaneous channel capacity with respect to H can be
written as [1]
C
eq
=
N
t

i=1
log
_
1 +
P
N
t

2
z

2
i
_
(2.4)
In the following gure, numerical results of ergodic (average) channel
capacity for 2 2, 3 3 and 4 4 Rayleigh at fading channels are shown
in Figure 2.2.
0 5 10 15 20 25
0
5
10
15
20
25
SNR (dB)
C
a
p
a
c
i
t
y

(
b
i
t
/
s
/
H
z
)
Equal power


2x2
3x3
4x4
5x5
Figure 2.2: Capacity for equal power allocation.
15
2.2.2 Water-Filling
If the transmitter has full knowledge about the channel, the most judicious
strategy is to allocate the power to each spatial subchannel by water-lling
principle [1]. It allocates more power when a spatial subchannel has larger
gain (i.e. {
i
}) and less when a subchannel gets worse. With V at the
transmitter and U at the receiver, the original MIMO channel is converted to
r uncoupled parallel SISO channels. The WF scheme is shown in Figure 2.3.
Constellation
Mapper
Ant-1
Ant-Nt
Ant-1
Ant-Nr
~
y
Binary
Info.
source Binary
Info.
Out
V U
Channel
Estimation
Feedback
P
1
x
1
S/P
D
e
t
e
c
t
o
r
P
o
w
e
r

A
l
l
o
c
a
t
i
o
n
Nt-r 0's
H
P
r
x
r
Figure 2.3: Water-lling scheme.
For spatial subchannel i, i = 1, 2, . . . , r, the received signal is
y
i
=
_
P
i

i
x
i
+ z
i
(2.5)
where
r

i=1
P
i
= P as a constraint and z
i
the is AWGN variable with zero mean
16
and
2
z
variance. Following the method of Lagrange multipliers, the optimal
P
i
can be found as [1]
P
i
= max
_
1
Lln 2


2
z

i
, 0
_
, i = 1, 2, . . . , r (2.6)
where L is the Lagrange multiplier. The instantaneous channel capacity is
for this spatial subchannel is
C
wf,i
= log
_
1 +
P
i

2
z

2
i
_
(2.7)
Then the total channel capacity with respect to H is
C
wf
=
r

i=1
C
wf,i
(2.8)
The WF scheme maximizes the channel capacity by power allocation over
spatial subchannels. Since
1

2
. . .
r
, we have P
1
P
2
. . . P
r
and
C
wf,1
C
wf,2
. . . C
wf,r
. In practice, if each spatial subchannel requires
the same error rate performance, the spatial subchannel with larger gain
(i.e.

P
i

i
) can have higher rate; while if each spatial subchannel has the
same rate, the spatial with larger gain will have better performance. The
application of these two approaches will depend on the type of service. For
example, the important data needs high quality but the voice can tolerate
low quality.
In the following gure, numerical results of ergodic (average) channel
capacity for 2 2, 3 3 and 4 4 Rayleigh at fading channels are shown
in Figure 2.4.
In the WF scheme, the feedback bandwidth for the perfect CSI grows
with respect to the number of transmit and receive antennas. In practice,
the bandwidth for CSI feedback is often very limited. The large feedback
bandwidth also aects the transmission eciency.
17
0 5 10 15 20 25
0
5
10
15
20
25
SNR (dB)
C
a
p
a
c
i
t
y

(
b
i
t
/
s
/
H
z
)
Waterfilling


2x2
3x3
4x4
5x5
Figure 2.4: Capacity for water-lling.
2.2.3 Eigen-Beamforming
To save feedback bandwidth, beamforming can be considered. Beamform-
ing is a linear signal processing technique that control the complex weights
of the transmit and receive antennas jointly to optimize the signal-to-noise
ratio (SNR) in one direction [7] [10]. For the MIMO model, the optimal
beamforming is called eigen-beamforming. In the following context, this
optimal solution is referred to as beamforming.
In this scheme, the eigenvector associated with the maximum singular
value
1
from the transmitter side, i.e., v
1
, is feedback to the transmitter as
18
the transmission beamformer. The Eigen-beamforming scheme is shown in
Figure 2.5.
Constellation
Mapper
Detector
Ant-1
Ant-Nt
Ant-1
Ant-Nr
~
y
1
Binary
Info.
source
Binary
Info.
Out
v
1
u
1
Channel
Estimation
Feedback
H
Px
1
Figure 2.5: Eigen-beamforming scheme.
We assume one symbol, saying x
1
, is transmitted. At the receiver, the
received vector can be written as
y
1
=

PHv
1
x
1
+z
1
(2.9)
where z
1
is the additive white Gaussian noise vector with i.i.d. symmetrical
complex Gaussian elements of zero mean and variance
2
z
.
The eigenvector associated with the maximum singular value from the
19
receiver side, i.e., u
1
, is applied as the receiver beamformer. Then we have
y
1
= u
H
1
y
1
=

P
1
x
1
+ z
1
(2.10)
where z
1
is Gaussian with zero mean and variance
2
z
. As can be seen from
(2.10), only the spatial subchannel associated with the maximum singular
value
1
is applied.
The associated instantaneous channel capacity with respect to H can be
written as
C
bf
= log
_
1 +
P

2
z

2
1
_
(2.11)
In the following gure, numerical results of ergodic (average) channel capacity
for 22, 33 and 44 Rayleigh at fading channels are shown in Figure 2.6.
The eigen-beamforming scheme can save feedback bandwidth and is op-
timized in terms of SNR [10]. However, since only one spatial subchannel is
considered, this scheme suers from loss of channel capacity [17], especially
when the number of antennas grows. Also, as can be seen from the gure,
the gap of capacity between dierent numbers of antennas is nearly the same
for all the SNR region.
20
0 5 10 15 20 25
0
1
2
3
4
5
6
7
8
9
10
SNR (dB)
C
a
p
a
c
i
t
y

(
b
i
t
/
s
/
H
z
)
Beamforming


2x2
3x3
4x4
5x5
Figure 2.6: Capacity for eigen-beamforming.
21
Chapter 3
Recent Research Results
In this chapter, the current achievements will be presented as follows.
3.1 Beam-Nulling
Inspired by the eigen-beamforming scheme, we will propose a new beamforming-
like scheme called beam-nulling (BN). This scheme uses the same feedback
bandwidth as beamforming. That is, only one eigenvector is fed back to the
transmitter. Unlike the eigen-beamforming scheme in which only the best
spatial subchannel is considered, in the beam-nulling scheme, only the worst
spatial subchannel is discarded. Hence, the loss of channel capacity as com-
pared to the optimal water-lling scheme can be reduced. The Beam-Nulling
scheme is shown in Figure 3.1.
In this scheme, the eigenvector associated with the minimum singular
value from the transmitter side, i.e., v
N
t
, is feedback to the transmitter.
By a certain rule, a subspace orthogonal to the weakest spatial channel is
22
Constellation
Mapper
Ant-1
Ant-Nt
Ant-1
Ant-Nr
~
y
Binary
Info.
source
Binary
Info.
Out
U
Channel
Estimation
Feedback
x
1
S/P
x
Nt-1
D
e
t
e
c
t
o
r
H
Figure 3.1: Beam-Nulling scheme.
23
constructed. That is, the following condition should be satised.

H
v
N
t
= 0 (3.1)
The N
t
(N
t
1) matrix = [g
1
g
2
. . . g
Nt1
] spans the subspace. Note that
the rule to construct the subspace should also be known to the receiver.
An example to construct the orthogonal subspace is presented as follows.
We construct an N
t
N
t
matrix
A = [v
N
t
I

] (3.2)
where I

= [I
(N
t
1)(N
t
1)
0
(N
t
1)1
]
T
. Applying QR decomposition to A, we
have
A = [v
N
t
] R (3.3)
where R is an upper triangular matrix with the (1,1)-th entry equal to 1.
is the subspace orthogonal to v
Nt
.
At the transmitter, N
t
1 symbols denoted as x

are transmitted only

over the orthogonal subspace . The received signals at the receiver can be
written as
y

P
N
t
1
Hx

+z

(3.4)
where z

is additive white Gaussian noise vector with i.i.d. symmetrical

complex Gaussian elements of zero mean and variance
2
z
.
Substituting (2.2) into (3.4) and multiplying y

by U
H
, we have

y =

P
N
t
1

_
_
_
B
0
T
_
_
_x

z (3.5)
where

z is additive white Gaussian noise vector with i.i.d. symmetrical com-
plex Gaussian elements of zero mean and variance
2
z
. With the condition in
24
(3.1),
V
H
=
_
_
_
B
0
T
_
_
_ (3.6)
where
B =
_
_
_
_
_
_
_
_
_
_
v
H
1
g
1
v
H
1
g
2
. . . v
H
1
g
Nt1
v
H
2
g
1
.
.
.
. . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
v
H
Nt1
g
1
. . . . . . v
H
Nt1
g
Nt1
_
_
_
_
_
_
_
_
_
_
(3.7)
B is an (N
t
1) (N
t
1) unitary matrix. As can be seen from (3.5), the
available spatial channels are N
t
1. Since the weakest spatial subchannel is
nulled in this scheme, power can be allocated equally among the left better
N
t
1 subchannels. Equation (3.5) can be rewritten as

P
N
t
1

Bx

(3.8)
where

y

and

z

are column vectors with the rst (N

r
1) elements of

y and

z, respectively, and

= diag[
1
,
2
, . . . ,
(Nt1)
]. From (3.8), the associated
instantaneous channel capacity with respect to H can be found as
C
bn
=
N
t
1

i=1
log
_
1 +
P
(N
t
1)
2
z

2
i
_
(3.9)
In the following gure, numerical results of ergodic (average) channel capacity
for 22, 33 and 44 Rayleigh at fading channels are shown in Figure 3.2.
25
0 5 10 15 20 25
0
5
10
15
20
25
SNR (dB)
C
a
p
a
c
i
t
y

(
b
i
t
/
s
/
H
z
)
Beamnulling


2x2
3x3
4x4
5x5
Figure 3.2: Capacity for beam-nulling.
As can be seen, the beam-nulling scheme only needs one eigenvector to
be fed back. However, since only the worst spatial subchannel is discarded,
this scheme can increase channel capacity signicantly as compared to the
conventional beamforming scheme.
3.2 Comparisons Among the Four Schemes
In this section, we compare the new proposed beam-nulling scheme with the
other schemes. Water-lling is the optimal solution among the four schemes
26
for any SNR.
Let

C
eq
,

C
bf
, and

C
bn
denote the ergodic channel capacities of equal
power, beamforming and beam-nulling, respectively. That is,

C
eq
= E(C
eq
) (3.10)

C
bf
= E(C
bf
) (3.11)

C
bn
= E(C
bn
) (3.12)
where E() denotes expectation with respect to H. Let = P/
2
z
denote SNR.
Dierentiating the above ergodic capacities with respect to respectively, we
have

C
eq

= E
_
_
N
t

i=1
1
+
N
t

2
i
_
_
(3.13)

C
bf

= E
_
_
1
+
1

2
1
_
_
(3.14)

C
bn

= E
_
_
Nt1

i=1
1
+
Nt1

2
i
_
_
(3.15)
The dierential will also be referred to as slope. The second order dier-
entials are listed as follows.

2

C
eq

2
= E
_
_
_
_
_
N
t

i=1

1
_
+
N
t

2
i
_
2
_
_
_
_
_
(3.16)

2

C
bf

2
= E
_
_
_
1
_
+
1

2
1
_
2
_
_
_ (3.17)
27

2

C
bn

2
= E
_
_
_
_
_
N
t
1

i=1

1
_
+
N
t
1

2
i
_
2
_
_
_
_
_
(3.18)
As can be seen, the above ergodic capacities are concave and monotonically
increasing with respect to . With the fact that
1

2
. . .
N
t
, it can
be readily checked that the slopes of ergodic capacities associate with equal
power and beam-nulling are bounded as follows.
E
_
_
N
t
+
Nt

1
_
_

C
eq

E
_
_
N
t
+
Nt

N
t
_
_
(3.19)
E
_
_
N
t
1
+
N
t
1

1
_
_

C
bn

E
_
_
N
t
1
+
N
t
1

(N
t
1)
_
_
(3.20)
For the case of N
t
= 2, beamforming and beam-nulling have the same
capacity for any SNR as can be seen from equations of capacity and slope.
If 0, i.e., at low SNR, it can be easily found that

C
bf

C
bn

C
eq

, 0 (3.21)
If , i.e., at high SNR, it can be easily found that

C
eq

C
bn

C
bf

, (3.22)
Hence, at medium SNR,


C
bn

has the largest value. That is, among equal

power, beamforming and beam-nulling, for low SNR beamforming has largest
capacity, for high SNR equal power has largest capacity, for medium SNR,
beam-nulling has largest capacity. Numerical results will be provided to
demonstrate our analysis.
28
0 5 10 15 20 25
0
2
4
6
8
10
12
SNR (dB)
C
a
p
a
c
i
t
y

(
b
i
t
/
s
/
H
z
)
2x2


EQ
WF
BF
BN
Figure 3.3: 2 2 Rayleigh fading channel.
From Figure 3.3 to 3.6, capacities of water-lling, beamforming, beam-
nulling and equal power are compared over 22, 33, 44 and 55 Rayleigh
fading channels, respectively. Note that since SNR is measured in dB, the
curves become convex. In these gures, EQ stands for equal power, WF
stands for water-lling, BF stands for beamforming and BN stands for
beam-nulling. In the gures with subgures, subgure (a) is for all region
SNR, subgure (b) is detailed for SNR region from low to medium, and
subgure (c) is detailed for SNR region from medium to high. Note that for
2 2 channel, beamforming and beam-nulling have the same capacity.
29
0 5 10 15 20 25
0
2
4
6
8
10
12
14
16
SNR (dB)
C
a
p
a
c
i
t
y

(
b
i
t
/
s
/
H
z
)
3x3


EQ
WF
BF
BN
(a) 0 25 dB
0 2 4 6 8 10
0
1
2
3
4
5
6
SNR (dB)
C
a
p
a
c
it
y

(
b
it
/
s
/
H
z
)
3x3


EQ
WF
BF
BN
(b) 0 10 dB
12 14 16 18 20 22
6
7
8
9
10
11
12
13
14
15
16
SNR (dB)
C
a
p
a
c
it
y

(
b
it
/
s
/
H
z
)
3x3


EQ
WF
BF
BN
(c) 12 22 dB
Figure 3.4: 3 3 Rayleigh fading channel.
30
0 5 10 15 20 25
0
2
4
6
8
10
12
14
16
18
20
22
SNR (dB)
C
a
p
a
c
i
t
y

(
b
i
t
/
s
/
H
z
)
4x4


EQ
WF
BF
BN
(a) 0 25 dB
0 1 2 3 4 5 6 7 8
0
1
2
3
4
5
6
SNR (dB)
C
a
p
a
c
it
y

(
b
it
/
s
/
H
z
)
4x4


EQ
WF
BF
BN
(b) 0 8 dB
16 17 18 19 20 21 22 23 24
10
11
12
13
14
15
16
17
18
19
20
SNR (dB)
C
a
p
a
c
it
y

(
b
it
/
s
/
H
z
)
4x4


EQ
WF
BF
BN
(c) 16 24 dB
Figure 3.5: 4 4 Rayleigh fading channel.
31
0 5 10 15 20 25
0
5
10
15
20
25
SNR (dB)
C
a
p
a
c
i
t
y

(
b
i
t
/
s
/
H
z
)
5x5


EQ
WF
BF
BN
(a) 0 25 dB
0 1 2 3 4 5 6 7 8
0
1
2
3
4
5
6
7
SNR (dB)
C
a
p
a
c
it
y

(
b
it
/
s
/
H
z
)
5x5


EQ
WF
BF
BN
(b) 0 8 dB
22 22.5 23 23.5 24 24.5 25 25.5 26
20
20.5
21
21.5
22
22.5
23
23.5
24
24.5
25
SNR (dB)
C
a
p
a
c
it
y

(
b
it
/
s
/
H
z
)
5x5


EQ
WF
BF
BN
(c) 22 26 dB
Figure 3.6: 5 5 Rayleigh fading channel.
32
As can be seen from all these gures, the water-lling has the best capac-
ity at any SNR region. The other schemes perform dierently at dierent
SNR regions as discussed in the following context.
3.2.1 At low SNR
At low SNR, the beamforming is the closest to the optimal water-lling, e.g.,
the SNR region below 3 dB for 3 3 fading channel, the SNR region below
3.2 dB for 4 4 fading channel, and the SNR region below 3.5 dB for 5 5
fading channel. At low SNR, the water-lling scheme can only allocate power
to one or two spatial subchannels. Especially, if only one spatial subchannel
can be use, the beamforming scheme is just the optimal water-lling, which
can be seen from the capacity of 2 2 fading channel in Figure 3.3.
3.2.2 At medium SNR
At medium SNR, the proposed beam-nulling is the closest to the optimal
water-lling, e.g., the SNR region from 3 dB to 16 dB for 33 fading channel,
the SNR region from 3.2 dB to 20.5 dB for 44 fading channel, and the SNR
region from 3.5 dB to 23.5 dB for 5 5 fading channel. The beam-nulling
scheme only discards the weakest spatial subchannel and allocates power to
the other spatial subchannels. As can be seen from the numerical results, the
beam-nulling scheme performs better than the other schemes in this case.
33
3.2.3 At high SNR
At high SNR, the equal power scheme is the closest to the optimal water-
lling, e.g., the SNR region over 16 dB for 3 3 fading channel, the SNR
region over 20.5 dB for 4 4 fading channel, and the SNR region over 23.5
dB for 5 5 fading channel. As can be seen from the gures, at high SNR,
the equal power scheme will converge to the water-lling scheme.
In summary, the application of the above four schemes shall depend on the
SNR region and the availability of CSI. At medium SNR, the proposed beam-
nulling scheme can achieve larger capacity than the beamforming scheme with
the same feedback bandwidth.
34
Chapter 4
Conclusions and Future Works
In this chapter, we will conclude the study and present our remaining tasks
together with their schedule.
4.1 Conclusions
Via singular-value decomposition, the original MIMO channel is converted
to uncorrelated spatial subchannels. Based on the concept of spatial sub-
channels, we studied various power allocation strategies for various channel
state information scenarios, such as equal power, water-lling, beamforming.
Inspired by the beamforming scheme, we proposed a novel scheme called
beam-nulling. Using the same feedback bandwidth as beamforming, the
new scheme exploits all spatial subchannels except the weakest one and thus
achieves signicant high capacity near the optimal water-lling scheme at
medium signal-to-noise ratio. Additionally, the capacities of equal power,
beamforming and beam-nulling were compared through theoretical analysis
35
rst and then numerical results of these three schemes are also compared
with the optimal water-lling scheme. The comparison showed that at low
signal-to-noise ratio, beamforming is the closest to the optimal water-lling,
at medium signal-to-noise ratio, beam-nulling is the closest to the optimal
solution, and at high signal-to-noise ratio, equal power is the closest to the
optimal solution.
4.2 Future Works
In the future, based on our current achievement, the remaining two works
are identied and will be presented together with their schedule as follows.
4.2.1 Performance and Application of Beam-Nulling
Currently, the capacity of beam-nulling was studied in the seminar. Based on
this knowledge, we will study how to achieve the promised capacity of beam-
nulling in practice. Additionally, we will study how to cooperate with the
other schemes and thus improve the performance further. For example, the
proposed scheme can concatenate with the linear dispersion code to achieve
better performance with more exibility.
4.2.2 Extended Adaptive Frameworks
The power allocation strategy of beam-nulling and beamforming can be ex-
tended to improve capacity further if more feedback bandwidth is available.
That is, if more than one eigenvector, e.g. k eigenvectors, can be avail-
able at the transmitter, the existing beamforming scheme and the proposed
36
beam-nulling scheme can be further extended, respectively. The extended
schemes will implement or discard k spatial subchannels and are referred to
as multi-dimensional (MD) beamforming and multi-dimensional beam-
nulling, respectively. The theoretical analysis and numeric results in terms
of capacity will be provided to evaluate the new extended schemes. We will
study the performance of the extended scheme. Similarly, we will study
how to cooperate with the other schemes and thus improve the performance
further.
4.3 List of Publications
Conferences
[1] Mabruk Gheryani, Z. Wu and Y. Shayan, SINR Analysis for Full-Rate
Linear Dispersion Code Using Linear MMSE, the 4th IEEE International
Symposium on Wireless Communication Systems, Oct. 2007.
[2] Mabruk Gheryani, Z. Wu and Y. Shayan, Design of Adaptive MIMO
System Using Linear Dispersion Code, Submitted to IEEE International
Conference on Communications (ICC2008), 2008.
[3] Mabruk Gheryani, Y. Shayan, X. Wang and Z. Wu, Error Perfor-
mance of Linear Dispersion Codes, Submitted to IEEE International Con-
ference on Communications (ICC2008), 2008.
[4] Mabruk Gheryani, Z. Wu and Y. Shayan, Power Allocation Strategy
for MIMO System Based on Beam Nulling, Submitted to IEEE Interna-
tional Conference on Communications (ICC2008), 2008.
37
Journals
[1] Mabruk Gheryani, Z. Wu and Y. Shayan, Design of Adaptive MIMO
System Using Linear Dispersion Code, A transaction paper has been sub-
mitted to IEEE Trans. on Wireless Communications.
[2] Mabruk Gheryani, Y. Shayan, X. Wang and Z. Wu, Error Perfor-
mance of Linear Dispersion Codes, A transaction letter has been submitted
to IEEE Trans. on Wireless Communications.
4.4 Schedule
The remaining tasks are scheduled in Table 4.1 and the associated Gantt
chart is also shown in Figure. 4.1.
Table 4.1: Schedule for the remaining tasks.
ID Task Name Schedule
1. Performance and Application of Beam-Nulling 2007/6 - 2007/11
2. Extended Adaptive Frameworks 2007/7 - 2007/12
3. Wrap-ups 2008/1 - 2008/3
4. Thesis Writing 2008/3 - 2008/5
38
ID Task Name
2007
Q1
1
Performance and Application of
Beam-Nulling
2 Extended Adaptive Frameworks
2008
Q2 Q3 Q4 Q1 Q2
3
4 Thesis Writing
Wrap-ups
Figure 4.1: Gantt chart for the remaining tasks.
39
Bibliography
[1] I. E. Telatar, Capacity of multi-antenna Gaussian channels, Eur. Trans.
Telecom., vol 10, pp. 585-595, Nov. 1999.
[2] G. J. Foschini, M. J. Gans, On limits of wireless communications in
a fading environment when using multiple antennas, Wireless Personal
Communications, vol. 6, no. 3, pp. 311-335, 1998.
[3] J. K. Cavers, Variable-rate transmission for Rayleigh fading channels,
IEEE Transactions on Communications, COM-20, pp.15-22, 1972.
[4] A. J. Goldsmith and S.-G. Chua, Variable rate variable power MQAM
for fading channels, IEEE Trans. Commun., vol. 45, no. 10, pp.
12181230, Oct. 1997.
[5] C. E. Shannon, A mathematical theory of communication, Bell Syst.
Tech. J., vol. 27, pp. 379423 (Part one), pp. 623656 (Part two), Oct.
1948, reprinted in book form, University of Illinois Press, Urbana, 1949.
[6] S. Zhou,and G. B. Giannakis, Adaptive modulation for multiantenna
transmissions with channel mean feedback, IEEE Trans. Wireless
Comm., vol.3, no.5, pp. 1626-1636, Sep. 2004.
[7] S. Zhou and G. B. Giannakis, Optimal transmitter eigen-beamforming
and space-time block coding based on channel mean feedback IEEE
Transactions on Signal Processing, vol. 50, no. 10, October 2002.
[8] P. Xia and G. B. Giannakis, Multiantenna adaptive modulation with
beamforming based on bandwidth-constrained feedback, IEEE Trans-
actions on Communications, vol.53, no.3, March 2005.
40
[9] B. Mondal and R. W. Heath, Jr., Performance analysis of quantized
beamforming MIMO systems, IEEE Transactions on Signal Processing
, vol. 54, no. 12, Dec. 2006.
[10] J. K. Cavers, Single-user and multiuser adaptive maximal ratio trans-
mission for Rayleigh channels, IEEE Trans. Veh. Technol., vol. 49, no.
6, pp. 20432050, Nov. 2000.
[11] Z. Luo, H. Gao,and Y. Liu,J. Gao Capacity Limits of Time-Varying
MIMO Channels, IEEE International Conference On Communications
vol.2, Mar.2003.
[12] Z. Shen, R. W. Heath, Jr., J. G. Andrews, and B. L. Evans, Comparison
of Space-Time Water-lling and Spatial Water-lling for MIMO Fading
Channels, in Proc. IEEE Int Global Communications Conf. vol. 1, pp.
431 435, Nov. 29-Dec. 3, 2004, Dallas, TX, USA.
[13] Z. Zhou and B. Vucetic Design of adaptive modulation using imperfect
CSI in MIMO systems, 2004 Eelectronics Letters vol. 40 no. 17, Aug.
2004.
[14] X. Zhang and B. Ottersten, Power allocation and bit loading for spatial
multiplexing in MIMO systems, IEEE Int. Conf.on Acoustics, Speech,
and Signal Processing, 2003. Proceedings (ICASSP 03) vol.5 pp. 54-56,
Apr. 2003.
[15] S. Zhou and A. Yener, MIMO-CDMA Systems: Signature and Beam-
former Design With Various Levels of Feedback, IEEE Transations on
Signal Proceeding , VOL.54, NO.7, JULY 2006
[16] S. Serbetli and B. Ottersten, Power allocation and bit loading for
spatial multiplexing in MIMO systems, IEEE Int. Conf.on Acoustics,
Speech, and Signal Processing, 2003. Proceedings (ICASSP 03) vol.5 pp.
54-56, Apr. 2003.
[17] S. Zhou and G. B. Giannakis, How accurate channel prediction
needs to be for transmit-beamforming with adaptive modulation over
41
Rayleigh MIMO channels, IEEE Trans. Wireless Comm., vol.3, no.4,
pp. 12851294, July 2004.
42