ChoAnd DAS Twireless PDF

SUBMISSION TO IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS 1
Downlink Performance and Capacity of Distributed

Antenna Systems in a Multicell Environment
Wan Choi and Jeffrey G. Andrews
Abstract
Distributed antenna systems (DAS) have been widely implemented in state-of-the art cellular communication
systems to cover dead spots. Recent academic studies have shown that in addition to coverage improvements, DAS
can also allow a capacity improvement. This paper analytically quanties the performance of downlink multicell
DAS in terms of power efciency, SINR improvement, and outage probability for two different transmission
strategies: selection diversity (where just one or two of the distributed antennas are used) and blanket transmission
(where all antennas in the cell broadcast data). Simple repeaters are a special case of our analysis. A generalized
information theoretic analysis is also provided to illuminate the fundamental limits of such systems. The results show
that DAS reduces other-cell interference in a multicell environment and hence signicantly improves performance
and capacity (by about 2-3x), with particularly large improvements for users near cell boundaries. Less obviously,
from a communication theory standpoint, it is shown that selection diversity is preferable to blanket transmission
even when maximum ratio transmission is used.
Index Terms
Distributed antennas systems, cellular communications, other-cell interference, capacity, diversity.
The authors are with the Wireless Networking and Communications Group, Department of Electrical and Computer Engineering, The
University of Texas at Austin, 1 University Station C0803, Austin, TX 78712, USA. Email:wchoi@mail.utexas.edu. Phone:1-512-471-0536.
This work was supported by SoLid Technologies Inc.
I. INTRODUCTION
Future wireless systems will need to provide high data rates to a large number of users. Experience
tells us that these networks will be interference-limited, since the interference from either in-cell users
(as in CDMA) or from other-cell users will determine how heavily the system can be loaded. Multiuser
receivers are one potential way to reduce interference in cellular systems (see e.g. [1] and the references
therein), but this requires a more complicated receiver. An alternative strategy is to try to reduce the overall
transmit power (and hence other-cell interference) using distributed antennas, which has the additional
merit of providing better coverage and increasing battery life [2]. Although distributed antennas systems
(DAS) were originally introduced to simply cover the dead spots in indoor wireless communications [3],
recent studies have identied other potential advantages such as power and system capacity, and expanded
its applications [4][9].
In DAS, antenna modules are geographically distributed to reduce access distance instead of centralizing
at a location. Each distributed antenna module is connected to a home base station (or central unit) via
dedicated wires, ber optics, or an exclusive RF link. Although the connections via the same RF link as
that used in a cell are possible, the connections with the same RF link construct information-theoretic
relay channels categorized as another research area called cooperative communications [10][13]. DAS
is rather similar to repeater systems [14][16] from the fact that the distributed antenna modules and the
home base station are physically connected. However, DAS is distinguished from conventional repeater
system by the fact that each distributed antenna module is able to transmit different data in the downlink
whereas repeaters just repeat signals from the home base station. Therefore, DAS is a generalization of
conventional repeater systems. Since the distributed antenna modules and the home base station together
construct a macroscopic multiple input single output (MISO) vector channel, DAS can be also interpreted
as a macroscopic multiple-antenna system [17], [18].
From an architectural point-of-view, DAS has manifest advantages over conventional communication
systems. DAS can reduce the cost of installing system and simplify maintenance because DAS can reduce
the required number of base stations within a target service area. Furthermore, blocking probability can
be improved owing to the principle of trunking efciency because resources for signal processing such
as channel cards/elements are centralized and shared at the home base station (or central unit) [19]. In
addition to these architectural advantages, DAS also has been shown to possess advantages in terms
of power, signal-to-interference-plus-noise ratio (SINR), and capacity owing to macro-diversity and the
reduced access distance [4][8]. Based on these advantages, many cellular service providers or system
manufacturers are seriously considering replacing legacy cellular systems with distributed antenna systems
or adopting the distributed antenna architecture in the future. However, most of the recent work on DAS
has been focused on investigating those advantages and analyzing its performance in the uplink because
of its analytical simplicity [4][8]. On the other hand, there are few studies on the downlink performance
of DAS although the demand for high speed data will be dominant in the downlink. There are also few
papers that consider the advantages of DAS in a multicell context.
A recent paper [9] addressed downlink performance of code division multiple access (CDMA) DAS
in a multiple cell environment but it relied on computer simulations and only investigated SIR levels
perceived at mobile stations. Computer simulations are useful but mathematical analysis enables us to
efciently identify the key design parameters and to understand their effects. Particularly, information-
theoretic analysis provides intuition on achievable system capacity. For this reason, the authors of [17],
[18] analyzed downlink capacity in terms of multiple-antenna (MIMO) information theory. However, they
did not consider a multicell environment, either. In addition, their analysis was limited to the case with
specic assumptions: (1) channel state information is known to the transmitter and to the receiver, and (2)
transmit power is shared among distributed antenna modules, i.e., there is no per antenna-module power
constraint.
In this context, we analyze the realistic potential gains of downlink DAS in a multicell environment. We
rst investigate the downlink advantages of DAS in terms of power efciency, SINR improvement, and
outage probability for two different transmission strategies: selection diversity (where just one or two of the
distributed antennas are used) and blanket transmission (where all antennas in the cell broadcast the data).
Simple repeaters are a special case of our analysis. We then develop a generalized analysis of achievable
capacity and transmit diversity performance from an information-theory standpoint. The result of this
paper shows that a distributed antenna architecture effectively addresses other-cell interference (OCI) in
a multicell environment, especially at the cell boundaries, and achieves a non-trivial capacity increase
and performance improvement over a conventional cellular architecture. An additional contribution of
this paper is a theoretical framework for analyzing the achievable capacity and diversity performance of
cellular repeater systems, a special case of DAS.
The rest of this paper is organized as follows. Section II provides a system model for analysis, and
describes the transmission and multiple access strategies. In section III, the architectural advantages of
downlink DAS in a multicell environment are investigated. Transmit diversity performance of downlink
DAS in a multicell environment is analyzed in section IV. In section V, an information-theoretic capacity
analysis for multicell DAS is developed, and compared to the previous results which were based on
communications theory. Finally, conclusions are drawn in section VI.
II. SYSTEM MODEL
A. Cellular Architecture
In DAS, the main processing modules such as channel cards are centralized at a location (central
unit) and are connected with distributed antenna modules. Each distributed antenna module is physically
connected with a home base station via dedicated wires, ber optics, or an exclusive RF link. As
previously mentioned, the dedicated connections do not construct the information-theoretic relay channel
and thus differentiate DAS from cooperative communications. A general architecture of DAS in a multicell
environment is given in Figure 1, where a cell is covered by a small base station and six distributed antenna
modules. In contrast, the same area is covered by only a single high-power base station in traditional
cellular systems. Alternatively, the small base station and 6 distributed antenna modules can be viewed
as an alternative to 7 small traditional base stations (pico/micro cells). The actual number of distributed
antenna modules would be determined by coverage, user densities, and other environmental factors but
we only consider 6 distributed antenna modules as a reasonable example. The total transmit power of the
ith distributed antenna module of the jth cell is P
(j)
i
, where the small base station of each cell and home
cell are indexed by i = 0 and j = 0, respectively. Throughout this paper, we assume that

6
i=0
P
(j)
i
= P
where P is the total transmit power of the conventional base station for a fair comparison. In Figure 1,
polar (r, ) coordinates are employed and the location of the base station 0 corresponds to the origin of the
coordinates. The radius of a cell (a bold dotted circle) is R and the radius of a coverage of a distributed
antenna module is r. We also consider the 1-tier cellular structure with universal frequency reuse, where
a given cell is surrounded by one continuous tier of six cells. Although this assumption of only 1-tier of
interfering cells is optimistic, a pessimistic assumption that all the distributed antenna modules and the
base stations are transmitting full power all the time easily compensates.
B. Transmission Strategy and Multiple Access Scenario
In cellular DAS, there are several possible transmission strategies using multiple distributed antenna
modules. Either all or some of the antenna modules can be used in transmission. Although many methods
of using the distributed antenna modules are possible, we consider three likely transmission strategies: the
blanket transmission scheme, single transmit selection scheme, and dual transmit selection scheme. The
blanket transmission scheme is to transmit signals through all the distributed antenna modules and the
home base station. In this scheme, either the same signals or different signals are transmitted through each
distributed antenna module. Therefore, the distributed antenna modules construct a macroscopic multiple
antenna system. In the single transmit selection scheme, only a single distributed antenna module or
the home base station is selected for transmission by the criterion of minimizing propagation pathloss.
Although smarter selection algorithms can potentially be used such as maximizing SINR or capacity, we
consider this simple one for simplicity of analysis and since it should minimize the required transmit
power (and hence the interference caused to other cells). This scheme exploits macroscopic selection
diversity and is expected to additionally reduce OCI because the number of OCI sources is reduced.
Finally, in the dual transmit site selection, signals are transmitted through the home base station and a
distributed antenna module selected by the minimum propagation pathloss criterion. This scheme also
exploits macroscopic selection diversity.
We also assume a single user scenario for simplicity of analysis, but this assumption holds for most
practical multiuser systems as well because time division multiple access (TDMA), frequency division
multiple access (FDMA), orthogonal code division multiple access (CDMA), and orthogonal frequency
division multiple access (OFDMA) systems basically correspond to this assumption since there is only a
single user transmitting in any time/frequency/code dimension.
C. Channel Model and Received Signal
When the blanket transmission scheme is used, the multiple distributed antenna modules and the home
base station together construct a macroscopic MISO vector channel given by
h = [
_
L
(0)
0
h
(0)
0
_
L
(0)
1
h
(0)
1

_
L
(0)
6
h
(0)
6
] (1)
where h
(j)
i
denotes short term fading from the ith distributed antenna module in the jth cell and is an
i.i.d. complex Gaussian random variable CN(0, 1). The fading channel is assumed to be static during
a symbol duration. L
(j)
i
denotes propagation pathloss and shadow fading from the ith distributed antenna
module in the jth cell. Let the transmitted signal vector be x = [x
(0)
0
x
(0)
1
x
(0)
6
]
T
, where x
(j)
i
is the
transmitted signal from the ith distributed antenna module in the jth cell with E[|x
(j)
i
|
2
] = P
(j)
i
. Then,
the received signal at a mobile station at a given symbol duration is given by
y = hx +
6
j=1
6
i=0
h
(j)
i
_
L
(j)
i
x
(j)
i
+ n (2)
= hx + z (3)
where n is the additive Gaussian noise with variance E[nn
H
] =
2
n
. Since the number of interfering source
is sufciently large and interfering sources are independent of each other, the interference plus noise is
assumed to be a complex Gaussian random variable z with variance
2
z
by the Central Limit Theorem
(CLT).
When the single transmit selection scheme is used, the received signal is given by
y =
_
L
(0)
m
h
(0)
m
x
(0)
m
+
6
j=1
h
(j)
m
_
L
(j)
m
x
(j)
m
+ n (4)
=
_
L
(0)
m
h
(0)
m
x
(0)
m
+ z
1
(5)
where m = arg max
i{0,1, ,6}
{L
(j)
0
, L
(j)
1
, , L
(j)
6
} and z
1
is a complex Gaussian random variable with
variance
2
z
1
, representing interference plus noise.
When the dual transmit selection scheme is used, the received signal is given by
y =
_
L
(0)
0
h
(0)
0
x
(0)
0
+
_
L
(0)
m
h
(0)
m
x
(0)
m
+
6
j=1
h
(j)
0
_
L
(j)
0
x
(j)
0
+
6
j=1
h
(j)
m
_
L
(j)
m
x
(j)
m
+ n (6)
=
_
L
(0)
0
h
(0)
0
x
(0)
0
+
_
L
(0)
m
h
(0)
m
x
(0)
m
+ z
2
(7)
where m = arg max
i{1, ,6}
{L
(j)
1
, , L
(j)
6
} and z
2
is a complex Gaussian random variable with variance
2
z
2
, representing interference plus noise.
III. ARCHITECTURAL ADVANTAGES OF CELLULAR DISTRIBUTED ANTENNA SYSTEMS
In this section, we investigate basic architectural advantages of the cellular DAS over conventional
cellular systems in terms of basic performance measures such as power efciency, signal-to-interference-
plus-noise ratio (SINR), and outage probability.
A. Power efciency
Power efciency is computed by comparing the required power for supporting the coverage area of a
single cell, and is calculated based on propagation pathloss. In the distributed antenna architecture given
in Figure 1, the area covered by a small base station and 6 distributed antenna modules is (3+6
3)r
2
(=
7r
2
24(r
2
/3
3/2r
2
)). Then, the radius of a circle with the same area becomes r
_
12
3/ 1.
For a fair comparison, we regard this circle as the effective coverage of a high-power base station in a
traditional cellular system instead of the bold dotted circle in Figure 1. Let the required transmit power for
each distributed antenna module and the small base station to support the circular area with radius r be
P. Then, the total required power to support the effective coverage area is 7P in the distributed antenna
structure, whereas that for the big base station with the same effective coverage area in the traditional
cellular structure becomes (12
3/ 1)
2
P if the propagation pathloss is assumed to be L = d
where
is the pathloss exponent. Then, the power efciency of the distributed antenna structure is given by
=
(12
3/ 1)
2
P
7P
=
(12
3/ 1)
2
7
(8)
For the case that the path loss exponent is 4, the power efciency gain is about 4.51(= 6.54dB). This
result shows that a distributed antenna structure requires much less power to support the same coverage
area.
B. Signal-to-interference-plus-noise ratio (SINR)
In this subsection, we show how the distributed antenna structure can reduce OCI and improve SINR in
downlink by increasing the received strength of the desired signal and reducing the power of the received
interference. Since the outage-limiting scenario in cellular networks is usually a mobile station at the
cell boundary, maintaining the SINR of the users near cell boundaries above a given level is particularly
important [20].
We assume that the same uncoded data is transmitted through multiple antenna modules. For a fair
comparison with the traditional cellular structure, we also assume that the transmit power of each dis-
tributed antenna module is 0.1P and the transmit power of the home base station is 0.4P, for a total
transmit power of P in both the distributed antenna structure and the traditional cell structure.
First, we consider the blanket transmission scheme in cellular DAS. For a given location of the target
mobile station, the expected SINR over short term fading is given by
E[] =
6
i=0
L
(0)
i
P
(0)
i
6
j=1
6
i=0
L
(j)
i
P
(j)
i
+
2
n
(9)
The expected SINR of the other transmission schemes in cellular DAS can be easily obtained similarly
to the form of (9).
Figure 2 shows the expected SINRs of different transmission schemes according to the normalized
distance from the home base station in the direction of the worst case position W in Figure 1. The SINR
of each transmission scheme is calculated when the pathloss exponent is 4.0 but lognormal shadowing is
not considered. Figure 2 demonstrates that DAS with the single transmit selection scheme outperforms a
conventional cellular system at all the distances whereas DAS with blanket transmission are worse than
conventional cellular systems only near the home base station. However, all the transmission schemes of
DAS have substantially higher SINR than conventional cellular system beyond the normalized distance
0.5. It is perhaps surprising that the macroscopic selection diversity scheme has the highest achieved SINR
of the proposed transmit schemes.
Compared to the computing SINR at a location, the overall average SINR in a cell can also be calculated
if we assume a uniform distribution of the target mobile stations, i.e.,
avg
=
_
S
E[(r, )]dS (10)
where S is the cell area covered by 6 distributed antenna modules and the home base station. Average
SINR is closely related to average system throughput, and the achievable capacity is developed in section
V. Figure 3 shows average SINR values of four different systems versus path loss exponent. The cellular
DAS with transmit site selection has the largest SINR regardless of the path loss exponent and all the
transmit schemes in DAS signicantly outperform conventional cellular systems, especially when the
pathloss is severe. Similarly to Figure 2, Figure 3 also suggests that the single transmit selection scheme
most effectively reduces OCI and improves SINR.
C. Outage Probability
If we consider lognormal shadowing, instantaneous SINR at a location for the cellular DAS with the
blanket transmission scheme is given by
inst
=
6
i=0
L
(0)
i

(0)
i
(h
(0)
i
)
2
P
(0)
i
6
j=1
6
i=0
L
(j)
i

(j)
i
(h
(j)
i
)
2
P
(j)
i
+
2
n
(11)
where
(j)
i
denotes lognormal shadowing from the ith distributed antenna module in the jth cell.
When the fading environment is a superposition of both fast and slow fading, i.e., log-normal shadowing
and Rayleigh fading, a common performance metric is combined outage probability and average error
probability where the outage occurs when the slow fading falls below some target value, i.e., target signal-
to-interference ratio averaged over the fast fading [21]. Since the background noise power is negligible
compared to the interference power, the SIR averaged over the fast fading can be given by
=
6
i=0
L
(0)
i

(0)
i
P
(0)
i
6
j=1
6
i=0
L
(j)
i

(j)
i
P
(j)
i
. (12)
Since the sum of lognormal random variables can be well approximated as a lognormal random variable
[22], the numerator in (12) can be approximated as a lognormal random variable with mean
d
and variance
2
d
as
d
=
6
i=0
L
(0)
i
P
(0)
i
exp
_
_
ln 10
10

(0)
i
+
_
ln 10
10
_
2
(
(0)
i
)
2
2
_
_
(13)
2
d
=
6
i=0
_
L
(0)
i
P
(0)
i
_
2
exp
_
_
2 ln10
10

(0)
i
+
_
ln10
10
_
2
(
(0)
i
)
2
_
_
_
exp
_
_
_
ln 10
10
_
2
(
(0)
i
)
2
_
_
1
_
(14)
and the denominator can also be approximated as a lognormal random variable with mean
I
and variance
2
I
as
I
=
6
j=1
6
i=0
L
(j)
i
P
(j)
i
exp
_
_
ln 10
10

(j)
i
+
_
ln 10
10
_
2
(
(j)
i
)
2
2
_
_
(15)
2
I
=
6
j=1
6
i=0
_
L
(j)
i
P
(j)
i
_
2
exp
_
_
2 ln10
10

(j)
i
+
_
ln 10
10
_
2
(
(j)
i
)
2
_
_
_
exp
_
_
_
ln 10
10
_
2
(
(j)
i
)
2
_
_
1
_
(16)
where
(j)
i
and
(j)
i
are the mean and the standard deviation of of
(j)
i
in dB value, respectively. Since the
ratio of lognormal random variables is also a lognormal random variable, the SIR becomes a lognormal
random variable with mean
and variance
2
in dB by

=
10 ln(
d
)
ln 10

10
2 ln10
ln
_
2
d
2
d
+ 1
_
10 ln(
I
)
ln 10
+
10
2 ln10
ln
_
2
I
2
I
+ 1
_
(17)

2
=
_
10
ln 10
_
2
ln
_
2
d
2
d
+ 1
_
+
_
10
ln 10
_
2
ln
_
2
I
2
I
+ 1
_
(18)
Then, the outage probability of the mobile station at a given location is given by
Pr[ <
req
] = 1 Q
_
10 log
10
(
req
)
_
(19)
where
req
is the required SINR to guarantee a desired QoS. Since the analysis of the blanket transmission
scheme in DAS is a general case and includes analysis of other schemes, the outage probabilities of other
transmission schemes in DAS and conventional cellular system can be obtained in the same way.
The outage probability versus the normalized distance from the home base station in the direction of the
worst case position W is shown in Figure 4 when the pathloss exponent is 4.0. We assume power control
compensates for the lognormal shadowing in the desired cell but the standard deviation of the lognormal
shadowing for the signals from neighbor cells is 6 dB. The outage threshold is assumed as 10dB. Figure 4
shows that all the transmission schemes in the cellular DAS substantially outperform conventional cellular
system.
IV. TRANSMIT DIVERSITY PERFORMANCE IN A DISTRIBUTED ANTENNA NETWORK
As previously mentioned, DAS (particularly, the blanket transmission scheme) constructs a macroscopic
MISO vector channel. When the channel is known to the transmitter, transmit-maximal ratio combining
(transmit-MRC) [23] is known to maximize transmit diversity gain in a MISO vector channel. Therefore,
we analyze symbol error probability performance of DAS with the transmit-MRC in this section, which
can be thought as an upper-bound to other schemes such as space time block coding (STBC). In transmit-
MRC, the signal is transmitted from each distributed antenna module after being weighted appropriately,
so that the signals arrive in phase at the mobile station and add coherently [23], [24]. For simplicity of
analysis, we assume that the transmit power of each distributed antenna module and the home base station
in DAS is P/7 whereas the transmit power of the base station in the traditional cellular system is P.
Then, the signal at the mobile station during a symbol period is given by
y = hwx + z (20)
where w is a 7 1 transmit weight vector and x is the transmitted data with E[|x|
2
] = P/7. The weight
vector is chosen to maximize the received SINR = |hw|
2
P/7
2
z
. By the Cauchy-Schwarz inequality, it
holds that

h
2
w
2
P
7
2
z
(21)
where is the L
2
vector norm and equality holds if and only if w is proportional to h
H
. Therefore,
the weight vector is h
H
/h and the corresponding SINR is given by
=
h
2
P
7
2
z
(22)
Then, the symbol error probability is obtained as [25]
P
e
E
_
N
e
Q
__
d
2
min
2
__
(23)
where N
e
and d
min
are the average number and minimum distance of nearest neighbors of the given signal
constellation. Since =
1
7
2
z
6
i=0
|h
(0)
i
|
2
L
(0)
i
P is a weighted chi-squared distributed random variable with
p.d.f. given by
f
() =
6
i=0
7
2
z
i
L
(0)
i
P
exp
_
7
2
z
L
(0)
i
P
_
where
i
=
6
k=0,k=i
L
(0)
i
L
(0)
i
L
(0)
k
, (24)
the symbol error probability can be obtained by [26]
P
e,blanket

_

0
N
e
Q
__
d
2
min
2
_
f
()d (25)
=
6
i=0
i
N
e
2
_
_
1
_
d
2
min
L
(0)
i
P
d
2
min
L
(0)
i
P + 28
2
z
_
_
(26)
When the single transmit selection scheme is used in the cellular DAS, the symbol error probability can be
easily obtained from (26) because the derivation of (26) for this MISO vector channel is a generalization
of a single input single out (SISO) channel. Then, the symbol error probability for the single transmit
selection scheme is given by
P
e,sel
=
N
e
2
_
_
1
_
d
2
min
L
(0)
m
P
d
2
min
L
(0)
m
P + 28
2
z1
_
_
(27)
where m = arg max
i{0,1, ,6}
{L
(j)
0
, L
(j)
1
, , L
(j)
6
}. Similarly, the symbol error probability for the con-
ventional cellular system is given by
P
e,conv
=
N
e
2
_
_
1
_
d
2
min
L
(0)
0
P
d
2
min
L
(0)
0
P + 4
2
c
_
_
(28)
where
2
c
is the interference-plus-noise power in the conventional cellular structure.
In Figure 5, symbol error probability (SER) of cellular DAS with transmit-MRC is shown versus the
normalized distance from the home base station in the direction of the worst case position W. In this
gure, QPSK modulation is used and the effects of lognormal shadowing are not considered. Figure
5 demonstrates that cellular DAS exploits macroscopic transmit diversity and thus shows substantially
better symbol error performance than the conventional cellular system beyond the normalized distance 0.4.
Although the blanket transmission scheme in DAS achieves maximum transmit diversity gain by employing
transmit-MRC, the single transmit selection scheme still outperforms since performance is limited by OCI
in interference-limited cellular networks. This somewhat surprising result is due to the effective reduction
of OCI, and hence improvement of SINR. However, as the pathloss exponent increases, the effect of OCI
is generally reduced so the SER performance of the blanket transmission scheme approaches that of the
single transmit selection scheme by virtue of the diversity gain of transmit-MRC.
V. ACHIEVABLE CAPACITY OF DISTRIBUTED ANTENNA NETWORKS
In this section, the Shannon capacity of the cellular DAS is derived in order to upper bound the
theoretically achievable system capacity. This will provide a useful contrast with the prior SINR-based
approach.
A. Ergodic Capacity only with CSIR
If we assume the channel state information is known only at the receiver (CSIR) and the channel is
ergodic, the ergodic Shannon capacity at a given location of the target mobile station can be achieved by
C
e
= E
h
_
log
2
_
1 +
1
2
z
hSh
H
__
(29)
where S is the covariance matrix of the transmitted vector x and given by diag(P
(0)
0
, P
(0)
1
, , P
(0)
6
).
If ergodicity of the channel is assumed, the ergodic capacity can be obtained as
C
e
= E
h
_
log
2
_
1 +
1
2
z
6
i=0
|h
(0)
i
|
2
L
(0)
i
P
(0)
i
__
=
_

x=0
log
2
(1 +)f
()d (30)
where =
1
2
z
6
i=0
|h
(0)
i
|
2
L
(0)
i
P
(0)
i
is a weighted chi-squared distributed random variable with p.d.f. given
by
f
() =
6
i=0
2
z
i
L
(0)
i
P
(0)
i
exp
_

2
z
L
(0)
i
P
(0)
i
_
where
i
=
6
k=0,k=i
L
(0)
i
P
(0)
i
L
(0)
i
P
(0)
i
L
(0)
k
P
(0)
k
, (31)
Then, the ergodic capacity can be obtained in a simple form by [26]
C
e
=
1
ln 2
6
i=0
i
exp
_

2
z
L
(0)
i
P
(0)
i
_
Ei
_

2
z
L
(0)
i
P
(0)
i
_
(32)
where Ei(x) is the exponential integral function (Ei(x) =
_
x
e
t
/t dt) and can be easily calculated
with popular numerical tools such as MATLAB, MATHEMATICA, and MAPLE.
Since the derivation for this MISO vector channel is a generalization of a SISO channel, the ergodic
capacities of the single transmit selection scheme and the conventional cellular system are given, respec-
tively, by
C
e,sel
=
1
ln 2
exp
_

2
z1
L
(0)
m
P
(0)
m
_
Ei
_

2
z1
L
(0)
m
P
(0)
m
_
(33)
C
e,conv
=
1
ln 2
exp
_

2
c
L
(0)
0
P
_
Ei
_

2
c
L
(0)
0
P
_
(34)
Figure 6 shows the ergodic capacity of cellular DAS versus the normalized distance from the home
base station in the direction of the worst case position W when the pathloss exponent is 4.0. Again,
the transmit power of each distributed antenna module is 0.1P and the transmit power of the home
base station is 0.4P in DAS whereas the transmit power of the base station in the conventional cellular
system is P. As expected from the SINR-based results, the single transmit selection scheme achieves
the highest throughput owing to the OCI reduction and macroscopic selection diversity. Although the
achieved throughput of the blanket transmission scheme in the cellular DAS is slightly lower than that
of conventional cellular system near the home base station due to reduced transmit power, the achieved
throughput of the blanket transmission scheme in DAS has substantially higher throughput beyond the
normalized distance 0.5. Note that 80% of the users will be in (0.5R, R) assuming a uniform distribution,
while the other 20% have very high SINR since they are close to the home base station and farther from
the interfering cells.
Assuming that the target mobile stations are uniformly distributed, we can obtain the average Shannon
capacity which represents achievable average cell throughput. In Figure 7, the achievable average cell
throughput is provided according to pathloss exponents. When the pathloss exponent is 3.5, the throughput
of the blanket transmission scheme and the single transmit selection scheme in cellular DAS are about
120% and 190% of throughput of conventional cellular system, respectively. This large improvement is
due to the fact that three times more users are outside of the radius 1/2R within a cell assuming uniform
distribution.
B. Ergodic Capacity with CSIT
When channel is known to the transmitter side, the ergodic Shannon capacity at a given location of the
target mobile station can be given by [27]
C
e
= E
h
_
max
S
ii
P
(0)
i
i
log
2
_
1 +
1
2
z
hSh
H
_
_
(35)
where S
ii
is the ith diagonal element of S. The maximum capacity is achieved by a proper design of the
transmit power spectrum matrix S under individual power constraints of the distributed antenna modules
and the home base station. If power cooperation among the distributed antenna modules and the home
base station is allowable, this problem reduces to a standard water-lling problem in a macroscopic MISO
vector channel [17], [27] and the maximum capacity is simply given by
C
e
= E
h
_
log
2
_
1 +
Ph
2
2
z
__
(36)
where P =
6
i=0
P
(0)
i
. However, the per-antenna power constraints make it meaningless to apply the
simple water-lling algorithm. Therefore, we should nd another solution instead of the conventional
water-lling algorithm.
Let the eigenvalue decomposition of the symmetric positive denite matrix S be S = UU
H
where
= diag(
0
,
1
, ,
6
). Then, the capacity formula in (35) can be given by
C
e
= E
h
_
max
i
P
(0)
i
i
log
2
_
1 +
1
2
z
ff
H
_
_
(37)
= E
h
_
max
i
P
(0)
i
i
log
2
_
1 +
1
2
z
6
i=0
|f
i
|
2
i
__
(38)
where f = hU and f
i
is the ith element of a 1 7 vector f.
Since the log function is a monotonic increasing function with respect to
i
i, the maximum capacity
is achieved if
i
= P
(0)
i
i as
C
e
= E
h
_
log
2
_
1 +
1
2
z
6
i=0
|f
i
|
2
P
(0)
i
__
(39)
We should emphasize here that the best transmission strategy of DAS with per antenna module power
constraints is for each antenna module and home base station to transmit allowable full power. This
result indicates that the channel state information at the transmitter (CSIT) in DAS does not provide
any advantage over the case that the channel state information is provided only to the receiver if there
are per-antenna module power constraints. However, if each antenna module and home base station has
multiple antennas, CSIT will certainly help increase capacity.
For simplicity of analysis, we assume P
(0)
i
= P/7 i. Then, since U is the unitary matrix, it holds that
6
i=0
|f
i
|
2
= hh
H
=
6
j=0
|h
(0)
j
|
2
L
(0)
j
and the capacity formula is given by
C
e
= E
h
_
log
2
_
1 +
P
7
2
z
6
i=0
|h
(0)
i
|
2
L
(0)
i
__
(40)
=
1
ln 2
6
i=0
i
exp
_
7
2
z
L
(0)
i
P
_
Ei
_
7
2
z
L
(0)
i
P
_
(41)
where
i
=
6
k=0,k=i
L
(0)
i
L
(0)
i
L
(0)
k
. Equation (41) is derived from the fact that

6
i=0
|h
(0)
i
|
2
L
(0)
i
is a weighted
chi-squared distributed random variable.
Figure 8 shows the ergodic capacity when channel state information is known to the transmitter
according to the normalized distance from the home base station when the pathloss exponent is 4.0.
In this gure, we assume that the transmit power of each distributed antenna module and the home base
station in DAS is P/7 whereas the transmit power of the base station in the traditional cellular system
is P. Similarly to the CSIR case where the transmit power of each distributed antenna module is 0.1P
and the transmit power of the home base station is 0.4P, the single transmit selection scheme achieves
the highest throughput owing to the OCI reduction and macroscopic selection diversity. Although the
achieved throughput of DAS is slightly lower than that of conventional cellular system near the home
base station due to reduced transmit power, the achieved throughput of DAS has substantially higher
throughput beyond the normalized distance 0.5. When the target mobile stations are uniformly distributed
in space, the average Shannon capacity is plotted versus pathloss exponent in Figure 9. This gure also
shows the similar result to the CSIR case.
VI. CONCLUSIONS
In this paper, the architectural advantages of downlink DAS in a multicell environment have been
investigated in terms of power efciency, SINR improvement, and outage probability. We have also
analyzed transmit diversity performance and achievable capacity of various transmission schemes in
cellular DAS from an information theoretic standpoint. The results of this paper suggest that distributed
antenna systems effectively reduce OCI and improve SINR compared to conventional cellular systems in an
interference-limited multicell environment. As a result, distributed antenna systems achieve lower symbol
error probability and higher capacity than conventional cellular systems. Based on the results of this paper,
the distributed antenna architecture might be an effective solution to reduce OCI in an interference-limited
cellular environment. An additional contribution of this paper is a theoretical framework for analyzing the
achievable capacity and diversity performance of conventional repeater systems, a special case of DAS.
REFERENCES
[1] J. G. Andrews, Inteference cancellation for celluar systems: A contemporary overview, IEEE Wireless Communications Magazine,
Apr. 2005.
[2] F. Adachi, Wireless past and future - Evolving mobile communication systems, IEICE Trans. on Funamentals of Elec. and Comm.,
vol. E84-A, no. 1, pp. 5560, Jan. 2001.
[3] A. A. M. Saleh, A. J. Rustako, and R. S. Roman, Distributed antennas for indoor radio communications, IEEE Trans. on
Communications, vol. 35, pp. 12451251, Dec. 1987.
[4] M. V. Clark and et al., Distributed versus centralized antenna arrays in broadband wireless networks, in Proc., IEEE Veh. Technology
Conf., May 2001, pp. 3337.
[5] L. Dai, S. Zhou, and Y. Yao, Capacity with MRC-based macrodiversity in CDMA distributed antenna systems, in Proc., IEEE
Globecom, Nov. 2002, pp. 1721.
[6] W. Roh and A. Paulraj, Outage performance of the distributed antenna systems in a composite fading channel, in Proc., IEEE Veh.
Technology Conf., Sept. 2002, pp. 15201524.
[7] A. Obaid and H. Yanikomeroglu, Reverse-link power control in CDMA distributed antenna systems, in Proc., IEEE Wireless
Communications and Networking Conf., Sept. 2000, pp. 608612.
[8] R. E. Schuh and M. Sommer, WCDMA coverage and capacity analysis for active and passive distributed antenna systems, in Proc.,
IEEE Veh. Technology Conf., May 2002, pp. 434438.
[9] R. Hasegawa, M. Shirakabe, R. Esmailzadeh, and M. Nakagawa, Downlink performance of a CDMA system with distributed base
station, in Proc., IEEE Veh. Technology Conf., Oct. 2003, pp. 882886.
[10] A. Sendonaris, E. Erkip, and B. Aazhang, User cooperation diversity part I and part II, IEEE Trans. on Communications, vol. 51,
no. 11, pp. 19271948, Nov. 2003.
[11] A. Nosratinia, T. Hunter, and A. Hedayat, Cooperative communication in wireless networks, IEEE Communications Magazine, vol. 42,
no. 10, pp. 7480, Oct. 2004.
[12] J. N. Laneman and G. W. Wornell, Distributed space-time coded protocols for exploiting cooperative diversity in wireless networks,
IEEE Trans. on Info. Theory, vol. 49, no. 10, pp. 24152425, Oct. 2003.
[13] M. Janani and et al., Coded cooperation in wireless communications: Space-time transmission and iterative decoding, IEEE Trans.
on Signal Processing, vol. 52, no. 2, pp. 362371, Feb. 2004.
[14] W. C. Y. Lee and D. J. Y. Lee, The impact of repeaters in CDMA systems, in Proc., IEEE Veh. Technology Conf., 2001, pp.
17631767.
[15] M. Rahman and P. Ernstrom, Repeaters for hotspot capacity in DS-CDMA networks, IEEE Trans. on Veh. Technology, vol. 53, no. 3,
pp. 626633, May 2004.
[16] W. Choi, B. Cho, and T. Ban, Automatic on-off switch repeaters for DS/CDMA reverse link, IEEE Communications Letters, vol. 4,
no. 5, pp. 138141, Apr. 2001.
[17] H. Zhuang, L. Dai, L. Xiao, and Y. Yao, Spectral efciency of distributed antenna systems with random antenna layout, Electronics
Letters, vol. 39, no. 6, pp. 495496, Mar. 2003.
[18] L. Xiao, L. Dai, H. Zhuang, S. Zhou, and Y. Yao, Information-theoretic capacity analysis in MIMO distributed antenna systems, in
Proc., IEEE Veh. Technology Conf., Apr. 2003, pp. 779782.
[19] Y. Chung and D. H. Cho, Performance analysis of handoff algorithm in ber-optic microcell/picocell radio system, in Proc., IEEE
Veh. Technology Conf., May 2000, pp. 24082412.
[20] A. Ghosh, D. R. Wolter, J. G. Andrews, and R. Chen, Broadband wireless access with WiMax/802.16: Current performance benchmarks
and future potential, IEEE Communications Magazine, vol. 43, no. 2, pp. 129136, Feb. 2005.
[21] A. Goldsmith, Wireless Communications, will be published by Cambridge in 2005, now at http://wsl.stanford.edu/andrea/EE359Reader/.
[22] S. Schwartz and Y. Yeh, On the distribution function and moments of power sums with lognormal components, Bell Syst. Tech. Jour.,
vol. 61, no. 7, pp. 14411462, Sept. 1982.
[23] T. Lo, Maximal ratio transmission, IEEE Trans. on Communications, vol. 47, no. 10, pp. 14581461, Oct. 1999.
[24] A. Paulraj, R. Nabar, and D. Gore, Introduction to Space-Time Wireless Communications. Cambridge, UK: Cambridge, 2003.
[25] J. Proakis, Digital Communications. New York: McGraw-Hill, 1995.
[26] I. Gradshteyn and I. Ryzhik, Table of Integrals, Series, and Products. London, UK: Academic Press, 2003.
[27] T. Cover and J. Thomas, Elements of Information Theory. New York: Wiley-Interscience, 1991.
R
BS : Base station
: Remote Anteannas Module (RNA)
BS
0
R2
R1
R6
R4
R5
R3
BS
5
R2
R1
R6
R4
R5
R3
BS
6
R2
R1
R6
R4
R5
R3
BS
3
R2
R1
R6
R4
R5
R3
BS
4
R2
R1
R6
R4
R5
R3
BS
2
R2
R1
R6
R4
R5
R3
BS
1
R2
R1
R6
R4
R5
R3
W
W : Example worst case location
Fig. 1. Structure of a distributed antenna system
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
20
0
20
40
60
80
100
120
Normalized distance from the home base station (distance/R)
S
i
g
n
a
l
t
o
I
n
t
e
r
f
e
r
e
n
c
e

r
a
t
i
o

(
d
B
)
Pathloss exponent () = 4.0
DAS w/ blanket transmission scheme
DAS w/ single transmit selection scheme
DAS w/ dual transmit selection scheme
Conventional cellular system
Fig. 2. Signal-to-interference-plus-noise ratio (SINR) versus the normalized distance from the home base station.
2.5 3 3.5 4
0
5
10
15
20
25
30
35
Pathloss exponent
A
v
e
r
a
g
e

S
I
N
R

(
d
B
)
Fig. 3. Signal-to-interference-plus-noise ratio (SINR) versus the path loss exponent
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
P
o
u
t

=

P
r
o
b
(
S
I
N
R
<
1
0
d
B
)
Fig. 4. Outage probability versus the normalized distance from the home base station (outage threshold =10 dB)
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
10
5
10
4
10
3
10
2
10
1
10
0
S
y
m
b
o
l

E
r
r
o
r

P
r
o
b
a
b
i
l
i
t
y
(a) pathloss exponent = 3.0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
10
6
10
5
10
4
10
3
10
2
10
1
10
0
S
y
m
b
o
l

E
r
r
o
r

P
r
o
b
a
b
i
l
i
t
y
(b) pathloss exponent = 4.0
Fig. 5. Uncoded symbol error probability versus the normalized distance from the home base station
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0
5
10
15
20
25
30
35
C
a
p
a
c
i
t
y

(
b
p
s
/
H
z
)
Pathloss exponent () = 4
Fig. 6. Ergodic capacity with CSIR versus the normalized distance from the home base station
2.5 3 3.5 4 4.5
1
2
3
4
5
6
7
8
9
Pathloss exponent
A
v
g
.

S
h
a
n
n
o
n

c
a
p
a
c
i
t
y

(
b
p
s
/
H
z
)
Fig. 7. Average Ergodic capacity with CSIR versus the pathloss exponent
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0
5
10
15
20
25
30
35
Normalized distance from a home base station (distance/R)
C
a
p
a
c
i
t
y

(
b
i
t
s
/
H
z
)
Pathloss exponent () = 4
Fig. 8. Ergodic capacity with CSIT versus the normalized distance from the home base station
2.5 3 3.5 4 4.5
1
2
3
4
5
6
7
8
9
Pathloss exponent
A
v
g
.

S
h
a
n
n
o
n

c
a
p
a
c
i
t
y

(
B
p
s
/
H
z
)
Fig. 9. Average Ergodic capacity with CSIT versus the pathloss exponent

ChoAnd DAS Twireless PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ChoAnd DAS Twireless PDF

Uploaded by

Copyright:

Available Formats

SUBMISSION TO IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS 1

Downlink Performance and Capacity of Distributed

You might also like