Welcome to Scribd. Sign in or start your free trial to enjoy unlimited e-books, audiobooks & documents.Find out more
Download
Standard view
Full view
of .
Look up keyword
Like this
0Activity
0 of .
Results for:
No results containing your search query
P. 1
MIMO Wireless Communication

MIMO Wireless Communication

Ratings: (0)|Views: 3|Likes:
Published by Sarbaz Irani
MIMO Wireless Communication
MIMO
Wireless
Communication
MIMO Wireless Communication
MIMO
Wireless
Communication

More info:

Published by: Sarbaz Irani on Sep 08, 2013
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

09/08/2013

pdf

text

original

 
• BLISS, FORSYTHE, AND CHAN
 MIMO Wireless Communication
VOLUME 15, NUMBER 1, 2005 LINCOLN LABORATORY JOURNAL
97
 
MIMO WirelessCommunication
Daniel W. Bliss, Keith W. Forsythe, and Amanda M. Chan
Wireless communication using multiple-input multiple-output (MIMO)systems enables increased spectral efficiency for a given total transmit power.Increased capacity is achieved by introducing additional spatial channels that areexploited by using space-time coding. In this article, we survey the environmentalfactors that affect MIMO capacity. These factors include channel complexity,external interference, and channel estimation error. We discuss examples of space-time codes, including space-time low-density parity-check codes and space-time turbo codes, and we investigate receiver approaches, including multichannelmultiuser detection (MCMUD). The ‘multichannel’ term indicates that thereceiver incorporates multiple antennas by using space-time-frequency adaptiveprocessing. The article reports the experimental performance of these codes and receivers.
M
-
multiple-output (MIMO) sys-tems are a natural extension of developmentsin antenna array communication. While theadvantages of multiple receive antennas, such as gainand spatial diversity, have been known and exploited forsome time [1, 2, 3], the use of transmit diversity hasonly been investigated recently [4, 5]. The advantagesof MIMO communication, which exploits the physi-cal channel between many transmit and receive anten-nas, are currently receiving significant attention [6–9]. While the channel can be so nonstationary that it can-not be estimated in any useful sense [10], in this article we assume the channel is quasistatic.MIMO systems provide a number of advantagesover single-antenna-to-single-antenna communication.Sensitivity to fading is reduced by the spatial diversity provided by multiple spatial paths. Under certain envi-ronmental conditions, the power requirements associ-ated with high spectral-efficiency communication canbe significantly reduced by avoiding the compressive re-gion of the information-theoretic capacity bound. Here,spectral efficiency is defined as the total number of in-formation bits per second per Hertz transmitted fromone array to the other. After an introductory section, we describe the con-cept of MIMO information-theoretic capacity bounds.Because the phenomenology of the channel is impor-tant for capacity, we discuss this phenomenology andassociated parameterization techniques, followed by ex-amples of space-time codes and their respective receiversand decoders. We performed experiments to investigatechannel phenomenology and to test coding and receivertechniques.
Capacity 
 We discuss MIMO information-theoretic performancebounds in more detail in the next section. Capacity in-creases linearly with signal-to-noise ratio (SNR) at low SNR, but increases logarithmically with SNR at highSNR. In a MIMO system, a given total transmit powercan be divided among multiple spatial paths (or modes),driving the capacity closer to the linear regime for eachmode, thus increasing the aggregate spectral efficiency. As seen in Figure 1, which assumes an optimal highspectral-efficiency MIMO channel (a channel matrix with a flat singular-value distribution), MIMO systemsenable high spectral efficiency at much lower requiredenergy per information bit.
 
• BLISS, FORSYTHE, AND CHAN
 MIMO Wireless Communication
98
LINCOLN LABORATORY JOURNAL VOLUME 15, NUMBER 1, 2005
The information-theoretic bound on the spectral ef-ficiency is a function of the total transmit power andthe channel phenomenology. In implementing MIMOsystems, we must decide whether channel estimationinformation will be fed back to the transmitter so thatthe transmitter can adapt. Most MIMO communica-tion research has focused on systems without feedback. A MIMO system with an uninformed transmitter
 
(without feedback) is simpler to implement, and at highSNR its spectral-efficiency bound approaches that of aninformed transmitter (with feedback).One of the environmental issues with which com-munication systems must contend is interference, ei-ther unintentional or intentional. Because MIMO sys-tems use antenna arrays, localized interference can bemitigated naturally. The benefits extend beyond thoseachieved by single-input multiple-output systems, thatis, a single transmitter and a multiple-antenna receiver,because the transmit diversity nearly guarantees thatnulling an interferer cannot unintentionally null a largefraction of the transmit signal energy.
Phenomenology 
 We discuss channel phenomenology and channel pa-rameterization techniques in more detail in a later sec-tion. Aspects of the channel that affect MIMO systemcapacity, namely, channel complexity and channel sta-tionarity, are addressed in this paper. The first aspect,channel complexity, is a function of the richness of scat-terers. In general, capacity at high spectral efficiency increases as the singular values of the channel matrixincrease. The distribution of singular values is a mea-sure of the relative usefulness of various spatial pathsthrough the channel.
Space-Time Coding and Receivers 
In order to implement a MIMO communication sys-tem, we must first select a particular coding scheme.Most space-time coding schemes have a strong connec-tion to well-known single-input single-output (SISO)coding approaches and assume an uninformed trans-mitter (UT). Later in the article we discuss space-timelow-density parity-check codes, space-time turbo codes,and their respective receivers. Space-time coding canexploit the MIMO degrees of freedom to increase re-dundancy, spectral efficiency, or some combinationof these characteristics [11]. Preliminary ideas are dis-cussed elsewhere [6]. A simple and elegant solution that maximizes diver-sity and enables simple decoupled detection is proposedin Reference 12. More generally, orthogonal space-timeblock codes are discussed in References 13 and 14. A general discussion of distributing data across transmit-ters (linear dispersive codes) is given in Reference 15.High SNR design criteria and specific examples are giv-en for space-time trellis codes in Reference 16. Unitary codes optimized for operation in Rayleigh fading arepresented in Reference 17. Space-time coding withoutthe requirement of channel estimation is also a com-mon topic in the literature. Many differential coding schemes have been proposed [18]. Under various con-straints at the transmitter and receiver, information-theoretic capacity can be evaluated without condition-ing on knowledge of the propagation channel [19, 20].More recently, MIMO extensions of turbo coding havebeen suggested [21, 22]. Finally, coding techniques forinformed transmitter systems have received some inter-est [23, 24].
Experimental Results 
Because information-theoretic capacity and practicalperformance are dependent upon the channel phenom-enology, a variety of experiments were performed. Bothchannel phenomenology and experimental proceduresare discussed in later sections. Experiments were per-
05 –51020151050
b
/
0
(dB)
    S   p   e   c   t   r   a    l   e    f    f   i   c   i   e   n   c   y    (    b   i   t   s    /   s   e   c    /    H   z    )
= 16
= 8
= 4
= 1
FIGURE 1.
Spectral-efficiency bound as a function of noise-spectral-density-normalized energy per information bit(
b
/
0
). The graph compares four different
 
×
 
multiple-input multiple-output (MIMO) systems, assuming channelmatrices with flat singular-value distribution.
 
• BLISS, FORSYTHE, AND CHAN
 MIMO Wireless Communication
VOLUME 15, NUMBER 1, 2005 LINCOLN LABORATORY JOURNAL
99
formed in an outdoor nonstationary environment in a mixed residential, industrial, and light urban settings.Intentional high-power interference was included.
Information-Theoretic Capacity 
The information-theoretic capacity of MIMO systemshas been widely discussed [7, 25]. The development of the informed transmitter (“water filling”) and unin-formed transmitter approaches is repeated in this sec-tion, along with a discussion of the relative performanceof these approaches. (The concept of “water filling” isexplained in the sidebar entitled “Water Filling.”) Inaddition, we introduce the topic of spectral-efficiency bounds in the presence of interference, and we discussspectral-efficiency bounds in frequency-selective envi-ronments. Finally, we summarize alternative channelperformance metrics.
Informed Transmitter 
For narrowband MIMO systems, the coupling betweenthe transmitter and receiver for each sample in time canbe modeled by using 
z Hx n
= +
,
 (1) where
is the complex receive-array output,
H
×
n n
R
is the
n
 
×
 
n
(number of receive by transmit antenna)
 WATER FILLING
 W 
 
is a metaphor for the solutionof several optimization problems related tochannel capacity. The simplest physical example isperhaps the case of spectral allocation for maximaltotal capacity under a total power constraint. Let
 x 
 denote the power received in the
th frequency cell, which has interference (including thermal noise) de-noted
n
. If the total received power is constrainedto be
 x 
, then the total capacity is maximized by solving 
max log( )max 
{ }{ }
 x x k  x x
k k
 x n
: =: =
+ /=
1
k
n x n
+ .
log( ) log( )
Use Lagrange multipliers and evaluate
+       
 x n x x
 j  j  j 
log( )
µ 
to find a solution. The solution satisfies
 x 
+
n
=
µ 
–1
for all nonzero
 x 
. Figure A illustrates the solu-tion graphically as an example of water filling. Thedifference between the water level (blue) and thenoise level (red) is the power allocated to the signal
NoiseFrequency
      P    o    w    e    r
FIGURE A.
Notional water-filling example.
in each frequency cell. The volume of the water isthe total received power of the signal. Note that cells with high levels of interference are not used at all. A similar solution results when the capacity is ex-pressed by 
log( )1
+
g
k
for gains
 g 
. One can write the gains as
 g n
k
=
1
 and use the water-filling argument above. In thiscontext, cells with low gains may not be used at all.

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->