The use of
multiple antennas at both the transmit and receive ends has become one of the
most important paradigms for the deployment of existing and emerging wireless
communications systems. The importance of Multiple-Input Multiple-Output (MIMO)
systems is witnessed by their presence in many recent standards, such as
IEEE802.11n, Worldwide Interoperability for Microwave Access (WiMAX) and Long
Term Evolution (LTE). Interestingly, like many other brilliant solutions
developed in these years in the field of digital communications, the MIMO
concept is not a new idea but dated back to the 1970s and was studied as a
model for multi-pair telephone cables.
It is part of the
“natural” evolution of systems that a first version is standardized and
implemented, further versions following, improving specific aspects of the
standard, including additional features, toward an increase of performance in
various directions. These improvements can be dealing with very specific
aspects of the system, hence, not being directly usable for other systems, or
can be the result of integrating some general technology into the system, thus,
taking advantage of a broader knowledge and of a larger number of suppliers.
The use of MIMO in cellular networks and WLANs is clearly in the latter case.
Sometimes, the implementation of features in a system requires only the effort
of infrastructure manufacturers, in the sense that these features do not affect
mobile terminals; the use of diversity in a base station is a good example of
this situation.
However, many
other times, a joint work and collaboration between infrastructure and mobile
terminals manufacturers is required and essential, in order to achieve the
desired target. Obviously, the latter situation increases the complexity of
system improvement. In the case of MIMO, the latter rule also applies, since it
requires the use of multiple antennas in both base station/access point and
mobile terminal.
The increase of
system performance can also be seen from two different perspectives. On one
hand, some features are totally independent of users, being fully controlled by
operators; the use of half-rate in GSM is a good example, among many others. On
the other hand, many performance parameters do depend on the type of terminal
users have, and on how they use it; of course, this creates problems in taking
full advantage of the technology and in achieving the goals of optimum/maximum
performance. Again, MIMO is the latter perspective. It is clear that, although
MIMO is a very powerful technology for the increase of performance (i.e.,
capacity, data rate, quality of service, etc.) of mobile and wireless systems,
really profiting from it can be a complex and nonguaranteed process. Still, the
potential of MIMO is so high that, regardless of these barriers, it is being
implemented in current upgrades of these systems; for example, LTE, WiMAX, and
WiFi.
Nowadays, major
manufacturers for cellular networks infrastructure already deliver MIMO products
for base stations; four antennas are already made available, and trials with
eight antennas have also been announced. However, the situation is quite
different as far as mobile terminals are concerned: at this end of the link,
terminals are severely size constrained and the space available to locate
multiple antennas is often desperately limited. Fortunately, there is a lot of
work being done on co-location of multiple antennas in mobile phones with a
relatively low correlation, especially for orthogonal linear polarizations.
On the other
hand, one should not look into this problem by taking only (current) mobile
phones as user’s terminals, other devices being real alternatives as well;
personal computers (to be used not only for WiFi, but also for cellular
networks) may include two or four antennas in a relatively decoupled way (there
are a number of publications on this topic), and even other hypothesis that do not
present space problems can be considered; for example, the use of cars as the ”interface”
between the mobile phone and wave propagation (this is already a reality today,
for high-end cars). So, although there may currently be some real problems in
the implementation of MIMO in mobile and wireless systems, by taking advantage
of multiple antennas at both ends of the link, the conditions for the efficient
deployment of MIMO systems will definitely be favorable in the near future.
In the recent
standard [IEE09a], up to four antennas are foreseen, but most of the proposed
schemes apply only to two, like the two schemes that are mandatory for use in
downlink, but are optional for uplink, i.e., Space Time Block Coding and
Spatial Multiplexing; still, a combination of these two schemes is being
considered. Again, MIMO improves system performance, as expected.
LTE has been
designed having MIMO in mind from the beginning [3gp09]. The effort in standardization
was focused on specifying efficient schemes for downlink single user MIMO (that
is, no indications are given for a transmission to multiple users such as
broadcast), while for uplink the system was designed on the assumption that the
mobile terminal front-end comprises a single signal chain; hence, for the
latter, the support of single user MIMO is limited to adaptive transmit antenna
switching; multiuser schemes are not excluded but are left to manufacturers
implementation. The techniques and schemes considered in LTE for MIMO are not
that much different from the ones taken in the previous systems, namely because
they all share basic aspects of the multiple access technique.
However, MIMO
development and usage is not going to be finished by the standardization of the
previously mentioned systems, and one may see quite a number of developments in
the future, some of them listed below;
MIMO Moves to MM-Wave Bands............
Due to the fact
that the UHF band has excellent propagation characteristics for mobile and
wireless communications, the race on higher data rates has not implied (yet)
the move to higher frequency bands, and basically all mobile and wireless
systems working today use this band. Moreover, with the digital dividend (the
use of part of the spectrum released from broadcast systems to mobile and
wireless ones), the pressure to go higher in frequency has decreased. The
achievement of higher data rates are usually obtained by more efficient
multiple access techniques and modulation schemes, better management of the
radio resources, and so on, rather than by investigating in the area of
propagation and channels. MIMO was the exception confirming the rule, i.e., by
taking advantage of the randomness of the propagation channel, it was possible
to profit from the “parallel channels,” and therefore, to enable higher data
rates. Nevertheless, one cannot avoid going up in frequency, if higher data
rates are to be obtained, especially because efficiency and performance
boundaries are being attained today. The knowledge of propagation and channels
up to the UHF band is almost complete, but beyond that (including mm waves and
up to the THz band) there is still a lot of work to be done, and a thorough
characterization of MIMO in these high frequency bands is still to be
performed.
MIMO Goes Green...................
Mobile
communications gave rise many years ago to the fears from (electromagnetic) radiation.
Basically, the extremely fast introduction of a new technology in the mass
market, without a proper explanation of the system behavior, generated a lot of
health concerns, which are still present today. One can currently envision that
environmental concerns, which are already of key importance nowadays (“green communications”
is in the agenda of many initiatives, projects and fora), will become even more
important, with a huge impact on systems and networks, concerning their
development, deployment and operation. Therefore, energy efficiency has to be
(is already being) taken into account in mobile and wireless communications. MIMO
can play an important role in this matter, through a better exploitation of the
spatial/polarization degrees of freedom toward more efficient communications,
at a lower transmission power. Moreover, by combining MIMO with beamforming,
one can further decrease the transmission power (given the increase in the
corresponding antenna gain); hence, providing additional contribution to
increase energy efficiency.
MIMO as an Enabler of the RF SIM Card..............
Another trend
regards mobile terminals. The evolution has been such that one can foresee that
users may carry just an RF SIM card for their identification by the network,
using any terminal/device at hand for communication, ranging from a PC to a “TV
screen,” encompassing terminals embedded in cars and in the office, and
reaching spectacles as the replacement of today’s mobile phones. Therefore, short
range communications, in the vicinity of a person’s body, will play a major
role. The exploration of MIMO in such distance ranges has not deserved much
attention, most probably due to the low randomness of the propagation channel,
as well as to the difficulties in having more than one antenna available at
either end of the link. Still, given the development of technology (in antenna
design, signal processing, nanodevices, etc.), together with an increase in
frequency band, one can easily envision that the conditions for deploying MIMO
in these kind of applications may arise in the near future.
MIMO Technologies and the Propagation
environment.............
Propagation in
not so usual environments needs to be studied as well, for example: several
indoor scenarios, with a differentiation between business and residential
environments; for personal systems, in-, on- and off-body propagation; for car
communications, propagation in between cars, from a heavy traffic situation in
a motorway (addressing high speed as well) to a city street; public transport
scenarios, including buses, trains, planes, and boats, among others. The
characterization of all these scenarios will allow one to estimate how far they
are appropriate for MIMO technologies, and how to optimize the location of
antennas.
Geographical Mapping of Channel Conditions............
Location based
services are already popular nowadays, where a user can know his/her location
via the mobile phone, and take advantage of them to navigate or to find a nearest
shop, among other applications. Basically, this means that the user is
accessing the network for location purposes. However, one can invert this
concept, that is, the network knowing where users are, and take advantage of
it. For example, by exploiting users’ terminals (and other devices) as channel
sensors, the network can establish a geographical map of channel conditions,
hence, of channel quality, and with that information forecast services
availabilities in an efficient way (specifically for each user). The creation
of such geographical channel mapping for better MIMO usage (i.e., enabling the
system to know where and when to use MIMO, and then, increasing data rate
accordingly), in a given scenario (from indoors to streets, including bodies and
cars, and many others), would definitely increase the overall system
performance.
M2M Communication....................
Machine-to-machine
communications (recently renamed as the Internet of Things) no longer
requires a dedicated introduction. The increasing number of applications of
this type of communications, where the user has no intervention, is obvious,
and definitely will surpass human communications (voice or data) in terms of
traffic volume, as data has already surpassed voice in terms of exchanged
number of information bits. This also extends to car-to-car and
car-to-infrastructure communications, and other types of systems. This means
that one will have devices all over the surroundings, at any location, many
times with low power consumption requirements, communicating not only among
each other but also to a control or information network. Again, MIMO can play
an important role in this type of communication, and conditions for deployment
should be explored.
The Internet of Things and MIMO................
Sensors can be
considered as subset of the Internet of Things, not only for personal use
(e.g., on the body), but also through myriad devices that start being implanted
in our surroundings (e.g., in cars or houses). Being a subset of the Internet
of Things, the same type of problems apply, but one can envision in this
case that power consumption and efficiency of communications will play enhanced
roles. This opens a wide range of study cases, by extending the MIMO concept to
the use of different multiple antennas, located at somehow random locations
(e.g., buttons on clothes); hence, creating the need to analyze system behavior
under those conditions.
In summary, the
possibilities for deploying and amplifying MIMO technologies in the future are
immense, bringing to the conclusion that the exploitation of these technologies
is just in its infancy.
