Multiple Antenna Techniques RefGuide v1.0
Multiple Antenna Techniques RefGuide v1.0
Multiple Antenna Techniques RefGuide v1.0
Antenna
Techniques
Reference
Guide
Version 1.0
Phone: +1.972.664.0727
Fax: +1.972.664.0729
If you have any questions, concerns or comments regarding this course please write to us at:
[email protected]
All rights reserved. This course book and the material and information contained in it ("course material") are
owned by Award Solutions, Inc. ("Award"). The course material shall not be modified, reproduced,
disseminated, or transmitted by or in any medium, form, or means, electronic or mechanical, including
photocopying, recording, or any information retrieval system, in whole or in part, without the prior express
written consent of Award. The unauthorized use, modification, reproduction, dissemination or transmission of
the course material, in whole or in part, is strictly prohibited.
This course reference guide is designed to be distributed as a student guide with the eLearning course and is
not designed to be a standalone course book. Award makes no representations or warranties and disclaims all
implied warranties with respect to the information contained herein or products derived from use of such
information and undertakes no obligation to update or otherwise modify the information or to notify the
purchaser or user of any update or obsolescence. Award’s total liability in connection with the course material
is the amount actually received by Award from the purchaser/user for the purchase of the course material.
Award Solutions, Inc. provides exceptional training and key points, repeat what’s important, and bring in analogies
consulting in advanced wireless and Internet technologies. and examples as needed. They are focused on knowledge
Our proven experience enables us to offer a complete suite transfer and don’t teach just “by the book,” instead
of services: Cutting edge technology training, customized adapting to the students’ needs. They bring invaluable
training solutions, and advanced technology consulting. knowledge into the classroom because they can relate the
theory to real-world experiences.
Our products and services provide our clients with
innovative, flexible, and cost-effective solutions that help We offer a multitude of delivery methods and services:
rapidly boost workforce productivity and competence to
more quickly meet market demands. Our areas of expertise On-site Instructor Led Training: Full classroom
include: experience and interactivity that comes with in-
person training.
• Emerging Trends • UMTS/HSPA/HSPA+
• LTE • 1x & 1xEV-DO Web-based Instructor Led Training:
• WiMAX • GSM & GPRS/EDGE A convenient alternative to on-site training,
• IP Convergence & IMS • Wireless Fundamentals without losing the benefits of a live instructor.
The level of technical depth in our training courses gives Mentoring Programs and Performance
students unique benefits that they can apply immediately. Workshops: A mix of classroom training and
We offer a range of courses appropriate for audiences hands-on applications of the concepts.
needing a high-level overview, as well as engineers looking
for technical details. Public Courses: Instructor Led training for
individuals or small groups of professionals. Visit
Knowledge Services our website for the full schedule of upcoming
As a knowledge-based company, Award Solutions is sessions.
recognized for being the expert, and equipping each client
to “become an expert.” Our instructors and consultants Self-Paced eLearning: A flexible, cost-effective
are best-in-class, having achieved substantial industry solution, allowing your team to take training
experience in areas such as product definition and online at their own pace when their schedule
development, network deployment, and network and permits.
systems engineering.
Custom Learning Solutions (CLS): Training
Award Solutions constantly keeps a finger on the pulse programs created and tailored to your specific
of the industry, always researching new technologies, needs.
and updating our curriculums to stay on the cutting edge
with changes to existing courses and the addition of new Consulting Services: Maximize your investment,
courses. augment your team or receive guidance from
Award Solutions’ experts.
Whether you are a training manager looking to train others,
or a team lead responsible for enhancing your team’s skills, If you are interested in or have any questions regarding
Award Solutions can meet your technology training needs. Award Solutions training or services, please visit Award’s
website at www.awardsolutions.com or contact us at +1-
Not only do our Subject Matter Experts (SMEs) understand 972-664-0727 ext. 306 or toll free at +1-877-472-9273 ext.
the technology, they know how to teach, emphasize the 306.
Table of Contents
1.0 COURSE OBJECTIVES........................................................................................................1
2.0 ANTENNA BASICS..............................................................................................................1
Simple Antennas ................................................................................................................................................. 1
Radiation Pattern: Where an Antenna Spreads Energy ..................................................................................... 2
Reciprocity of the Antenna Radiation Pattern .................................................................................................... 4
Active vs. Passive Elements ................................................................................................................................ 5
Antenna Gain ..................................................................................................................................................... 5
Antenna Beamwidth ........................................................................................................................................... 6
Omni Antenna vs. Directional Antenna .............................................................................................................. 6
How About a ”More Powerful” Omni Antenna? ................................................................................................ 8
Sector Antennas: A Closer Look ........................................................................................................................ 10
Antenna Down‐Tilt ........................................................................................................................................... 11
Why Multiple‐Antenna Techniques? Capacity ................................................................................................. 13
Why Multiple‐Antenna Techniques? Coverage ................................................................................................ 14
Why Multiple‐Antenna Techniques? Higher Data Rates .................................................................................. 15
Challenges of Advanced Antenna Techniques .................................................................................................. 15
3.0 TRANSMIT AND RECEIVE DIVERSITY............................................................................ 16
Motivation for Antenna Diversity ..................................................................................................................... 16
Concepts of Transmit and Receive Diversity ..................................................................................................... 16
Spatial Diversity as a Receive Diversity Example .............................................................................................. 17
What is Polarization? ....................................................................................................................................... 18
Polarization Diversity as a Receive Diversity Example ...................................................................................... 19
Transmit Diversity ............................................................................................................................................ 20
Time Diversity: A Transmit Diversity Example .................................................................................................. 20
Frequency Diversity: A Transmit Diversity Example ......................................................................................... 21
Space/Spatial Diversity: A Transmit Diversity Example .................................................................................... 22
Space Time Coding: An Advanced Transmit Diversity Technique ..................................................................... 23
Frequency Hopping Diversity: An Advanced Transmit Diversity Technique ..................................................... 24
Antenna Grouping: An Advanced Transmit Diversity Technique ...................................................................... 25
4.0 BEAMFORMING TECHNIQUES ...................................................................................... 26
Motivation for Antenna Beamforming ............................................................................................................. 26
Concept of Phase .............................................................................................................................................. 27
Concept of Phase: In‐phase Signals .................................................................................................................. 27
Concept of Phase: 90‐degree Phase Shift and Out‐of‐phase Signals ................................................................ 28
How Can I Form a Beam? ................................................................................................................................. 29
More on Receive Beamforming ........................................................................................................................ 31
Transmit Beamforming .................................................................................................................................... 32
Types of Beamforming ..................................................................................................................................... 33
A Switched‐Beam System ................................................................................................................................. 33
A Switched‐Beam System: Impact of User Mobility ......................................................................................... 34
Adaptive Beamforming .................................................................................................................................... 35
Switched‐Beam vs. Adaptive Beamforming ..................................................................................................... 35
5.0 MIMO SPATIAL MULTIPLEXING .................................................................................... 36
What is Spatial Multiplexing? .......................................................................................................................... 36
Separating Signals from Different Antennas .................................................................................................... 37
MIMO Terminology .......................................................................................................................................... 38
MIMO Performance Gain ................................................................................................................................. 39
Code Symbols and MIMO Transmission ........................................................................................................... 41
Vertical Encoding (MIMO‐Single Code Word) .................................................................................................. 42
Horizontal Encoding (MIMO‐ Multi Code Word) .............................................................................................. 43
Vertical vs. Horizontal Encoding ....................................................................................................................... 44
Motivation for MIMO with Antenna Selection Feedback ................................................................................. 45
MIMO with Antenna Selection Feedback ......................................................................................................... 46
Closed Loop MIMO with Precoding: Review of Underlying Principles .............................................................. 47
Inside the Closed Loop MIMO with Precoding .................................................................................................. 48
Space/Spatial Division Multiple Access (SDMA) ............................................................................................... 50
Latest and Greatest: MIMO + SDMA! ............................................................................................................... 51
Collaborative Spatial Multiplexing ................................................................................................................... 52
6.0 ACRONYMS ..................................................................................................................... 53
A half-wave dipole antenna radiates maximum energy in the center of the beam and zero energy along the axis
of the antenna. From another perspective, no energy is radiated in the direction of the electric current flowing
through the antenna wire or element.
The reason that this antenna is called a half-wave dipole is that the total length of the antenna is half of the
wavelength of the signal flowing through the antenna.
At 900 MHz frequency, the wavelength is about 1 foot and the length of the dipole antenna is about one-half of
a foot.
Consider a base station situated in the wide open, flat terrain with an antenna pointing due north. Assume that
the base station is transmitting the same amount of power to three mobile terminals that are exactly one
kilometer away. Relative to the base station, Mobile 1 is to the north, Mobile 2 is to the northwest, and Mobile
3 is to the northeast. For our example, please keep in mind that the north corresponds to 0 degrees and the
angle is measured in a clockwise direction from north.
Assume that Mobile 1 receives the signal from the base station at the power level of 1 micro Watt or -30 dBm.
Naturally, the maximum power is experienced in the center of the antenna beam.
The radiation pattern specifies gain in power relative to the center of the antenna beam as a function of the
angle. So, it has two parameters, angle and gain.
To calculate how much gain Mobile 3 at the 45 degree angle will experience relative to Mobile 1 at the 0
degree angle, draw a line from the base station to Mobile 3. This line intersects the antenna radiation pattern
at "-3 dB" gain. In other words, Mobile 3 will experience the signal strength that is 3 dB below the signal
strength experienced by Mobile 1. Specifically, Mobile 3 will receive the signal at -30 dBm -3 = -33 dBm.
Similarly, Mobile 2 is in the northwest at the angle of 315 degrees, it will also receive the signal at -33 dBm.
Here is another example, consider Mobile 4 which is also one kilometer away from the base station but at an
angle of 60 degrees. Since the radiation pattern at 60 degrees is associated with a relative gain of -30 dB,
Mobile 4 will experience the signal level of -30 dBm - 30 dB = -60 dBm.
Mobile 1, again, is in the center of the antenna beam and its signal is received by the base station at the power
level of 1 nano Watt or -60 dBm.
Since Mobile 4 is at an angle of 60 degrees, its signal is received 30 dBs weaker based on the antenna
radiation pattern, at the level of -60 dBm - 30 dB = -90 dBm.
In summary, a given radiation pattern can be used to analyze the signal levels in different directions.
In a cellular system, the power amplifier increases the signal power by amplifying the signal and thus is an
example of an active element. The antenna is a passive element; it cannot amplify a signal.
Antenna Gain
If the antenna is a passive element, how can it have a gain? The gain of a given antenna is expressed relative
to a reference antenna such as the hypothetical isotropic antenna or a dipole antenna.
If the gain of an antenna is 18 dB, it means that compared to an isotropic antenna, this antenna provides 18
dBs higher signal power in the center of the beam. You might remember from our discussion of the antenna
characteristics that the antenna gain reaches its maximum in the center of the beam and is lower away from
the center of the beam.
Antenna Beamwidth
Beamwidth is another important parameter that characterizes the antenna radiation pattern. Beamwidth
essentially indicates the region over which the antenna focuses energy. Typically, half-power beamwidth is
used.
Half-power beamwidth corresponds to the range of angles over which the power is greater than or equal to half
of the maximum power. Halving the power reduces the power by 3 dB. Let's consider our antenna radiation
pattern.
In the first radiation pattern, the half-power beamwidth is 90 degrees as the relative gain falls to -3 dB at 315
degrees and at 45 degrees.
In the second example of the radiation pattern, the half-power beamwidth is 60 degrees as the relative gain
falls to -3 dB at 330 degrees and at 30 degrees.
In summary, the antenna beamwidth tells us whether the antenna is focusing energy in a narrow area or a
wide area. Another side effect of a narrower beamwidth is that a strong signal reaches further distances.
A lamp without a shade spreads light in all directions and the light intensity decreases as the distance from the
lamp increases.
Similar to the lamp, a base station with an omni-directional antenna transmits in all directions. In the first
example shown here, all three mobile terminals receive the same signal from the base station.
Now, consider a flashlight or a desk light. It focuses light in a specific region and does not spread light in other
regions. Just like the flashlight, a directional or sector antenna has more intensity in some region and less
intensity in others. For example, the base station serves three sectors using three different sector antennas.
The sector antenna is focusing energy in a specific sector such as Sector Gamma. Sector Gamma does not
radiate appreciable energy in Sector Alpha or Sector Beta.
Now, how can we increase the gain of an omni antenna? One possible approach is the use of a collinear array
antenna, where multiple antenna elements are stacked inside the antenna enclosure. The antenna enclosure
is called a radome. What we usually see on the base station towers are radomes and the radomes house the
actual antenna elements.
Let's keep just one antenna element inside the radome, which results in small coverage. In this case, we need
several base stations to cover the target service area.
Now, instead of using just one element, let's use two antenna elements. This will increase the antenna gain by
3 dB compared to one element example as the antenna radiation patterns for two antenna elements add up
constructively. Now, we can cover the same area with fewer Base Stations.
Let's make the number of antenna elements four. This arrangement will further increase the antenna gain to 6
dB compared to the one element case, enabling larger coverage and requiring fewer base stations compared
to the two-antenna element example.
In summary, multiple-antenna elements help increase the gain of an antenna as electrical signals for different
antenna elements add up.
As you may recall, a sector antenna operates like a flashlight as it focuses energy in a specific region instead of
spreading energy in all directions. If you look inside the flashlight, you will find a lamp that spreads light in all
directions. It is the mirror-like reflecting elements that reflect light to focus the overall light in a specific
direction.
Similarly, a panel antenna utilizes an omni-directional antenna along with a reflecting element that directs
energy over a predefined beamwidth. The panel directional antennas are the most popular directional
antennas used today. A base station with three panel antennas is the most commonly deployed configuration
today.
In summary, the use of a reflecting element enables a panel antenna to focus energy over a specific region
instead of shooting energy all over the place.
Antenna Down-Tilt
Consider two base stations using sector antennas. The signal from one base station reaches far into another
base station, causing interference. Is there any way we can reduce the reach of the signal? The answer is YES!
One way to limit the extent of RF coverage is to use antenna down-tilting. It is a widely used mechanism.
Let's use the example of a flashlight to appreciate the process of down-tilting. If we hold a flashlight parallel to
the ground, you will see that the light reaches quite a distance.
Instead of holding the flashlight parallel to the ground, let's tilt the flashlight toward the ground. Notice that the
light covers a smaller area on the ground! The same idea is used for antenna down-tilting.
Let's use two tilted flashlights each of which is giving off a different color of light. Since they are tilted, there is
no area of significant overlap.
In a cellular system, these flashlights represent the base stations that are using down-tilted antennas. This
down-tilting reduces the range or extent of the RF energy, leading to lower interference.
In summary, antenna down-tilting is an efficient method of controlling the RF coverage. While an example of
mechanical down-tilting is shown here, it is also possible to use electrical down-tilting.
Let's say that we can simultaneously serve 100 users in a sector with one antenna at the base station and one
antenna at the mobile terminal.
If we use an advanced antenna technique, we may be able to serve more users, perhaps as many as 400
users! In practice, two antennas at the mobile terminal may not be visible; we don't want ugly cell phones... do
we? Antennas could be hidden inside the casing of the phone.
Without multiple antennas, a mobile terminal can maintain its communication with the base station up to 3
km, but with multiple antennas the terminal can maintain the communication up to 4 km. Please note that 4
km is just an example number.
Let's consider a service operator that has deployed regular base stations without the advanced antenna
techniques. One thousand base stations might be required to cover the Dallas/Fort Worth Metroplex.
After learning the benefits offered by the advanced antenna techniques, the service operator decides to
embrace these new technologies and deploys advanced base stations in the Dallas/Fort Worth Metroplex. The
Service Operator may need just 200 advanced base stations instead of 1000 normal base stations, leading to
significant savings in infrastructure cost.
Basically, an increase in coverage due to multiple-antenna techniques translates into the need for fewer base
stations to cover a given area.
Joe and Tom are working for Manager Steve. Both are on a business trip on their way to attend a conference
but their plane is late. So, they are waiting in an airport lounge. Manager Steve sends an urgent email to Joe
and Tom, specifying an assignment that involves studying part of a large document attached to an email. Both
begin downloading the email.
By the time Average Joe downloads the email, Savvy Tom has already replied to Manager Steve since
advanced antenna techniques enabled Tom to download the email much faster. Guess who will get a bigger
bonus?
Since the design of advanced antenna techniques requires a significant R&D investment, the cost of the
mobile device and the base station supporting such techniques would obviously be higher than that of a simple
device and base station.
Some multiple-antenna techniques are quite complex to design and implement. For example, advanced
techniques to estimate the prevailing radio channel conditions are required.
When multiple antennas are deployed, network optimization has an added dimension for optimization. New
skill sets, experience, and tools become necessary to optimize network performance.
Consider a scenario where one transmitter sends out a signal from one antenna and the receiver receives the
signal. This signal is susceptible to any fading that occurs in the radio channel. If we transmit multiple copies of
the same signal, then it is less likely that a fade will impact all copies of the signal in the same way. In effect,
we've increased the possibility that the signal was received correctly. This is referred to as diversity gain.
The diversity gain can be exploited in several ways. For a given transmit power, it helps achieve higher data
rates and better coverage. Or, at a given data rate, it requires less transmit power.
When multiple paths between the transmitter and the receiver are created by using multiple receive antennas,
the resulting diversity is called receive diversity. One transmit antenna suffices in the case of receive diversity.
Consider two or more antennas spaced an adequate distance apart receiving signals from a single transmit
antenna. Since the transmitted information is received at two different "spaces" or antennas, such diversity is
called space or spatial diversity. This receive diversity is also called Single Input Multiple Output or SIMO.
Single Input refers to a single transmit antenna, and, Multiple Output refers to multiple receive antennas.
Commercial 2G and 3G cellular networks exploit spatial diversity by using multiple receive antennas at the
base station.
It is also possible to use multiple receive antennas at the mobile station. Emerging 3G devices have begun to
exploit spatial diversity at the mobile terminal. You can easily imagine how ugly two antennas sticking out of
your cell phone would be. To address such aesthetic concerns, one antenna could be hidden inside the casing
of the device, while the second regular antenna could be visible on the outside.
What is Polarization?
Hopefully you are familiar with polarized sunglasses. These vertically polarized sunglasses reduce the glare of
the sunlight by blocking horizontal reflections.
Let's discuss polarization of the RF signals before getting into the details of polarization diversity. You may
recall that the signal being radiated by an antenna is an electromagnetic wave, which consists of an electric
field and a magnetic field.
When the electric field is perpendicular to the Earth's surface, the resulting polarization is called vertical
polarization. A radiating antenna, when held vertical relative to the surface of the Earth, would essentially yield
vertical polarization. Vertical polarization is quite popular in cellular networks and that's why you see base
station antennas situated vertically on the towers.
When the electric field is parallel to the Earth's surface, the resulting polarization is called horizontal
polarization. A radiating antenna that is held horizontal relative to the surface of the Earth would essentially
yield horizontal polarization. Horizontal polarization is used to broadcast television.
Generally speaking, the best performance is achieved when polarization of the transmit antenna and the
receive antenna match. In other words, if the transmit antenna is vertical, use of a vertical receive antenna
yields maximum signal strength.
In spatial diversity at the base station, two antennas with vertical polarization receive signals from the mobile
station. In a rural environment, spatial diversity typically outperforms polarization diversity as the polarization
of the electromagnetic signal is likely to be vertical. Spatial diversity requires more space to mount multiple
antennas that are spaced farther apart.
One way to exploit polarization is to use cross-polarization, where an antenna array has physical antenna
elements that are situated at an angle of 45 degrees. When the mobile terminal's signal reaches this antenna
array, some of its power will be captured by Group A's antenna elements from one plane of the received
electric field. Group B's antenna elements will capture power from another plane of the received electric field.
Cross-polarization diversity can yield better performance than spatial diversity in urban and suburban
environments when there are numerous multipaths reaching the base station antenna from different
directions. In this case, antenna elements with different polarization can capture significant power.
Transmit Diversity
Meet our friend Warren. Warren wants to quickly find out the current stock quote for Advanced Wireless
Communications so that he can decide whether to buy the stock now or wait for a future buying opportunity.
The stock quote is eventually transmitted over the air from the base station to Warren's mobile device in binary
numbers, 1s and 0s. However, we want to make sure the communication is reliable; we do not want our friend
Warren to make a bad financial decision! Let's help Warren by using transmit diversity.
Time diversity exploits the fact that fading in the wireless channel varies over time. Let's say that the stock
quote is being sent to Warren via a bit stream. The basic idea of time diversity is that the same information is
transmitted at different time instants. For example, information bit A is transmitted at time instant t1, t2, and
t3, and, information bit B is transmitted at time instant t4, t5, and t6.
In our scenario the radio channel causes the signal to become very weak at time t2, so the mobile device
makes an error in detecting bit A at time t2. However, there is no need to worry as the two other A bits are just
fine! If we had transmitted bit A only once, the bit would be in error, leading to retransmission delays. Think of
our friend Warren, who is anxious to receive the stock quote quickly.
In summary, time diversity minimizes the probability of error by sending the same information at different
times.
Frequency diversity exploits the fact that fading in the wireless channel is frequency-selective, at a given time
or instance, it may adversely affect one frequency but not another frequency. We can use frequency diversity
instead of time diversity to quickly and reliably provide the stock quote to Warren.
Now, at time instant t1, let's send out all of the A bits on different frequencies f1, f2, and f3 from a base
station antenna. Also, at another time instant t2, let's send B bits simultaneously but on different frequencies
from a base station antenna.
Now, assume that the radio environment causes a fade on frequency f3 at time instant t1. In this case, the
receiver makes an error in estimating the bit. However, since there is no fade on f1 and f2, the A bits on f1 and
f2 are received without errors and, our friend Warren can obtain the correct stock quote quickly.
In summary, frequency diversity minimizes the probability of error by sending the information at different
frequencies.
Space or spatial diversity exploits the fact that fading in the wireless channel depends on the propagation path
between the transmitter and the receiver; it affects two paths differently at a given time instant.
Let's help our friend Warren by exploiting space diversity instead of time or frequency diversity.
Now, at time instant t1, let's send A bits simultaneously but from two different antennas. Also, at another time
instant t2, let's send B bits simultaneously from different antennas.
Now, assume that the radio environment causes a fade at time instant t1 on the signal path between antenna
2 and the receiver. In this case, the receiver makes an error in estimating bit A. However, since there is no fade
on the signal path between antenna 1 and the receiver, bit A is transmitted from antenna 1 and is received
without an error, and, our friend Warren can obtain the correct stock quote quickly.
In summary, space diversity minimizes the probability of errors by sending the information from different
antennas or different spaces.
Let's enhance the reliability of communications for our friend Warren by exploiting STC. We can use combined
space and time diversity instead of just time diversity or just frequency diversity to quickly and reliably provide
the stock quote to Warren.
At time instant t1 let's send bit A from Antenna 1 and bit B from Antenna 2. Furthermore, at another time
instant t2, let's send bit A from Antenna 2 and bit B from Antenna 1. Since the same bit, Bit A, is transmitted
from two different antennas, we are using space diversity. Since the same bit, Bit A, is transmitted at different
time instants, we are using time diversity.
In summary, STC exploits both space diversity and time diversity to improve reliability.
Frequency Hopping Diversity Coding (FHDC) is an example of space frequency coding. It combines both spatial
diversity and frequency diversity, thereby benefiting from the advantages of both space diversity and frequency
diversity. Let's provide reliable communications to our friend Warren by using FHDC.
At time instant t1, let's send bit A from Antenna 1 on frequency f1 and bit B from Antenna 1 on frequency f2. At
exactly the same time t1, let's send bit A from Antenna 2 on frequency f3 and bit B from Antenna 2 on
frequency f4.
Since the same bit, such as Bit A, is transmitted from 2 different antennas, we are using space diversity. Since
the same bit, for example Bit A, is transmitted on different frequencies, we are using frequency diversity. In
practice the set of frequencies on which the information bits is transmitted keeps hopping from one time
period to the next.
In summary, FHDC exploits both space diversity and frequency diversity to improve reliability.
Let's assume that the base station has four transmit antennas but Warren's mobile device can handle data
transmissions from only two of these four antennas.
Warren's device measures the quality of the signals being transmitted from the four base station antennas and
sends feedback on the signal quality to the base station. In this example, the device measures the best signal
quality for Antenna 1 and Antenna 3. The device sends "Antenna 1 and Antenna 3 grouping" as its feedback.
Since the signals transmitted from Antenna 1 and Antenna 3 are received by the mobile device with better
quality, the base station sends data packets to the mobile device by using Antennas 1 and 3 and avoiding
Antennas 2 and 4.
In summary, the antenna grouping facilitates transmission on better propagation paths, thereby increasing the
reliability of communication.
Consider a typical deployment scenario where a sector transmits a signal to our friend John near the edge of
the sector. Even though John is near the edge, his signal is radiated throughout 120 degrees of the sector,
what a waste of an expensive resource! And, poor John, he just sees one "signal bar" on his cell-phone in his
own home!
Consider a system that forms a narrow beam in the direction of John. Instead of shooting John's signal energy
all over the sector, the base station focuses energy for John in a narrow region, significantly increasing the
signal power received by John.
Just look at John. He is happy now that he is getting four bars on his cell-phone in his home, sweet home.
In summary, beamforming focuses energy in a given direction, thereby improving the signal quality experienced
by the receiver.
Concept of Phase
Before we learn how to form a beam, let's discuss the concept of "phase". Phase essentially dictates how the
signal looks at a given time instant.
Imagine that we just took a picture of this RF signal using our futuristic camera. We have captured three full
cycles of the RF wave. Let's call Wave 1 our Reference Signal. Wave 1 has the wavelength of Lambda, which is
about 1 foot at 900 MHz.
Now, let's make a copy of our reference signal Wave 1 and call this copy Wave 2. Observe the middle cycle in
Wave 1 and Wave 2. Both Wave 1 and Wave 2 look identical and they are said to be in-phase.
Let's add these two in-phase signals and observe what we get. We will get a signal with much higher amplitude!
Hence, when we want to add two signals, we would like those signals to be in-phase.
Now, let's make another copy of our reference signal Wave 1 and call this copy Wave 3. Recall that one full
cycle corresponds to 360 degrees, a half-cycle corresponds to 180 degrees, and a quarter cycle corresponds
to 90 degrees. Now move Wave 3 by 90 degrees. Wave 3 is said to have a phase shift of 90 degrees relative to
Wave 1.
As another example of a phase shift, let's make yet another copy of Wave 1 and call it Wave 4. Let's move
Wave 4 by 180 degrees. Wave 4 is a phase-sifted version of Wave 1 with the phase shift being 180 degrees.
Wave 1 and Wave 4 are said to be out-of-phase.
Let's add these two out-of-phase signals. Our combined signal is now wiped out completely as a positive
portion of one signal cancels out the negative part of another signal! So, we should avoid adding the signals
that are out-of-phase.
Now that we know all about phase, let's see how the concept of phase is used to construct a beam.
Consider a mobile terminal sending an RF signal from its transmit antenna. The base station is using two
receive antennas. Assume that the signal received on Antenna 2 has a phase shift of 90 degrees with respect
to the signal received on Antenna 1. If the base station had only one antenna, the received power would have
been the square of the amplitude of the received signal, which is A squared. Since the base station has two
antennas, we can just add these two signals together. In our example, the 90 degree phase shift at the
receiver results in a combined signal of higher amplitude (> A). This results in a larger received power
compared to the case of just one receive antenna.
Beamforming cannot be achieved by simply adding the received signals. To form a beam, we need to make
sure that the signals that we add are in-phase. In our example, let's add the phase shift of 90 degrees to Signal
1. Now, both Signal 1 and Signal 2 are identical as they are in-phase. When we add these in-phase signals, the
amplitude of the combined signal is 2A. The received power is 4A squared, which is four times the power
corresponding to a single antenna reception. Recall that the simple addition of the received signals without any
phase adjustment gave us the received power of 1.44 A squared, which is much less than 4A squared.
In summary, beamforming is the process of introducing the correct amount of phase shift in the signals
received on different antenna elements so that the manipulated signals are added in-phase. Such processing
increases the overall amplitude or power of the received signal. When the manipulated signals are in-phase,
the receiver is said to be "looking in the direction of the transmitter."
The beamforming scenario that we just described was an example of receive beamforming. Receive
beamforming occurs when the phases of the received signals are adjusted at the receiver before the signals
are combined.
Observe that one transmit antenna is adequate, although two or more antennas are required at the receiver.
Now, the question is, how does the receiver figure out which is the correct amount of phase adjustment? In
one simple implementation, the receiver tries various phase adjustments such as 0 degrees, 45 degrees, 90
degrees, 135 degrees, and 180 degrees, and picks the one corresponding to the maximum power for the
combined signal. In our example, the receiver chooses 90 degrees as it results in the maximum power for the
combined signal.
Transmit Beamforming
The base station transmits the signals from two transmit antennas. Signal 1 is transmitted from Antenna 1 and
has a 90 degree phase shift relative to Signal 2 which is sent from Antenna 2. At the mobile terminal's receive
antenna, two signals arrive in-phase and they automatically add up, leading to a larger power received signal.
How does the transmitter know which is the correct amount of phase adjustment for the transmitted signals?
In one possible implementation, the receiver specifies the phase adjustments to the transmitter.
In our example, the receiver tries different phase adjustments and determines that 90 degrees is a good value.
The receiver provides this phase adjustment feedback to the transmitter. The transmitter then simply follows
the recommendations of the receiver while sending signals from multiple transmit antennas.
In summary, the transmitter adjusts the phases of the transmitted signals in transmit beamforming. The end
result is larger received power compared to the non-beamforming example.
Types of Beamforming
There are two basic ways in which beamforming is exploited by the radio network, a switched-beam system and
an adaptive beamforming system. A switched-beam system constructs fixed beams in a sector, while an
adaptive beamforming system constructs a dynamic beam for a user. First, we will look at the operating
principles of a switched-beam system, and then, we will consider an adaptive beamforming system.
A Switched-Beam System
Here is a base station serving three 120 degree sectors, Sector Alpha, Sector Beta, and Sector Gamma. Now
that we know how to form a beam, let's construct three beams in Sector Beta Beam 1, Beam 2, and Beam 3.
These beams are fixed in space. In other words, they always point in the same direction. Each beam transmits
its own reference signal so that the device can measure the signal quality of a given beam.
Let's imagine that our friend Jeff is giving us a ride in his nice car. While Jeff is driving, we are watching a high
resolution news video on our 4G phone. When we are in the area illuminated by Beam 1, our 4G phone reports
to the base station that the reference signal associated with Beam 1 is the strongest among the three
reference signals. Since Beam 1 is able to provide the best quality signal to our phone, the base station sends
the data packet containing the video in Beam 1. As long as we stay in the same beam, our mobile phone
continues to report the channel quality for Beam 1 and the base station continues to send the data packet
containing the video in Beam 1.
Our friend Jeff can drive fast! So, now we are in the area illuminated by Beam 2. Our 4G phone reports to the
base station that Beam 2 is providing us the best quality reference signal.
Since Beam 2 provides the best quality signal, the base station now sends the data packet containing the
video in Beam 2.
In summary, a switched-beam system communicates with the device in one beam at any given time and, as the
user moves from one beam to the next, the beam carrying the data packet for the user is switched.
Adaptive Beamforming
While beams are fixed in a switched-beam system, the beam follows the user in an adaptive beamforming
system.
Imagine yourself sitting in the middle of a lecture hall, where the presenter is describing advanced antenna
techniques. Further imagine that you have closed your eyes and that you are listening to what the lecturer is
saying! Make sure that you stay awake, though! At this time, the lecturer is in the front of the audience and you
can sense the lecturer's voice coming directly from the front.
Now, the agile lecturer moves to the right. Even with your eyes still closed, you perceive the lecturer's voice
coming from the right. Feel free to open your eyes in your imagination!
Let's transform you into a base station and the lecturer into a mobile device. Just as human ears can focus in a
given direction, the base station can construct the beam in the direction of the user. As the user moves, the
base station follows the user by constructing a new beam in the direction of the user.
In summary, adaptive beamforming constructs the beam in the direction of the user on an on-going basis
rather than using one of a fixed set of beams for the user.
Since adaptive beamforming customizes a beam for a user, it is able to concentrate more energy in the
direction of the user, thereby significantly improving the signal strength. Generally speaking, a switched-beam
system provides a smaller improvement in the signal strength compared to adaptive beamforming.
The main factor that favors one technique over the other is the priority of performance compared to
implementation complexity.
Meet Bill who is watching a high-definition video on his mobile device. Bill's home overlooks a lake and there is
a base station reasonably close to his home. Bill is such a valued customer that the base station forms a beam
toward Bill's multimedia cell phone, yielding excellent signal strength.
Assume that the base station has a maximum of 100 radio channels. The base Station gives all these
channels, f1 through f100, to Bill to make Bill happy. Bill's device is able to get a data rate of about 2 Mbps.
Now, Bill wants to download an email with an attached presentation. This presentation is about the new
software product that his company is about to launch. Now, the base station has a tough problem to solve. If it
divides the data rate of 2 Mbps between two simultaneous applications such as high-definition video and
email, the quality of the video would suffer and it would take a little longer to download the presentation.
Is there anything we can do to the base station so that Bill can continue to watch the video with high quality
while downloading the presentation more quickly? The answer is yes. It is Spatial Multiplexing that will save the
day!
Let's make the base station and Bill's multimedia device more powerful. We now have two antennas at the
base station and two antennas at the mobile device. Furthermore, both the base station and the mobile device
support an advanced antenna technique called spatial multiplexing.
The base station uses all 100 radio channels f1 through f100 to send a video from Antenna 1. Furthermore,
the base station uses the same channels, f1 through f100, to send an email from Antenna 2. The email
application experiences speeds of 2 Mbps and the video application experiences 2 Mbps as well. Bill can now
download the email with the presentation without adversely impacting the quality of the video!
How can the mobile device separate radio channel f1 coming from Antenna 1 and radio channel f1 coming
from Antenna 2? The receiver has to implement algorithms to separate the multiple signals coming from
multiple antennas using the same radio channels. In other words, Spatial Multiplexing enables us to use the
same radio channel on multiple antennas. We will discuss the issue of separating the antenna signals next.
In a radio environment, these flashlights are nothing but the antennas. The way the receiver can separate the
signals from two antennas is by observing the way the radio environment influences these signals.
Signal 1 being transmitted from Antenna 1 is affected by the radio environment in one way, while Signal 2
being transmitted from Antenna 2 is affected by the radio environment in a different way. For example, Signal
1 may not experience any phase shift, while Signal 2 may experience a 180 degree phase shift. Please note
that the base station is not introducing any phase shifts at the transmitter. It is the radio environment that is
changing the signals transmitted by the base station.
Typically, each transmit antenna sends a reference signal over the air. The receiver observes the reference
signals being transmitted from different antennas and learns how these signals are affected by the radio
environment. Then, it applies this channel estimation knowledge to extract data packets being transmitted
from different antennas.
MIMO Terminology
MIMO is defined to be Multiple Input Multiple Output. MIMO facilitates the implementation of Spatial
Multiplexing. Multiple Input indicates the use of multiple transmit antennas and Multiple Output refers to the
use of multiple receive antennas.
Recall that forward link or downlink is the communications link from the base station to the mobile terminal,
while the reverse link or uplink is the communications link from the mobile terminal to the base station.
Let's discuss Downlink (2 x 4) MIMO as an example. The first number, 2 in our example, specifies the number
of transmit antennas and the second number, 4 in our example, gives the number of receive antennas. Since
we are discussing MIMO for the Downlink, the base station is using two transmit antennas and the mobile
station is using four receive antennas. Typically, the number of receive antennas is equal to or greater than the
number of transmit antennas.
When Transmit Antenna 1 of the base station sends out a signal, all four antennas at the mobile terminal
receive the signal. Thus, four receive paths come into existence for one transmitted signal. Similarly, when
Transmit Antenna 2 of the base station sends out a signal, all four antennas at the mobile terminal receive this
signal. So, four more receive paths are formed
The number of transmit antennas in a MIMO system dictates the MIMO throughput gain. If we use (2 x 2)
MIMO for the downlink, each transmit antenna uses the same set of 100 radio channels, leading to 100 + 100
= 200 virtual radio channels. Since we have doubled the number of radio channels from 100 to 200, our
throughput increases by approximately a factor of 2 and becomes 2 Mbps * 2= 4 Mbps.
As another example, if we use (4 x 4) MIMO for the downlink, we will have 4 * 100 = 400 virtual radio channels
instead of just 100 regular radio channels. As long as the receiver can separate out the signals coming from
four different transmit antennas, we can increase throughput by approximately a factor of 4 (i.e., from 2 Mbps
to 8 Mbps).
In summary, the number of transmit antennas dictate how much throughput gain can be achieved relative to
the system consisting of a single transmit antenna and a single receive antenna. Recall from our previous
discussion that the number of receive antennas is typically the same as or greater than the number of transmit
antennas. A system with an equal number of transmit and receive antennas is the most common.
When we send a packet on a wireless link, we add redundancy so that the original packet can be retrieved in
spite of the loss of some bits. In our example, our packet consists of the information bits A, B, C, and so on.
Channel coding such as Convolutional coding or Turbo coding introduces redundancy. For example,
information bit A is represented by the bits A and A' at the output of Channel Coding and information bit B is
represented by the bits B and B' at the output of Channel Coding. The bits coming out of Channel Coding are
typically referred to as code symbols or encoded symbols.
In MIMO transmission, we have two types of encoding approaches based on how code symbols are sent over
the air, vertical encoding and horizontal encoding. We will look at these approaches next.
Vertical Encoding defines a method for sending code symbols from multiple transmit antennas. Vertical
Encoding is also referred to as MIMO-Single Code Word (MIMO-SCW). The input to the Channel Coding block is
a packet with information bits and the output of the Channel Coding block is a code word. The term "Single
Code Word" refers to the fact that only one Channel Coding block is used.
In Vertical Encoding, the set of code symbols that corresponds to the given information bit is transmitted from
different antennas. In our example, information bit A is associated with two code symbols, A and A'. One of
these two code symbols is sent from Antenna 1, and, the other code symbol is sent from Antenna 2. Observe
that these two code symbols are NOT transmitted from the same antenna.
As another example, information bit B corresponds to two code symbols, B and B'. These two code symbols are
again transmitted from two different transmit antennas. The receiver has to gather code symbols across
multiple receive paths to decode the data packet.
Horizontal Encoding defines another method for sending code symbols from multiple transmit antennas.
Horizontal Encoding is also referred to as MIMO-Multi Code Word (MIMO-MCW). Remember that the input to
the Channel Coding block is a packet with information bits, and the output of the Channel Coding block is a
code word. The term "Multi Code Word" implies the use of multiple Channel Coding blocks.
In Horizontal Encoding, multiple packets undergo channel coding in parallel. For example, information bits of
the packet P pass through a channel coding block and information bits of the packet Q pass through another
channel coding block.
Furthermore, the code symbols that correspond to a given packet are transmitted from the same antenna. In
our example, all the code symbols that corresponding to packet P are sent from Antenna 1, and all code
symbols of packet Q are sent from Antenna 2.
The number of processing blocks in a transmitter and a receiver is an important design consideration. If there
are more processing blocks, it would increase the cost and increase the processing power requirements. At the
transmitter, Vertical Encoding requires just one channel coding block, while Horizontal Encoding requires
multiple channel coding blocks. At the receiver, Vertical Encoding requires just one channel decoder, while
Horizontal Encoding requires multiple parallel channel decoders. Hence, Horizontal Encoding is likely to cost
more and would need more processing power.
Recall that spatial diversity involves transmission of the same information from multiple antennas. In Vertical
Encoding, the same information bit is effectively sent from multiple antennas in the form of code symbols.
However, Horizontal Encoding involves the transmission of all code symbols of a packet from the same
antenna. Hence, Vertical Encoding gives more spatial diversity gain compared to Horizontal Encoding.
In the case of Horizontal Encoding, since all code symbols for a given information bit come from the same
antenna, the decoder is relatively simpler. However, in the case of Vertical Encoding, the decoder at the
receiver needs to process the code symbols coming from different antennas to recover a given information bit.
Such processing makes the decoder more complex for Vertical Encoding.
Let's assume that a system has the capability of implementing a (4 x 4) MIMO for the downlink.
Recall that each base station transmit antenna sends out a Reference Signal so that the mobile terminal can
measure the channel conditions between the base station transmit antenna and its receive antenna.
The radio environment is quite dynamic. At a given time instant, one transmit antenna provides a better quality
signal, while another transmit antenna provides a poor quality signal.
At another time instant, the situation may have completely reversed. The antenna previously yielding a poor
quality signal may now provide a better quality signal.
In MIMO with the antenna selection feedback, we exploit variations in channel conditions among various
antennas. The mobile terminal informs the antennas that it prefers. The base station, in turn, uses these
antennas for a packet transmission to the mobile terminal. Let's take a closer look at this technique.
Here is a system that has the capability of implementing a (4 x 4) MIMO for the downlink. However, at a given
time instant, the mobile terminal cannot fully distinguish among the four reference signals. The mobile
terminal learns that it can differentiate just two signals from two antennas under the current channel
conditions.
Let's define a term called "Rank". Rank essentially indicates the number of base station transmit antennas that
the mobile terminal would be able to use reliably under current channel conditions. In our example, the mobile
terminal sends feedback with Rank = 2, indicating to the base station that it can differentiate just two signals
at this time.
Now, the base station sends a data packet to the mobile terminal using just two antennas instead of four
antennas. The mobile terminal is then able to decode the packet correctly.
If the base station had used four transmit antennas to send this packet, the mobile terminal would not have
been able to decode the packet correctly. Note that either Horizontal or Vertical Encoding could be used.
In summary, the rank feedback from the receiver to the transmitter allows the transmitter to select a suitable
number of antennas for a packet transmission.
It is possible to combine Transmit Beamforming with Spatial Multiplexing. The result is Closed Loop MIMO with
precoding. The term "closed loop" indicates that there is some feedback from the receiver to the transmitter.
The term "MIMO" indicates that Spatial Multiplexing is used. The term "Precoding" refers to the phase
adjustments for beamforming.
Transmit Beamforming involves phase adjustments at the transmit antennas so that the receiver experiences
a significant increase in the received signal strength.
Spatial Multiplexing uses the same radio resources on multiple antennas to send different streams of bits,
which significantly increases the throughput.
The Closed Loop MIMO with precoding combines the beamforming technique with the Spatial Multiplexing
technique to get best of both techniques.
Let's consider a downlink (4x4) MIMO system. The mobile terminal observes the reference signals being
transmitted from different antennas so that it can determine the channel quality of the paths between the
transmit and receive antennas.
The mobile terminal then sends feedback to the base station. This feedback can include (i) the rank for Spatial
Multiplexing and (ii) the suggested phase shifts for beamforming. Remember that the rank specifies the
number of transmit antennas that the receiver can utilize reliably. In our example, the mobile terminal
recommends the use of two transmit antennas instead of four transmit antennas and it suggests a phase shift
of 90 degrees between the two antenna signals.
The base station now forms a beam in the direction of the mobile terminal and sends two parallel streams of
data using two antennas to implement spatial multiplexing.
In summary, closed loop MIMO with precoding combines beamforming with spatial multiplexing.
Consider a traditional adaptive beamforming scenario, where the base station has constructed two beams for
two users, Bill and Warren. Bill and Warren are in Sector Beta. Sector Beta has a total of 100 radio channels
for data transmission and it has allocated Bill and Warren 50 radio channels each. Bill and Warren are given
different radio channels so that the channels for Bill do not interfere with the channels for Warren and vice
versa. Assume that Sector Beta is able to support 5 Mbps for both, yielding a total of 10 Mbps sector
throughput.
Let's try to double the sector throughput using the concept of Space or Spatial Division Multiple Access.
Instead of distributing 100 radio channels of Sector Beta between Bill and Warren, let's use all 100 radio
channels for Bill in Beam 1 and let's use the same 100 radio channels for Warren in Beam 2. Beam 1 and
Beam 2 are now SDMA beams, with each SDMA beam using the same radio channels. Now, each SDMA beam
has access to 100 radio channels and each SDMA beam can yield 10 Mbps. Now we have two SDMA beams,
the total sector throughput is 2 * 10 Mbps = 20 Mbps! We just doubled the sector throughput by using two
beams.
Please keep in mind that phase adjustments are used to construct a beam. The only difference between
regular beams and SDMA beams is that two regular beams in a sector use different radio channels and two
SDMA beams in a sector use the same radio channels.
The signals for Bill and Warren are now separated just in space and not in radio channels. Hence, such
systems are called Space Division Multiple Access.
If SDMA is so good at increasing throughput, why not use it all the time? Well, the use of the same radio
channels in a given sector significantly increases the likelihood of interference, especially when the SDMA
beams are not formed properly.
Let's first construct two SDMA beams in Sector Beta and re-use all 100 radio channels in each SDMA beam for
Bill and Warren. Currently, we have 10 Mbps for Bill, 10 Mbps for Warren, and a sector throughput of 20 Mbps.
Let's add the power of Spatial Multiplexing by using (4x4) MIMO in the downlink for both Bill and Warren. Each
of the four transmit antennas provides 10 Mbps to Bill, giving Bill throughput of 40 Mbps. Similarly, Warren
experiences 40 Mbps. The total sector throughput becomes 80 Mbps! Bill and Warren's devices need to be
capable of supporting such data rates. The cost of the network also goes up due to increased complexity.
In summary, SDMA allows different SDMA beams within a sector to re-use the same radio channels, while
MIMO Spatial Multiplexing allows the re-use of the same radio channels on different antennas between the
base station and the mobile terminal.
The goal of Collaborative Spatial Multiplexing is to increase uplink sector throughput with just one transmit
antenna at the mobile terminal. Note that the cost of the mobile terminal is kept low in this case. If we start
using multiple transmit antennas at the mobile terminal, we would then need to worry about the cost of the
antennas, transmit power requirements, processing power, and complexity. To avoid all of these issues and to
keep the cost of the device low, we typically use only one transmit antenna at the mobile device.
Consider a system, where we want to implement uplink (2x2) MIMO with Collaborative Spatial Multiplexing. To
have two transmit antennas in the uplink, we will use one antenna from one mobile terminal and another
antenna from another mobile terminal. The base station uses two receive antennas to receive signals from two
transmit antennas. The basic operation of uplink MIMO is similar to that for downlink MIMO. Basically, uplink
MIMO is like downlink MIMO-MCW (MIMO-Multiple Code Word) or MIMO with Horizontal Encoding, where one
transmit antenna sends out one code word, and another transmit antenna sends out another code word.
If a system with one transmit antenna and one receive antenna yields sector throughput of 10 Mbps in the
uplink, uplink (2x2) MIMO with Collaborative Spatial Multiplexing can provide sector throughout of 20 Mbps in
the uplink. Remember that the number of transmit antennas dictates the throughput in MIMO.
In summary, Collaborative Spatial Multiplexing extends the concept of MIMO to uplink with just one transmit
antenna at the mobile terminal. Compared to non-MIMO, this approach significantly increases the uplink sector
throughput.
6.0 Acronyms
2
2G Second Generation Wireless Systems
3
3G Third Generation Wireless Systems
4
4G Fourth Generation Wireless Systems
F
FHDC Frequency Hopping Diversity Coding
M
Mbps Megabits per second
MHz Megahertz
MIMO Multiple Input Multiple Output
MIMO-MCW MIMO-Multi Code Word
MIMO-SCW MIMO-Single Code Word
R
RACH RF Radio Frequency
S
SDMA Space (or Spatial) Division Multiple Access
SIMO Single Input Multiple Output
STC Space Time Coding
STTD Space Time Transmit Diversity