White Paper

Designed for Speed

Network Infrastructure in an 802.11n World

Peter Thornycroft
May 2007 V2.7
1 Preparing for the 802.11n adoption wave ............................................................................................................... 2
1.1 Summary........................................................................................................................................................ 2
1.1.1 Introduction................................................................................................................................ 2
1.1.2 Benefits for the Enterprise ......................................................................................................... 2
1.1.3 Benefits for the user .................................................................................................................. 3
1.1.4 Reasons these benefits will initially be elusive .......................................................................... 3
1.1.5 802.11n migration strategies for enterprises ............................................................................. 4
1.2 Network design with 802.11n........................................................................................................................ 4
1.2.1 New hardware for APs and clients ............................................................................................ 4
1.2.2 RF planning & AP placement .................................................................................................... 5
1.2.3 Wired backhaul from APs & LAN implications ........................................................................... 5
1.2.4 Implications for rogue APs and IDS........................................................................................... 6
1.2.5 Good-neighbour (or bad-neighbour) strategies ......................................................................... 6
2 Technology in 802.11n............................................................................................................................................ 7
2.1 Techniques for high-throughput PHY ........................................................................................................... 7
2.1.1 High Throughput PHY: MIMO................................................................................................... 7
2.1.2 High throughput PHY: transmit beamforming (TxBF) ............................................................. 11
2.1.3 MIMO, SDM, beamforming… terminology and techniques ..................................................... 12
2.1.4 802.11n MIMO configurations and terminology ....................................................................... 12
2.1.5 High Throughput PHY: 40 MHz channels............................................................................... 13
2.1.6 High Throughput PHY: Shorter guard interval ........................................................................ 13
2.1.7 High Throughput PHY: More subcarriers ............................................................................... 14
2.1.8 High Throughput PHY: New Modulation Rates ...................................................................... 14
2.1.9 High Throughput PHY: Duplicate format ................................................................................ 16
2.2 Techniques to enhance the MAC................................................................................................................. 17
2.2.1 MAC layer enhancements: Frame aggregation ...................................................................... 17
2.2.2 MAC layer enhancements: Multiple Traffic ID Block Acknowledgement (MTBA)................... 18
2.2.3 MAC layer enhancements: Reduced inter-frame spacing (RIFS)........................................... 18
2.2.4 MAC layer enhancements: Spatial multiplexing power save (SM power save)...................... 19
2.2.5 MAC layer enhancements: Power Save Multi-poll (PSMP) .................................................... 19
2.3 Compatibility Modes and Legacy Support in 802.11n ................................................................................ 20
2.3.1 Phased coexistence Operation (PCO) .................................................................................... 22
2.3.2 Other mechanisms for coexistence: Dual-CTS protection (CTS-to-self) ................................ 23
2.3.3 Other mechanisms for coexistence: 40 MHz-intolerant indication.......................................... 24
2.3.4 Using 802.11n in the 2.4 GHz band ........................................................................................ 24
3 Product development and capabilities................................................................................................................... 26
3.1 Standards and Certification framework ....................................................................................................... 26
3.2 Expected product introduction timelines ..................................................................................................... 27
4 Migration strategies............................................................................................................................................... 28
4.1 Different paths to enterprise-wide 802.11n ................................................................................................. 28
4.1.1 Greenfield ................................................................................................................................ 29
4.1.2 Summary of ‘Greenfield’ recommendations: ........................................................................... 30
4.1.3 AP-overlay ............................................................................................................................... 30
4.1.4 Summary of ‘Overlay’ recommendations:................................................................................ 31
4.1.5 AP substitution......................................................................................................................... 32
4.1.6 Summary of ‘AP substitution’ recommendations: .................................................................... 32
4.2 Other considerations when planning an upgrade......................................................................................... 32
5 Conclusion ............................................................................................................................................................ 33
6 Appendix............................................................................................................................................................... 34
6.1 Note on expected ‘real-world’ cell capacity with 802.11n .......................................................................... 34
6.2 Note on expected ‘real-world’ data rates with 802.11n............................................................................... 35
6.3 Glossary of terms used in this note.............................................................................................................. 41

1 802.11n Theory and Practice Aruba Networks

1 Preparing for the 802.11n adoption wave

1.1 Summary
Welcome to the new world of 802.11n. And forget everything you thought you knew about wireless. With
this technology, you often want to avoid ‘line of sight’: it actually works better when there are signal
reflections in the radio path from transmitter to receiver; this means performance often improves as
distance increases. 802.11n will bring higher data rates, longer range and more reliable coverage than
previous Wi-Fi technology; it represents a significant upgrade in performance.

However, the migration to 802.11n poses some challenges. Most of the benefits will only be realized
when 802.11n-capable clients are used with similar infrastructure, and even a few legacy (802.11a/b/g)
clients in the cell will drastically reduce overall performance compared to a uniform 802.11n network.
Also, the wiring to the access points will require upgrading for gigabit Ethernet, and edge switch ports
must also accommodate this protocol and probably the new 802.3at power-over-Ethernet standard.
Migration strategies will require careful planning and patience.

1.1.1 Introduction
Wi-Fi technology has traced a curve of ever-increasing performance from the earliest pre-802.11
standards through 802.11b to 802.11a/g, with peak data rates rising from 2Mbps to 54Mbps. The latest
set of innovations is a package known as 802.11n. This standard is still in development at the IEEE, but
‘pre-n’ silicon has been shipping for some months, a ‘draft-n’ certification will be available soon, and
Enterprise infrastructure vendors will be introducing 802.11n access points in the first half of 2007. This
note is intended to explain the important technology in 802.11n, but also to give non-specialists sufficient
information to plot their own upgrade strategy.

1.1.2 Benefits for the Enterprise

802.11n includes a number of technological advances. They are explained in detail later in this paper, and
some of them are complicated. However, a focus on complex technology can distract from the overall
concepts and implications for network design. These concepts can be explained relatively easily.

• Increased capacity. 802.11n enables increased data rates, improving the capacity of a cell from
perhaps 15-20Mbps with 802.11a/g to 100-200Mbps (see note below). Given that this will be
spread over a number of simultaneous users, one can say that performance should be very close
to that of a wired 100Mbps Ethernet connection, the standard for desktop connectivity.

• Improved range. An 802.11g connection from AP to client can usefully extend at best 60 metres in
open, unobstructed areas but 20 metres in office environments. 802.11n will increase this, but for
a significant improvement, both the access point and the client should support at least two ‘spatial
streams’ for 802.11n. In some parts of the network this will lead to simpler design, with larger cells
and fewer APs required to cover a given area: we expect this to be applied especially in areas with
low client densities.

• More uniform coverage (‘reliable’ coverage). Coverage in Wi-Fi networks is notoriously spotty. A
user may have a good signal, but moving the client a short distance, stepping in front of it, or
opening and closing of a door across the room can significantly affect the signal strength received,
and hence the performance. The best technology to counter this today is antenna diversity –
nearly every Wi-Fi device sports two antennas, and switches between them so when one is in a
multipath null, the other should still have a workable signal. 802.11n uses more sophisticated
technology that should much reduce these effects.

Aruba Networks 802.11n Theory and Practice 2

 • Lower network costs. This is a consequence of improved range and more uniform coverage. APs
can now be further apart. This reduces costs in a number of ways: fewer APs, lower installation
costs, possibly fewer edge-switch ports, and fewer outdoor APs to cover campus areas between
buildings. Against this, 802.11n APs will initially be more expensive than older Wi-Fi, but this
premium will be eroded over time.

1.1.3 Benefits for the user

The user also benefits from higher data rates, higher throughput and more uniform coverage. For a given
number of users in a cell, each will now have access to more bandwidth. The WLAN will no longer be the
limiting factor for application performance, and a significant obstacle to un-wiring the desktop is overcome.
Video, both interactive and broadcast, is an application that will benefit from 802.11n. Many organizations
are beginning to use video broadcasts as an internal communications tool, and while earlier 802.11
networks can support it, the demands for bandwidth and contention are sufficient to stress current
installations. 802.11n provides more bandwidth, but also incorporates many features to optimize video
performance.

1.1.4 Reasons these benefits will initially be elusive

Unfortunately, there are many reasons it will be very difficult for Enterprises to realize the expected
benefits of 802.11n in the short-term:

The presence of legacy clients negates 802.11n advantages. The benefits above are easy to achieve in a
‘greenfield’ network, where all APs and all clients are 802.11n-capable. However, most Enterprises
upgrade PCs (the most common Wi-Fi client) over a cycle of 2-3 years or longer. As long as older clients
exist, they will affect the performance of the network by connecting at much lower rates than 802.11n
clients, effectively slowing cells to near-802.11a/g rates. The overall throughput falls even below the
average of the rates: a client connecting at 6Mbps and sending 4Mbps of traffic effectively takes nearly all
the bandwidth available, even if other clients are using 600Mbps rates. As long as even a few legacy
clients exist, the expected capacity improvements are unlikely to be realized. Legacy devices are also
unable to take advantage of improvements in range and uniformity of coverage: if 802.11n APs are
spaced farther apart to lower costs, legacy clients will likely run into more coverage problems than before.

The first paragraph of this paper suggested that moving to 802.11n will drive equipment upgrades in other
parts of the network. While 802.11n requires new access point hardware incorporating new silicon, and
gigabit Ethernet and power-over-Ethernet upgrades will often be required for wiring and edge switches,
the WLAN controllers that are used by state-of-the-art centralised architectures should be considered
vendor-by-vendor. For a controller to be upgraded for 802.11n, it must support the new radio and other
features of the technology from a technology perspective, and this is not likely to be a problem. The
increased data (certainly peak data rates) generated by an 802.11n WLAN may challenge some current
controllers using general-purpose computing platforms. (Aruba’s WLAN mobility controllers will not require
upgrading to support 802.11n data rates, as they already incorporate hardware-acceleration for the data
path.)

There is already a wave of pre-standard implementations. The Wi-Fi industry is technology-driven and
consumer-led. Since consumers are less sensitive to standards than Enterprises (‘So long as the handful
of devices in my living room work together, I have compatibility - and I can ensure this by using one
vendor’s products everywhere.’). Thus there are already first and second waves of ‘pre-n’ silicon and
consumer products on the markets, and major vendors are preparing to build-in pre-n support into mass-
market PCs. This is helps move the new technology into the field, but it is pre-empting the standards
process, which is not yet complete. There is no guarantee that pre-n products shipped today will be fully-

3 802.11n Theory and Practice Aruba Networks

compatible with the eventual standard. There is no concern that such mixed networks will not work, but
they may not prove capable of realizing the full benefits listed above. Indeed, the Wi-Fi Alliance, an
industry group, is developing an ‘official’ ‘draft-n’ standard, which will add a measure of predictability to
these issues, but risks further confusing customers.

Thus, while the eventual benefits are easy to see, it may prove more difficult to ‘get there from here’. It is
the migration plan that requires careful consideration.

1.1.5 802.11n migration strategies for enterprises

802.11n includes a great variety of mechanisms for successful coexistence with legacy 802.11 networks.
As noted above, these operating modes negate many of the performance advantages of a greenfield
802.11n network, but they offer complete support for older 802.11 clients. Thus, 802.11n access points
can be installed before any 802.11n clients are present, and operate in 802.11a/b/g mode. It will be
possible to mix 802.11n access points within 802.11a/b/g network, replacing selected existing access
points, or deploying new 802.11n access points as an overlay in parallel with an existing 802.11 network:
the section on migration strategies gives more thoughts on this.

Since the IEEE will not finalize its technical specification until 2008, the only short-term option for network
managers wishing to move to 802.11n will be the ‘draft-n’ products referred to above. For enterprises
looking to test early 802.11n products, we would recommend starting with ‘official pre-n’ products that
have the Wi-Fi Alliance ‘draft-n’ certification (we expect this certification to be available in the second or
third quarter of 2007). These will certainly offer good legacy 802.11 support, but it is currently unknown
how many features of the ‘official draft-n’ products will be incompatible with the final 802.11n standard.
Our feeling at Aruba is that such differences are likely to be minor, but facts will only become apparent
with the passage of time. We would certainly endorse a strategy of installing pre-802.11n access points
and testing them with draft-802.11n clients in a pilot or trial of the technology, and for some enterprises a
more widespread deployment may be appropriate, after a consideration of the technology risks.

Finally, before embarking on a technology migration it is useful to assess the requirements and drivers.
The most compelling reason to move to 802.11n is to provide greater bandwidth for users. This reason is
valid: 802.11n will provide increased raw data rates, but for many enterprises there is limited need to
provide greater bandwidth on the wireless network: most current users’ performance is not limited by the
WLAN, but by other bottlenecks in the application-delivery network. For those current 802.11 networks
where performance is an issue, opportunities exist to increase performance by making more use of the 5
GHz band (in similar fashion to a practical 802.11n network) and by providing more access points,
reducing the size of the 802.11 cells and increasing the capacity of each cell even while it serves fewer
users. In some WLAN architectures with distributed encryption, the encrypted throughput of the access
point may become a bottleneck; the section above has a brief discussion of the need (or otherwise) to
upgrade WLAN controllers to accommodate increased network bandwidth. These techniques for
designing high-performance 802.11 enterprise WLANs have been refined by current vendors over a
number of years. While the overall network efficiency and performance will fall short of a ‘greenfield’
uniform 802.11n network, such a mid-life upgrade can provide the desired performance improvements for
another 12-24 months, by which time the technology risks of moving to 802.11n will have abated.

1.2 Network design with 802.11n

1.2.1 New hardware for APs and clients
New hardware will be required to support 802.11n. The radio characteristics are different, old silicon
cannot support the new protocols and an AP needs multiple RF chains and antennas to support the new
standard.

Aruba Networks 802.11n Theory and Practice 4

1.2.2 RF planning & AP placement
For RF planning purposes, 802.11n differs from 802.11a/g in utilizing MIMO and the option of a 40MHz
channel., which provides approximately twice the data rate of a 20 MHz channel, but uses twice the RF
spectrum. Because the 2.4 GHz band has a limited number of channels, even at 20 MHz, the 40 MHz
option is unlikely in this band, although the many other enhancements in 802.11n will still provide
significant performance improvements. However, the 40 MHz channel will be popular in the 5 GHz band
where channels are plentiful.

Most RF planning tools rely on modeling the reduction of signal strength over distance and across through
RF obstructions. For most purposes, this reduces to a simple calculation of dB loss at a given distance
from the AP, and knowing the transmit power, the data rate achievable at a given distance from the AP
can be easily calculated. To plan an Aruba network, the required inputs are the dimensions of the
floorplan, the minimum data rate desired and the cell overlap factor. Since the propagation characteristics
for 802.11n signals should be no different from 802.11a/g, existing RF planning tools such as this will need
little modification.

The significant extra factors that need to be considered are the number of antennas on both AP and client,
and whether the design should account for legacy clients.

The output of the RF planning tool will be as before, a ‘grid’ of suggested AP locations that satisfy the
input parameters. A benefit of 802.11n should be that a greenfield design with 802.11n clients can be built
with fewer APs, but still give better performance than an equivalent 802.11a/g design.

The type, placement and attitude of the antennas, particularly on the AP, may be important to 802.11n
performance; in general, Enterprise office deployments will use the captive antennas supplied on the AP,
and they should be deployed vertically. Spacing the antennas by at least half a wavelength (6.25cm for
2.4GHz or 2.7cm for 5.5GHz) will give better performance.

1.2.3 Wired backhaul from APs & LAN implications

With 802.11a/g, the maximum data rate on the air is 54Mbps, and actual, achievable data throughput tops
out at about 36Mbps: easily supported on a 10/100Mbps Ethernet connection. 802.11n rates, however,
can move well beyond 100Mbps, at least for peak traffic. This is too much for a single connection, so
vendors have three options:

• Use multiple 10/100 Ethernet (probably two) connections to the AP. Thus two drops and two edge
switch ports will be needed for each AP location.
• Set up with one 10/100 Ethernet, and accept that at peak loads this may not be enough. But in
many cases these peaks will be widely spaced, because clients are unlikely to connect at data
rates much higher than this in practice.
• Gigabit Ethernet. This comprehensively carries the traffic, but may require that cabling is upgraded
to category 5e or category 6, and edge switches provide GE ports. Vendors are most likely to
provide this option (GE ports) on new 802.11n APs.

Other areas of the LAN should be checked for traffic capacity but will probably not require augmentation.
The traffic on the upstream connection from the edge switch still represents the same aggregate number
of users and applications as when clients were all wired.

The power consumption of 802.11n APs is likely to be greater than for 802.11a/g APs. The power draw is
likely to exceed 802.3af (12.95W maximum can be delivered to a Class 3 device under 802.11af ) which
means the edge switches or in-line power injectors must support the (unratified) 802.3at which provides
phantom power over all four pairs, rather than just two as in 802.3af. Alternatively, the AP must use a
local power brick.

5 802.11n Theory and Practice Aruba Networks

1.2.4 Implications for rogue APs and IDS
Aruba expects the earliest deployments of 802.11n, apart from limited pilot trials, may be for RF
monitoring purposes. This is because we expect the ‘rogue AP’ problem to reappear, as employees use
802.11n consumer products at home, find they work and bring them to the office. The only way to reliably
identify these devices is to use another 802.11n device: an 802.11n AP in monitor mode.

Once 802.11n is deployed, Intrusion Detection Services will need some enhancements to identify
characteristics of 802.11n-specific attacks, but as most attacks are at higher levels than the PHY, the
current IDS functionality will be portable to 802.11n.

1.2.5 Good-neighbour (or bad-neighbour) strategies

802.11n networks will often overlap with other 802.11 networks. In some cases the network manager will
wish to cooperate with these networks, in others it may be better to use non-coexisting modes, supporting
higher throughput, and to ignore occasional interference. 802.11n allows a number of strategies.

Several forms of interference are possible:

• If an 802.11n network is operating in ‘greenfield’ mode, its transmissions will not be

comprehensible to 802.11a/b/g APs and clients. Therefore, even though neighbouring cells may
be operating in the same RF channel, the transmissions from one cell (AP and clients) will appear
as noise bursts to the other. This interference will work in both directions: 802.11n transmissions
are likely to be affected as much as the legacy network.
• An 802.11n network using a 40 MHz channel will have just this effect in the 5 GHz band. In 2.4
GHz, because the channel boundaries of 802.11n do not line up with the traditional channels used
in 802.11b/g (channels 1, 6, 11), transmissions will inevitably overlap in frequency. This is the
reason such 40 MHz operation is not recommended. However, it is quite reasonable to use
802.11n in 20 MHz channels in the 2.4 GHz band.
• An 802.11n network operating in mixed-mode is entirely compatible with overlapping 802.11a/b/g
cells. In this mode, all transmissions are prefixed by a preamble that uses legacy modulation,
allowing such networks to coexist.
• Another optional mechanism, Phased coexistence Operation (PCO) allows an AP to alternate
between a 40 MHz channel and each of the constituent 20 MHz channels. The AP sets the
Network Availability Vector (NAV) on the 20 MHz channels to inhibit transmissions for a time,
during which it and whatever 40 MHz-capable clients it supports switch to 20 MHz operation.

Aruba Networks 802.11n Theory and Practice 6

2 Technology in 802.11n
There are several key technologies and features in the 802.11n standard. This section includes a brief but
technical treatment of all significant technologies, including:

• High throughput PHY (physical layer): MIMO. The new PHY supports OFDM modulation with
additional coding methods, preambles, multiple streams and beam-forming. These can support
higher data rates, and a much larger range of data rates than earlier 802.11 standards. The MIMO
technique that is synonymous with 802.11n belongs in this section.
• High throughput PHY: 40 MHz channels. Two adjacent 20 MHz channels are combined to create a
single 40 MHz channel. This relatively simple technique (already used in some point-to-point
bridges and consumer equipment) more than doubles the effective data rate under a given set of
RF conditions.
• Efficient MAC: MAC aggregation. Two MAC aggregation methods are supported to efficiently
pack smaller packets into a larger frame. This reduces the number of frames on the air, and
reduces the time lost to contention for the medium, improving overall throughput.
• Efficient MAC: Block Acknowledgement. Particularly for streaming traffic such as video, a
performance optimization where one acknowledgement can cover many transmitted frames: an
ack is no longer required for every frame. This technique was first introduced in 802.11e.
• Power Saving: power save multi-poll. This is an extension of the U-APSD and S-APSD concepts
introduced in 802.11e.

Each of these is considered in more depth:

2.1 Techniques for high-throughput PHY

2.1.1 High Throughput PHY: MIMO
802.11n achieves significant improvements in performance through the use of MIMO (Multiple Input,
Multiple Output). In MIMO, the transmitting and receiving stations each have multiple RF chains with
multiple antennas. We will consider a MIMO system with two transmitting and two receiving antennas
(2:2).

2
A
B
1

Baseband Tx Rx Baseband
MAC MAC
signal signal
etc etc
processing Tx Rx processing

Each antenna is connected to its own RF chain for transmit and receive: this is architecturally important,
as it allows signals received at each antenna to be decoded independently. The baseband processing on
the transmit side can synthesize different signals to send to each antenna, while at the receiver the signals
from different antennas can be decoded individually. (We will simplify this explanation by showing only
one direction of transmission and a 2x2 MIMO example, practical systems will transmit in both directions
and may use up to 4 antennas at each station.)

7 802.11n Theory and Practice Aruba Networks

Under normal, line of sight conditions, the receiving antennas all ‘hear’ the same signal from the
transmitter. Even if the receiver uses sophisticated techniques to separate the signals heard at antennas
1 and 2, it is left with the same data. If the transmitter attempts to send different signals to antennas A and
B, those signals will arrive simultaneously at the receiver, and will effectively interfere with each other.
There is no way under these conditions to better the performance of a non-MIMO system: one might as
well use only one antenna at each station.

2
A
B
1

(Nearly all 802.11 stations built to date actually use two antennas. However, this is not MIMO – the radio
switches a single radio chip from one antenna to the other, so only one is used at any time. Using two
antennas in this way helps to negate the effects of multipath, as when one antenna is in a multipath ‘null’,
the other is likely to have a better signal. It is generally reckoned that using antenna diversity in this way
improves overall reception by perhaps 3-6 dB, although the effect is of course statistical. MIMO is
different in that both antennas are driven by or receiving signals at all times, and those transmitted signals
need not be identical.)

2
A
B
1

However, if there is sufficient RF distortion in the path, receiving antennas will see different signals from
each transmit antenna. This distortion of the RF channel is extremely complicated, but for our purposes it
amounts to multipath reflections. The transmit antenna radiates a signal over a broad arc, and it reflects
off various objects in the surrounding area. Each reflection entails a loss of signal power, and the longer
the reflected path, the more delay is introduced relative to a line-of-sight signal. In the past, multipath has
been the enemy of radio systems, as the receiver sees a dominant signal (line of sight if it is present), and
all the multipath signals tend to interfere with it, effectively acting as noise or interference and reducing the
overall throughput of the system. Multipath effects also change over time, as objects in the path move,
and movement of reflecting objects results in a Doppler shift of the frequency of the received signal,
further complicating the control mechanisms needed to counter multipath.

To understand how MIMO works, first consider the signal each receive antenna sees in a multipath
environment.

Aruba Networks 802.11n Theory and Practice 8

 b c
a
2
A
B
Received power level 1

Dominant signal at Rx antenna 1 is

a from path a. Path b, c and other
multipath signals cause some
b degradation to the dominant signal
– a similar effect to higher
background noise.
c
time

Total throughput (per unit time) from transmitter to receiver = a - b

In this example there are 3 multipath signals arriving at antenna 2. The strongest is signal a, and the
information carried in this signal will be decoded. Other multipath signals will arrive at lower power levels,
and they are likely to be time-shifted (or phase-shifted) compared to a, so it is likely they will degrade the
overall signal-to-noise ratio associated with a.

B2
A2
2
A B1
B A1
1

Dominant signal at Rx antenna

A1 1 is from Tx antenna A (B1
Received power level

signal and other multipath

B1 signals cause some degradation)

B2 Dominant signal at Rx antenna

2 is from Tx antenna B (A2
signal and other multipath
A2 signals cause some degradation)

time

Total throughput (per unit time) from transmitter to receiver = +

When multiple antennas are considered, however, MIMO offers considerable gains in throughput. The
example above shows that each receive antenna receives its dominant signal from a different transmit
antenna: receiver 1 uses transmitter A while receiver 2 uses transmitter B. When the system
understands, it can take advantage by transmitting different signals from each antenna, knowing each will
be received with little interference from the other. Herein lies the genius of MIMO.

9 802.11n Theory and Practice Aruba Networks

 V11 U11

V21 U21
S1 in S1 out

MAC signal signal MAC

V12 U12
etc processing S2 in S2 out processing etc

V22 U22

The diagram above shows a more detailed explanation of MIMO implementation. At the transmit side,
signal processing provides real and imaginary outputs S1in and S2in. These are then mixed with different
weights V11 etc, before the signals are combined and delivered to the transmit antennas. A similar mixing
function processes signals from the receive antennas using weights U11 etc. Provided the RF
characteristics are known, the weights V11 … and U11… can be calculated and set for optimum throughput,
given the RF channel conditions.

The most favourable case would be where each transmit-receive pair operates with a completely
independent RF path: a 2x2 (2 antennas at each station) system will have double the throughput of a
single-antenna 1x1 system, and a 3x3 configuration could extend to triple the throughput. 802.11n defines
MIMO configurations from 2:1 to 4x4 antennas and spatial streams.

MIMO is the most difficult aspect of 802.11n to understand: multipath (reflected RF between transmitter
and receiver) is normally the enemy of performance, but with MIMO it can be used constructively. Line of
sight normally gives the best performance, but with MIMO it provides just baseline data rates. (Note,
however, that reflected signals are usually much weaker than primary, line-of-sight signals. Even though
losing line-of-sight may allow use of more RF paths and hence the additive MIMO effect, the signal-to-
noise ratio of each path may be considerably worse than previously. It is difficult to predict the relative
weight of these two opposing effects without extensive field testing.)

As noted above, MIMO works best when antennas are positioned more than half a wavelength apart. For
5.5 GHz, half a wavelength is about 2.7 cm or one inch. The 802.11n standard mandates at least two
spatial streams (antennas) for access points and one spatial stream for client devices, with a maximum of
4 spatial streams per device.

One key question in MIMO systems is how to tune the transmit signals at different antennas for optimum
reception at the receiver. 802.11n offers two methods for this. Implicit feedback requires the MIMO
receiver to transmit long training symbols, which are received by the MIMO transmitter. There is an
assumption that the RF channel’s characteristics are reciprocal, and that the transmit and receive radios at
each end are identical, and under these conditions the measurements will be valid in the MIMO transmit
direction.

Implicit feedback is not held to be a promising method in practice, so a second, explicit feedback
mechanism has been defined. The MIMO transmitter sends a long series of training symbols which the
MIMO receiver analyses so as to characterize the differential path loss and delay for each antenna and
spatial stream. It then sends this information back to the transmitter which can optimize its transmit
signals and modulation for the RF path characteristics.

This is a complicated procedure, as many calculations must be performed and RF conditions may be
changing rapidly, requiring continuous adjustment. Since 802.11n uses OFDM, each symbol is modulated
around many carriers rather than only a single carrier, and each carrier should be measured for each

Aruba Networks 802.11n Theory and Practice 10

transmit-receive antenna pair for full characterization. But explicit feedback (MRQ) is a key mechanism in
enabling MIMO, and it has been shown to work in plugfests and in the field.

2.1.2 High throughput PHY: transmit beamforming (TxBF)

Transmit beamforming is a technique that has been used in radio for many years, and has been applied to
earlier forms of Wi-Fi by some startup companies in recent years. In 802.11n, beamforming will allow an
AP to ‘focus’ its transmission to a particular client in the direction of that client (and vice versa for a client
with multiple antennas), allowing higher signal to noise ratios and hence higher data rates than would
otherwise be the case. Beamforming APs can target multiple clients, but only for unicast frames.

C
B
Normal isotropic
Directional pattern after
radiation pattern
beamforming
Transmitted power level

Received power level

A A

tca
B B

tcb
C C

time time
Identical RF signals are transmitted Transmit timing offsets are This results in constructive
from each antenna, but very slightly calculated to match the path delays interference, and a higher
offset in time (phase) of the different RF beams, so all signal-to-noise ratio at the
signals are directed at the target in- receive antenna
phase

By carefully controlling the time (or phase) of the signal transmitted from multiple antennas, it is possible to
shape the overall pattern of the received signal, emulating a higher-gain, or directional antenna in the
direction of the target. The same implicit and explicit feedback mechanisms used to characterize the
MIMO channel allow beamforming.

In practice, beamforming may be used when MIMO with SDM is not feasible, as in line-of-sight conditions.
This is because beamforming aims to produce a single, coherent RF signal at the receiver, while SDM
relies on multiple, independent signals. (Beamforming and MIMO SDM are in fact similar techniques:
MIMO has been described as beamforming to steer antenna pattern nulls to different receive antennas.)

Beamforming is a part of 802.11n that is not fully-agreed as of April 2007, so it will probably not be part of
the Wi-Fi Alliance certifications. Beamforming will only work well with 802.11n clients, as it relies heavily
on feedback messages from the client.

11 802.11n Theory and Practice Aruba Networks

2.1.3 MIMO, SDM, beamforming… terminology and techniques
The 802.11n standard uses the terms ‘MIMO’ and ‘SDM’ (spatial division multiplexing). For our purposes,
these refer to the same techniques, as discussed above: SDM provides a MIMO system with its superior
data throughput.

The figures below explain the difference between MIMO/SDM, transmit beamforming and transmit
antenna diversity. These are all techniques defined in 802.11n, although antenna diversity has been used
prior to 802.11n.

P1a
2
P1a A P1a
P1 B P1b P1
P1b 1 P1b

MIMO with Spatial Diversity Multiplexing (SDM) and Space Time Block Coding (STBC). Independent paths between pairs
of antennas allow data transmission in parallel: data packets (P1) are interleaved and mapped to different paths,
where they may be encoded at a different data rate for each path, depending on RF conditions. The receiver
interleaver re-builds the original packet.

P1 2
A P1
P1 P1
B
1
Transmit beamforming. The transmitter sends a single stream of data, adjusting the signal from each antenna to
ensure the optimal signal forms at the receiving antenna. This is used where there is little RF separation between the
different inter-antenna paths, so SDM is not useful.

2
A P1
P1 P1
B
1
Receive antenna spatial diversity. Working on only one transmitted signal, the receiver can use RF combining
techniques on signals from different receive antennas to achieve higher signal-to-noise ratios and higher data rates.

P1a
2 3
P1a P1a
A P1b
P1 P1
P1b
B P1b
C 1

SDM combined with antenna diversity. In this example, the path between the A-3 antenna pair has different RF
characteristics from the other antenna pairs: it offers RF diversity, and carries one spatial stream. The other inter-
antenna paths, B/C to 1/2, are not RF-isolated, so they cannot only be used for one further spatial stream. However,
transmit beamforming or receive antenna diversity may be used to optimise this spatial stream. In this case a system
with 3 transmit and 3 receive antennas nevertheless supports only 2 spatial streams.

2.1.4 802.11n MIMO configurations and terminology

A full specification of an 802.11n system includes a number of parameters. MIMO is often defined as
MxN: e.g. 2x2, 3x3. In this case M refers to the number of transmit antennas configured, and N to the

Aruba Networks 802.11n Theory and Practice 12

number of antennas at the receiver. Next the number of spatial streams must be specified. While for
most access points this will be the same as the number of antennas, many clients, particularly where
power consumption, processing or size is a concern, may have ‘asymmetric’ capabilities. 802.11n offers
MIMO specifications up to 4x4, with 4 spatial streams.

2.1.5 High Throughput PHY: 40 MHz channels

All earlier versions of 802.11 have used 20 MHz channels, defined in the 2.4 GHz and 5 GHz bands.
Some vendors have increased throughput, particularly in point-to-point bridging applications by using two
adjacent channels simultaneously, usually modulating each channel separately then combining the
streams at the far end.

802.11n specifies operation in the same 20 MHz channels used by 802.11b/g in the 2.4 GHz and 802.11a
in the 5 GHz bands, but adds a mode where a full 40-MHz wide channel can be used. As might be
expected, this offers approximately twice the throughput of a 20 MHz channel. However, while in the 5
GHz band the channels are defined as pairs of existing 20 MHz channels, they do not line up with
previously-defined 20 MHz channels in the 2.4 GHz band. This means that when a 40 MHz channel is
used in 2.4 GHz, it will interfere with all possible 802.11b/g channels: effectively, 802.11n can only be
used in 2.4 GHz in one channel, and in the absence of 802.11b/g users.

2.1.6 High Throughput PHY: Shorter guard interval

The diagram below shows how the guard interval is used in OFDM. 802.11 uses complex modulation
techniques with OFDM, where large numbers of bits of input data are coded into a single OFDM symbol at
RF.

a
2
A c
B
1
Guard Inter-symbol
interval interference
802.11n (also 802.11a/g) transmission is by
Received power level

OFDM symbols (examples N, N+1, N+2).

a N N+1 N+2 Multipath increases delay spread at the
receiver; the guard interval prevents inter-
b N N+1 N+2 symbol interference. In this example, path b
is within the guard interval while c causes
c N N+1 N+2 inter-symbol interference.

time

Previous 802.11 standards used a guard interval of 800nsec. 802.11n adds an option for 400nsec, negotiated
between receiver and transmitter, for cases where the worst-case multipath delay is low. (propagation in
free-space, delay = distance x 0.3 metres/nsec, so 400nsec is equivalent to 120metres path difference.)

13 802.11n Theory and Practice Aruba Networks

For best (least-error) decoding, the symbol must arrive at the receiver without any interference or noise.
Previous sections of this document have shown how 802.11n uses MIMO to improve reception of
multipath, but this only works symbol-by-symbol. Inter-symbol interference occurs when the delay
between different RF paths to the receiver exceeds the guard interval, causing a reflection of the previous
symbol to interfere with the strong signal from the current symbol: a form of self-interference.

The optional 400nsec short guard interval in 802.11n can be used when the path difference between the
fastest and slowest RF paths is less than that limit. The diagram includes a quick calculation for path loss,
but in reality multipath reflections can introduce RF phase changes and reach 400nsec relatively easily.
This means the shorter guard interval will be very useful in consumer settings, but cannot be relied upon
when planning for enterprise deployments.

2.1.7 High Throughput PHY: More subcarriers

Through advances in implementation, it is now possible to squeeze more OFDM subcarriers (each
subcarrier allows more data to be transmitted over the RF channel) in a 20MHz or 40 MHz channel.

26 carriers 26 carriers 28 carriers 28 carriers

-10MHz fc +10MHz -10MHz fc +10MHz

52 subcarriers (48 usable) for a 20 MHz 56 subcarriers (52 usable) for a 20

non-HT mode (legacy 802.11a/g) channel MHz HT mode (802.11n) channel

57 carriers 57 carriers

-20MHz -10MHz fc +10MHz +20MHz

114 subcarriers (108 usable) for a 40 MHz HT mode (802.11n) channel

The additional subcarriers effectively add bandwidth to the channel, allowing increased data rates for a
given modulation type (see the section below on new modulation rates). (The figures above include pilot
tones, used for dynamic calibration, hence the lower figure for ‘usable subcarriers’. The effect is not very
significant).

2.1.8 High Throughput PHY: New Modulation Rates

Radio systems have to adapt to the signal and noise characteristics of the RF path, and they accomplish
this by changing the modulation rate. For a given SNR (signal to noise ratio) the system will change the
modulation rate to provide the best compromise between data rate and error rate: at any point,
modulating for a higher data rate will increase the error rate. This is a continuous decision-making
process, the transmitter relying on feedback from the receiver about its SNR to adjust the transmit
modulation.

Aruba Networks 802.11n Theory and Practice 14

While 802.11a and g specify 7 rates (6, 12, 18, 24, 36, 48 and 54 Mbps), 802.11n provides many more:
over 300. However, the basic set is of 8 rates:

Basic rates of 802.11n (Mbps) 20 MHz channel; single stream; 800 nsec GI; equal modulation
6.5 13 19.5 26 39 52 58.5 65

This is a set of rates for one spatial stream in a 20 MHz RF channel and with an 800 nsec guard-interval
and ‘equal modulation’ on all spatial streams. This basic set of rates is comparable to the 802.11b/g rates
above: each rate is improved by about 8% (e.g. 18 to 19.5 Mbps) by using slightly wider bandwidth and
improved modulation. The 65 Mbps rate has no equivalent in 802.11a/g.

Other rates are generally derived in multiples of the basic rates above:

Rates of 802.11n (Mbps) 20 MHz channel; two streams; 800 nsec GI; equal modulation
13 26 39 52 78 104 117 130

Rates of 802.11n (Mbps) 40 MHz channel; single stream; 800 nsec GI; equal modulation
13.5 27 40.5 54 81 108 121.5 135

The 40 MHz channel allows slightly more than twice the data rate of two 20MHz channels.

Rates of 802.11n (Mbps) 20 MHz channel; single stream; 400 nsec GI; equal modulation
7.2 14.4 21.7 28.9 43.3 57.8 65 72.2

The shorter 400nsec guard interval allows slightly higher data rates than 800nsec.

Rates of 802.11n (Mbps) 20 MHz channel; two streams; 800 nsec GI; unequal modulation
39 52 65 58.5 78 97.5

For a given system, the range of choices will be smaller than the tables above or below would indicate,
because some of these factors are fixed for a given system:

• 20 MHz or 40 MHz channel. As discussed elsewhere in this paper, a 40 MHz channel will not often
be feasible for an enterprise deployment in the 2.4 GHz band. However 40 MHz channels are
likely to be widespread in the 5 GHz band.
• Spatial streams. As described above, a 3x3 system will normally support 3 spatial streams.
However, where there is insufficient RF path isolation between streams, even a 3x3 system may
not be able to support 3 diverse streams. Also, a 3x1 system only supports one spatial stream,
and many clients may initially be equipped with only one antenna.
• Guard interval. The guard interval is the time between OFDM symbols in the air. Normally it will
be 800 nsec: the option is for a 400 nsec guard interval, but as noted above, reliable detection
with this value may be challenging under normal enterprise conditions.
• Convolutional coding. When data arrives at the PHY layer for transmission, it is scrambled and
coded. This alters its spectral characteristics in order to achieve the best signal to noise ratio, and
also includes built-in error correction, known as convolutional coding. The 802.11n standard
includes BCC (block convolutional coding), as included in previous 802.11 standards, but also
adds an option for LDPC (low density parity check) coding, which can improve effective throughput
for given RF conditions.
• Modulation. All spatial streams may use the same (equal) modulation, or they may carry different
(unequal) modulation and coding. An example might be where there are three spatial streams with

15 802.11n Theory and Practice Aruba Networks

 good MIMO characteristics, but one stream has a high noise floor or a low signal level: under
these conditions the weak stream would support a lower data rate than the other streams. It is too
early to predict with confidence how often unequal modulation will be invoked.

Data rate ranges for various 802.11n system parameters

Equal Channel Guard Number of Minimum Maximum Number of
Modulation width (MHz) Interval streams PHY rate PHY rate rates
(nsec) (Mbps) (Mbps)
YES 20 800 1 6.5 65 8
YES 20 400 1 7.2 72.2 8
YES 20 800 2 13 130 8
YES 20 400 2 14.4 144.4 8
YES 20 800 3 19.5 195 8
YES 20 400 3 21.7 216.7 8
YES 20 800 4 26 260 8
YES 20 400 4 28.9 288.9 8
YES 40 800 1 13.5 135 8
YES 40 400 1 15 150 8
YES 40 800 2 27 270 8
YES 40 400 2 30 300 8
YES 40 800 3 40.5 405 8
YES 40 400 3 45 450 8
YES 40 800 4 54 540 8
YES 40 400 4 60 600 8
NO 20 800 2 39 97.5 6
NO 20 400 2 43.3 108.3 6
NO 20 800 3 52 156 14
NO 20 400 3 57.8 173.3 14
NO 20 800 4 65 214.5 24
NO 20 400 4 72.2 238.3 24
NO 40 800 2 81 202.5 6
NO 40 400 2 90 225 6
NO 40 800 3 108 324 14
NO 40 400 3 120 360 14
NO 40 800 4 135 445.5 24
NO 40 400 4 150 495 24

2.1.9 High Throughput PHY: Duplicate format

Non-HT (legacy) duplicate format defines the subcarriers used for transmitting in mixed-mode, where each
half of the 40 MHz channel is used separately: it’s the same format used by 802.11a/g devices. This
makes the 802.11n access point compatible with legacy devices.

Aruba Networks 802.11n Theory and Practice 16

 52 carriers 52 carriers

-20MHz -10MHz fc +10MHz +20MHz

52 x 2 usable subcarriers for a 40 MHz duplicate format (802.11n)

channel (applicable to greenfield and mixed-modes)

HT duplicate format is used only for the lowest data rate for a 40 MHz channel, for a single spatial stream.
It sends the same data on each 20 MHz sub-channel, providing better error performance for a given (high)
noise level.

2.2 Techniques to enhance the MAC

2.2.1 MAC layer enhancements: Frame aggregation
Since a client (or AP) must contend for the medium with every frame it wishes to transmit, and contention,
collisions on the medium and backoff delays waste time that would otherwise be used for sending useful
traffic, 802.11n incorporates mechanisms to aggregate frames at stations, and thus reduce the number of
contention events. Many tests have shown the efficacy of this effect in prior 802.11 standards. For
instance, in 802.11g, a given configuration can send 36 Mbps of data using 1500 Byte frames, but when
the frame length is reduced to 256 Bytes, throughput drops to 16 Mbps (it is important when conducting
these tests to isolate other effects such as the number of clients in the cell).

With MAC-layer aggregation, a station with a number of frames to send can opt to combine them into an
aggregate frame (MAC MPDU). The resulting frame contains less header overhead than would be the
case without aggregating, and because fewer, larger frames are sent, the contention time on the wireless
medium is reduced.

Two different mechanisms are provided for aggregation, known as Aggregated MSDU (A-MSDU) and
Aggregated-MPDU (A-MPDU). The figure below shows the general architecture:

Applications

P1 P2 P3 MSDU (MAC Service Data Unit) P1 P2 P3

P1 P2 P3

MAC processing MAC processing MAC processing

MAC MAC MAC MAC

P1 P2 P3 MPDU (MAC Protocol Data Unit) P1 P2 P3
header header header header

Aggregated MSDU format (A-MSDU) PHY layer Aggregated MPDU format (A-MPDU)

17 802.11n Theory and Practice Aruba Networks

In the A-MSDU format, multiple packets from higher layers are combined and processed by the MAC layer
as a single entity. Each original packet becomes a subframe within the aggregated MAC frame, with its
own sub-header containing source & destination addresses and length. Thus this method can be used for
packets with differing source and destination addresses, but only MSDUs of the same priority (access
class, as in 802.11e) can be aggregated.

An alternative method, A-MPDU format, allows concatenation of MPDUs into an aggregate MAC frame.
Each individual MPDU is encrypted and decrypted separately. Since MPDUs are packed together, this
method cannot use the earlier 802.11 per-MPDU acknowledgement mechanism for unicast frames. A-
MPDU must be used with the new Block Acknowledgement function of 802.11n.

In order to accommodate aggregated MAC frames, the maximum length accepted by the PHY is
increased from 4095 in previous standards to 65535.

2.2.2 MAC layer enhancements: Multiple Traffic ID Block Acknowledgement

(MTBA)
Earlier 802.11 standards demanded an ack frame for every unicast data frame transmitted. The new
block ack feature allows a single ack frame to cover a range of received frames. This is particularly useful
for streaming video and other high-speed transmissions, but if a frame has not been received, there will be
a delay before a non-acknowledge is received and re-transmission can be accomplished: this is not often
a problem with video, where re-transmission is often not feasible, given the time constraints of the media,
but may be problematic for other applications.

Block Ack covers many frames in one Ack

P4 header P3 header P2 header P1 header

header Ack P1, P2, … P4

Aggregate MPDU is a special case requiring Block Ack

P3 P2 P1 header
header Ack P1, P2, … P3

The format of the Block Ack is a bit-map to acknowledge each outstanding frame: it is based on a
mechanism originally defined in 802.11e. The bit-map identifies specific frames not received, allowing
selective retransmission of only those required.

2.2.3 MAC layer enhancements: Reduced inter-frame spacing (RIFS)

When a station (client or AP) has a number of frames to send sequentially, it must pause between frames.
However, these pauses constitute overhead for the overall network. Prior to 802.11n, the pause between
frames transmitted by the same station was set at SIFS (single inter-frame spacing). 802.11n defines a
smaller spacing, RIFS (reduced inter-frame spacing). RIFS cannot be used for the pause between frames
transmitted by different stations, and it can only be used when the station is transmitting in 802.11n HT
(greenfield) mode. It accomplishes similar goals to the MAC aggregation functions explained earlier, but

Aruba Networks 802.11n Theory and Practice 18

arguably with less implementation complexity. 802.11n defines a RIFS interval as 2 usec, whereas SIFS
is 16 usec.

2.2.4 MAC layer enhancements: Spatial multiplexing power save (SM power
save)
The basic 802.11n power save mode is based on the earlier 802.11 power save function. In this mode,
the client notifies the AP of its power-save status (intention to sleep), then shuts down, only waking for
ATIMs (Ad-hoc traffic indication maps) broadcast by the AP, while the AP buffers downlink traffic for
sleeping stations between ATIMs.

Power save in 802.11n is enhanced for MIMO operation with SM power save mode. Since MIMO requires
maintaining several receiver chains powered-up, standby power draw for MIMO devices is likely to be
considerably higher than for earlier 802.11 equipment.

A new provision in 802.11n allows a MIMO client to power-down all but one RF chain when in power save
mode. When a client is in the ‘dynamic’ SM power-save state, the AP sends a wake-up frame, usually an
RTS/CTS exchange, to give it time to activate the other antennas and RF chains. In static mode, the
client decides when to activate its full RF chains, regardless of traffic status.

2.2.5 MAC layer enhancements: Power Save Multi-poll (PSMP)

PSMP can be considered an extension of APSD: it has the same extensions, scheduled- and
unscheduled-PSMP (viz. S-APSD, U-APSD).

802.11n (HT)
AP

Incoming to AP
(wired side) Data Multicast Data

Buffers traffic for A Data Multicast Data Buffers traffic for A

AP

Sleep Trigger/Data Ack / Sleep

A

time

Unscheduled PSMP is simpler: it is very similar to U-APSD, supporting both trigger-enabled and delivery-
enabled options. Each sleep interval is considered and signaled independently, with the client determining
when to wake to receive or transmit data. In the diagram above, the ‘sleep’ frame informs the AP that the
client will stop receiving frames until further notice. When the client wishes to communicate, it sends a
regular or trigger frame to the AP, and both parties then transmit whatever data is queued. At the end of
this exchange, the client can indicate its return to sleep mode.

Scheduled PSMP is very similar to the S-APSD function introduced in 802.11e. The client requests a T-
Spec (traffic specification) from the AP, giving details of data rate, frame size, frame interval and access
class (QoS priority) of the traffic streams it wishes to send and receive. The AP, once it has admitted this

19 802.11n Theory and Practice Aruba Networks

T-Spec, defines and a polling schedule for the client. Since there may be several clients using S-PSMP,
the AP defines global PSMP SP (service period) for S-PSMP traffic, informing other stations they cannot
transmit during these intervals. Once a PSMP SP is declared, the AP first transmits data in the downlink
direction to all applicable S-PSMP clients during the DTT (downlink transmission time), then accepts traffic
from clients during the UTT (uplink transmission time).
A C
PSMP = power save
multi-poll
802.11n
802.11n
PSMP DTT = downlink
Not
AP PSMP transmission time
802.11n UTT = uplink
PSMP transmission time
B
Incoming to AP Data Data Data Data
(wired side) A B A B

Buffers traffic, no polling PSMP-DTT PSMP-UTT Buffers traffic, no polling PSMP-DTT PSMP-UTT
AP Publish Data Data Data Data
schedule A B A B

TSpec Data Data

Arequest

TSpec Data Data

B request

sends & receives frames Transmit inhibited by NAV sends & receives frames Transmit inhibited by NAV
C

time

S-PSMP is a very efficient way to transmit streaming or periodic traffic over 802.11n: there is no
contention for the medium, as everything depends on a published schedule.

2.3 Compatibility Modes and Legacy Support in 802.11n

The most vexing aspect of 802.11n is how it works in the presence of earlier 802.11 technologies.
Because it operates in the same bands as legacy 802.11, and is developed by the same standards
bodies, and because there are already more than 100 million Wi-Fi devices in use world-wide, 802.11n is
designed to support earlier forms of Wi-Fi. This includes:

• Support for legacy clients. 802.11a/b/g clients can connect to 802.11n APs. They will not be able
to use 802.11n features, and their performance will not be improved when connecting to an
802.11n AP.
• Awareness of neighbouring or overlapping 802.11a/b/g networks. This is particularly important
when using the new 40 MHz channel capability, which would impair the performance of such
networks.

Unfortunately, as explained elsewhere in this note, working with legacy 802.11 clients and networks
degrades the performance of 802.11n considerably, to the point where 802.11a/b/g clients will see very
comparable performance whether they are using an 802.11a/b/g or 802.11n access point. In addition,
working with legacy clients ‘poisons’ the 802.11n cell: its capacity will be severely degraded as soon as
even one legacy client is present. This does not negate the need for legacy operation, but it does
increase the urgency of upgrading the client population to 802.11n. The diagram below shows how

Aruba Networks 802.11n Theory and Practice 20

introducing an 802.11a client into an 802.11n cell reduces the throughput (based on data rates alone: co-
existence mechanisms will further reduce throughput).

A B
2SS SDM
802.11n 2SS SDM
802.11n
802.11n
52 Mbps
AP 104 Mbps

Data on AP to B 1KB B to AP 1KB B to AP 1KB

A to AP 1KB @ 52 Mbps AP to A 1KB @ 52 Mbps AP to A 1KB @ 52 Mbps
the air @ 104 Mbps @ 104 Mbps @ 104 Mbps
2N N N 2N N 2N time

Data transferred = 6KB, 3KB to/from A and 3KB to/from B ***(for a time interval of 9N*).

A B
2SS SDM
802.11n 2SS SDM

802.11n
802.11n 1SS 117 Mbps
52 Mbps AP
802.11a
24 Mbps C
Data on AP to B 1KB
the air A to AP 1KB @ 52 Mbps AP to C 1KB @ 24 Mbps AP to A 1KB @ 52 Mbps
@ 104 Mbps
2N N 4N 2N time

Data transferred = 4KB, 2 KB to/from A, 1KB to/from B, 1KB to/from C ***(for a time interval of 9N) .

*** In a time interval of 9N, ignoring contention time

Depending on the requirements to support legacy 802.11a/b/g clients, 802.11n defines three modes of
client compatibility: High Throughput (Greenfield), High Throughput Mixed-Format and Non-HT (legacy)
mode.

HT (High Throughput) Greenfield format

HT-GF-STF HT-LTF1 HT-SIG HT-LTF … HT-LTF HT-LTF … HT-LTF Data

Key
STF Short Training Field
Non-HT format LTF Long Training Field
SIG Signal
GF Greenfield
L-STF L-LTF L-SIG Data
L Legacy (e.g. pre-802.11n)
HT High Throughput (e.g. 802.11n)

Note: In 802.11n terminology, L = Non-HT

HT mixed format

L-STF L-LTF L-SIG HT-SIG HT-STF HT-LTF … HT-LTF HT-LTF … HT-LTF Data

High Throughput (HT). In HT or Greenfield mode, the AP does not expect to connect to any legacy 802.11
clients, and indeed, assumes that there are none operating in the area. Even so, the first part of the
preamble is a legacy short training sequence, enabling other devices and APs to sense that there is
802.11 equipment in the area. However, after that no indication is available that will allow older devices to
understand the remaining part of the transmission: it is all in HT-format. HT-mode is the only one of the
three that is not mandatory in the 802.11n standard.

21 802.11n Theory and Practice Aruba Networks

Non-HT format. This is essentially legacy mode. The frames are all in 802.11a/g format (PHY and MAC),
so they can be understood and decoded by 802.11a/b/g clients. This mode gives essentially no
performance advantage over legacy networks, but offers full compatibility. Non-HT mode cannot be used
with 40 MHz channels.

HT Mixed Format. As might be expected, this allows operation of 802.11n clients in HT mode, while
legacy clients are fully-supported. There is a full legacy preamble, then the option of using HT or legacy
format afterwards. The preamble allows legacy clients to detect the transmission, acquire the carrier
frequency and timing synchronization, and the L-SIG field allows them to estimate the length of the
transmission.

This (mixed) mode can be used in a 40 MHz channel, but to make it compatible with legacy clients, all
broadcast and non-aggregated control frames are sent on a legacy 20 MHz channel as defined in
802.11a/b/g, so as to be interoperable with those clients. And of course all transmissions to and from
legacy clients must be within a legacy 20 MHz channel.

2.3.1 Phased coexistence Operation (PCO)

Phased coexistence operation is designed to allow an 802.11n AP to support both 802.11n and 802.11a/g
clients while being a good neighbour to other legacy 802.11 APs operating in the area. This differs from
continuous ‘mixed mode’ operation because it allows 40 MHz operation for part of the time, whereas a
mixed mode cell must use only one 20 MHz channel.

As the diagram shows, the AP time-slices its cell, switching between 20 MHz, 802.11a/b/g compatible
operation in each of the constituent 20 MHz channels and full 40 MHz operation for 802.11n clients. To
maintain order and provide the best throughput, two mechanisms are used.

A C

802.11n
802.11a In 20
20 MHz AP MHz or
40 MHz
802.11a
20 MHz
B

Sends self- Sends self- Sends self- Sends self- Sends self-
addressed CTS addressed CTS addressed CTS addressed CTS addressed CTS
AP setting NAV to N setting NAV to N setting NAV to N setting NAV to N setting NAV to N
in upper 20 MHz in both 20 MHz in lower 20 MHz in both 20 MHz in upper 20 MHz
channel channels channel channels channel

AP and A and C
C C C C
channel

exchange traffic
20 MHz
Upper

T T AP and C in 802.11a 20 MHz T AP and C T

No transmission No transmission
S S exchange traffic channel (mixed S exchange traffic S
in 802.11n 40 mode) in 802.11n 40
MHz channel (HT MHz channel(HT
AP and B and C AP and B and C
C (greenfield) mode C C greenfield) mode
channel
20 MHz
Lower

exchange traffic exchange traffic

in 802.11a 20 MHz T T No transmission T in 802.11a 20 MHz
channel (mixed S S S channel (mixed
mode) mode)
N N N N N

time

Firstly, for 802.11n clients, the AP advertises a forthcoming switch of operation, allowing these clients to
continue communicating in all time slices, whether 20 MHz or 40 MHz. Clearly, throughput is lower during
20 MHz time slices, but nevertheless, two-way transmission between the client and the AP can continue

Aruba Networks 802.11n Theory and Practice 22

uninterrupted. While the diagram shows all time slices of equal length for simplicity, the AP can choose
between a number of defined time intervals for each time slice.

For legacy 802.11 clients, only one of the three modes of operation (40 MHz, upper 20 MHz, lower 20
MHz channels) will be possible at any time: these clients will only operate in one of the 20 MHz channels.
During time slices when the AP is in one of the other two modes, these clients must be informed that they
cannot transmit: the AP will not be able to receive their frames. This is achieved by the AP transmitting a
self-addressed CTS (clear to send, see below) frame with a ‘duration’ value equal to the next time-slice
duration. When clients hear this frame, they set their NAV (network allocation vector) to this value: under
the rules of all 802.11 standards, they are not allowed to attempt transmission until this timer has expired.

Phased coexistence operation is also a good-neighbour policy because APs and clients in range of the AP
will be able to hear the self-addressed CTS messages and set their NAV timers in similar fashion, avoiding
one form of co-channel interference. However, 802.11a/b/g cells operating in range of a PCO cell will
experience much-reduced capacity, as APs and clients will be inhibited from transmitting for a significant
percentage of the time previously available.

2.3.2 Other mechanisms for coexistence: Dual-CTS protection (CTS-to-self)

802.11n (HT)
20 MHz AP
802.11a (L)
20 MHz
B
L = Legacy (802.11a/b/g mode)
HT = High Throughput (802.11n mode)

CTS (HT) CTS (L) CTS (HT) Data frame (L) CTS (L) Data frame (HT)
AP
CTS to self CTS to self

RTS (HT) Sends data (HT) Sets NAV Receives data (HT)
A

Sets NAV Receives data (L) Sets NAV

B

time

In this mode, the AP transmits extra CTS (clear to send) frames, so every data frame, whether from a
client or the AP, is protected by a legacy and an HT CTS. When the traffic is generated by a station, it first
sends an RTS (request to send) to the AP. The AP responds with two CTS frames, one in HT and the
other in legacy format. The client is then free to transmit the data frame, while other clients in the same
and neighbouring cells set their NAV correctly so they do not transmit over the authorized frame,
interfering with it. When the AP has traffic to send, it uses a self-addressed CTS frame to perform the
same function.

Dual-CTS makes the network a good neighbour to overlapping or adjacent legacy 802.11 networks. It
also solves the ‘hidden node’ problem where different clients in a cell may not be able to hear each others’

23 802.11n Theory and Practice Aruba Networks

transmissions, although, by definition they can all hear the AP and its CTS frames. However, the use of
RTS/CTS further reduces the data throughput of the cell.

2.3.3 Other mechanisms for coexistence: 40 MHz-intolerant indication

A mechanism is defined in 802.11n that allows an AP to advertise in its beacon that it prohibits 40 MHz
operation in adjacent or overlapping cells. This indication can also be transmitted by a client, informing
the AP that it is required to prohibit 40 MHz operation by all other members of its cell.

2.3.4 Using 802.11n in the 2.4 GHz band

The 2.4 GHz band presents special challenges for 802.11n. While in the 5GHz band, usable channels
have been defined by regulators to be 20 MHz apart, the spacing in 2.4 GHz is 5 MHz. Combined with the
limited overall spectrum available in this band, this has led to customary usage where channels 1, 6 and
11 are used in a 3-channel plan (for the US: other regulators allow use to channel 13). This presents a
difficulty when defining a 40 MHz channel in 2.4 GHz, because it is not possible to build it on two adjacent
20 MHz channels, so the phased coexistence operation feature will not be useful when operating as part
of an existing network, and other coexistence methods will not work well when the 40 MHz channel and
overlapping networks have 5, 10 or 15 MHz offsets, even though there is spectrum in the band for one 40
MHz and one 20 MHz band. This channel overlap is illustrated below:

-19MHz +19MHz -9MHz +9MHz

0dBr 0dBr 0dBr 0dBr
-21MHz +21MHz -11MHz +11MHz
-20dBr -20dBr -20dBr -20dBr
-30MHz +30MHz -20MHz +20MHz
-28dBr -28dBr -28dBr -28dBr
-40MHz +40MHz -30MHz +30MHz
-40dBr -40dBr -40dBr -40dBr
-40MHz -30MHz -20MHz -10MHz fc +10MHz +20MHz +30MHz +40MHz -30MHz -20MHz -10MHz fc +10MHz +20MHz +30MHz

Spectral mask for 40 MHz channel Spectral mask for 20 MHz channel

3 4 5 6 7 8 9

1 6 11

Channel 1 2 3 4 5 6 7 8 9 10 11 12 13
Centre Frequency 2412 2417 2422 2427 2432 2437 2442 2447 2452 2457 2462 2467 2472

Channels defined for 2.4 GHz band, showing common 20 MHz channel plan and 40 MHz options

This means that in practice, it is unlikely that 40 MHz channels will be used in the 2.4 GHz band. But this
does not mean that 802.11n should not be used: there will be performance improvements even when a
20 MHz channel is used, although the presence of legacy clients will reduce the realized benefits.

This also illustrates one of the drivers for using 802.11n in handheld clients such as mobile Wi-Fi phones.
Design of these devices will offer challenges in terms of siting diverse antennas and power consumption –
already a concern – but allowing the device to work in greenfield 802.11n mode will avoid the need for the

Aruba Networks 802.11n Theory and Practice 24

cell to fall into a PCO or other coexistence mode, and greatly increase overall cell performance, even if a
single antenna with a single transmit/receive chain is used.

25 802.11n Theory and Practice Aruba Networks

3 Product development and capabilities
This section deals with three issues: IEEE 802.11n stipulations for mandatory and optional features; Wi-Fi
Alliance tests for certification; and chip-designers implementations. There is a fourth level, those features
the equipment vendors such as Aruba will choose to implement, but historically, most features provided
by the chip vendors have been supported – and of course, those that are not supported in silicon cannot
be implemented in equipment.

3.1 Standards and Certification framework

In the table below, the first two columns are correct as of April 2007, while the last column is based on
discussions with chip vendors, and the opinions and experience of Aruba’s engineering team.

Feature Mandatory / Optional in WFA Certification Likely to be implemented (initial, ‘draft-

(M = Mandatory, O = Optional) n’) by chip vendors and product
vendors?
APs Client: Client: Client:
PC HandHeld Consumer
Baseline Electronics
High Throughput
PHY
40 MHz channel in O O O O Supported
5 GHz
20 MHz channel in M M O M Supported
5 GHz
40 MHz channel in Supported (may be disallowed by Wi-Fi
2.4 GHz Alliance)
20 MHz channel in M M O O Supported
2.4 GHz
Concurrent O O Supported (this is for a dual-radio solution:
operation in 2.4 not splitting SS’s over two bands).
and 5 GHz bands
SDM (Spatial 2x2 / 2x2 / 4x4 1x1 / 2x2 2x2 / 3x3, Initial chipsets support only 2 spatial
Division 4x4 (equal 4x4 (equal streams, (despite public marketing positions
Multiplexing): mod at modulation & statements). We expect a maximum of 2
number of spatial client) only) or 3 antennas on commercial products for
streams supported enterprise, perhaps at the end of 2007.
STBC (Space-time M – Tx O We expect only symmetrical coding options
block coding): the 2x1 to be implemented initially. Explicit
modulation options feedback frames may not work across dis-
available similar vendors.
Spatial expansion O We don’t expect to see this capability
of legacy frames initially.
Reverse direction O We don’t expect to see this, except maybe if
same vendor’s chip sets at both ends, AP &
client.
TxBF (Transmit Not yet properly defined (target Q1 2008).
Beamforming) Not likely to be supported before 2008.
Greenfield Mode O O Optional but critical for improved
performance! All products will support this.
L-SIG legacy M Supported.
protection (mixed
mode)
High Throughput O O Supported.
duplicate mode

Aruba Networks 802.11n Theory and Practice 26

Short guard interval O O Supported.
MRQ/MFB. Fast M Supported.
Link adaptation
MAC Layer
Enhancements
A-MPDU, A-MSDU O Supported (although there are some
questions about a security vulnerability of A-
MSDU)
PSMP. Power Save M Supported.
Multipoll
MTBA Block Ack M Supported.
Quality of Service
WMM (including M M M M Will support WMM in 802.11n mode.
WMM Power Save)
Security
WPA2 M M M M Will certainly support WPA2 in all 802.11
modes.

3.2 Expected product introduction timelines

This section gives estimates of the type of 802.11n products likely to be offered to the enterprise market,
and when they might be introduced. The information is speculative – the passage of time will surely show
that much of it is wrong – but it represents the author’s best guess as of May 2007. The major players
determining this timeline are the silicon vendors and client (PC) designers – and of course enterprise
customers and their buying decisions. This section ignores the consumer market, which is likely to
develop much more quickly and may in the end be a significant influence on the enterprise market,
especially by affecting silicon prices and overall user acceptance of the technology.

We suggest that while ‘pre-n’ products present too much technology risk to use for more than isolated
tests in enterprise networks, ‘draft-n’ products certified by the Wi-Fi Alliance will be suitable for pilot
deployments, and to alleviate any local congestion points before full upgrade plans can proceed, either
with ‘draft-n’ products (after experience, the standards process and certification tests demonstrate reduced
risk) or with fully 802.11n-certified products.

Time interval Product Cert Notes

2Q-3Q 2007 AP draft-n APs will support two spatial streams. Many APs will be ‘dual-radio-
set’, capable of either dual-band 802.11n operation or with one
radio-set for 802.11n and the other for 802.11a/g.
PC draft-n As above. The technology for aftermarket PC NIC cards is quite
similar to that for an AP. It will take longer for the technology to be
integrated on motherboards at manufacture, but this will offer a
superior solution, as proper antenna spacing and placement is
easier when embedded in the PC case.
4Q 2007 – AP draft-n APs will support 3 spatial streams.
1Q 2008
PC draft-n New PCs will ship with 802.11n embedded as standard, as vendors
ramp production volume up to reduce costs and accelerate market
adoption. Unlikely to move to 3 spatial streams on the client
because the increased cost, complexity, heat generated and PC
board-space is unlikely to justify.
2Q-3Q 2009 AP, PC n Full 802.11n certification should be available in this timeframe.
Capabilities as above. APs will possibly extend to 4 antennas, but

27 802.11n Theory and Practice Aruba Networks

 unlikely 4 spatial streams, except for special circumstances like
point-point bridging applications. APs may become capable of
beamforming.
Phones n Wi-Fi phones have only just incorporated 802.11g, and very few yet
support 5 GHz operation.
2010 Handheld n Bar code readers, location tags & other specialized Wi-Fi clients will
devices take years to incorporate 802.11n because while 802.11a/g
technology is cheaper, more compact & lower-power (and networks
are backwards-compatible), designers will not be able to justify.

4 Migration strategies

4.1 Different paths to enterprise-wide 802.11n

There are essentially three strategies to get an Enterprise WLAN to 802.11n: greenfield, overlay and
interspersed. The four questions the network designer must answer are where to place the new 802.11n
APs, which channels to use, whether to remove some older 802.11a/b/g APs, and how to manage the
client population.

As indicated earlier, a uniform 802.11n network will allow greater range and hence wider AP spacing than
an 802.11a/b/g network. RF engineers insist that every site has unique conditions and usually avoid
giving any figures for ‘typical’ deployments, but a rule of thumb in ‘carpeted’ enterprise buildings is that
with design figures of 9-12 Mbps minimum rate and 150% cell overlap, and a ‘typical’ office layout and
user population, 802.11a/g APs can be spaced at about 15-21 metres (50–70 feet). With 802.11n APs
and 802.11a/b/g clients (remembering they must support at least two spatial streams), we would expect
the same guidelines to hold.

However, with 802.11n APs and 802.11n clients with at least two spatial streams (or antenna diversity
processing at both ends – an option that is unlikely to be available till 2008), Aruba estimates that AP
spacing can be increased to perhaps 20-25 metres (65-85 feet). In a new ‘greenfield’ installation, this
offers a considerable savings of AP hardware, cabling & installation, and even in edge switch ports, closet
space, power and cooling provisions.

As noted throughout this paper, 802.11n offers an opportunity to increase usage of the 5 GHz band where
channels are plentiful, interference is uncommon and 40 GHz channel usage is feasible. However, while
PC clients will be readily available with 802.11n and draft-n capability, specialized clients such as Wi-Fi
phones, location tags or handheld bar-code readers will be confined to 2.4 GHz for 12-24 months or more,
with few exceptions. For this reason, and because of the considerable installed base of 802.11b/g client,
we suggest that any enterprise network should provide coverage in the 2.4 GHz band. Since WLAN
vendors will be marketing access points supporting two radios: either with both capable of 802.11n, or
one 802.11n and the other 802.11a/b/g, this will be quite feasible, but remember that if legacy clients are
to be fully and reliably supported, AP spacing cannot be increased.

In general, we recommend that when 802.11n access points are used, they should support two radio-sets,
one operating with a 40 GHz channel at 5 GHz and the other with a 20 MHz channel at 2.4 GHz. This will
provide maximum performance and flexibility for a small installed price premium over an
802.11n+802.11a/b/g or a single 802.11n access point.

Even though phones, location tags and other handheld clients will be slow to adopt 802.11n, it is important
that they eventually achieve this goal. Even if they only support one antenna, one RF chain and a single

Aruba Networks 802.11n Theory and Practice 28

spatial stream, the upgrade will prevent access points from dropping into legacy-compatible modes and
hence will remove a serious inhibitor of performance as measured by cell capacity.

It is likely that an early driver for introducing 802.11n access points in an existing network will be
defensive: to provide rogue AP detection that extends to the APs on the shelves of consumer electronics
shops. If previous upgrades of Wi-Fi are an indication, the consumer market will lead the enterprise
market for several quarters, and the only way to detect and classify an 802.11n rogue will be to use an
802.11n access point. Most WLAN vendors are expected to continue their architecture where a single
access point can dynamically switch between coverage and monitoring mode, and in a Wi-Fi service
delivery network we expect the usual ratio of 15% to 25% extra access points in order to ensure full
monitoring even under heavy load, as well as dynamic network healing in the event of AP failure. If new
APs are used for the sole purpose of monitoring, they would normally be installed at wider spacing than
indicated above. This may change for 802.11n: since some of the range-enhancing techniques are not
symmetrical and depend on the client implementing and enabling particular features. We will be able to
give firmer guidelines after more testing with early 802.11n products.

4.1.1 Greenfield
This is not really a migration: it is building a network with new APs and new clients where there was no
WLAN before. This offers an opportunity to start with all-n coverage, and to ensure that clients are n-
capable, or can be rapidly phased-out in favour of n-capable substitutes. The foremost concern with this
strategy is that all clients must be 802.11n-capable: older clients will have shorter range, similar to
802.11a/g distances, and will experience very low connectivity rates and coverage holes, at the same time
reducing the potential performance of 802.11n clients in the cells where they are connected.

802.11n Access

802.11n coverage

As seen from the approximate figures above, it may be possible to reduce the number of APs required to
cover an area by a factor of 30% or more, when comparing 802.11n with older 802.11 technologies. This
may mean that the capacity of a cell becomes more important than the connection speed of any data
client, but this is easily incorporated into the planning stage. The savings indicated above may make it
worthwhile to upgrade clients to 802.11n in order to take advantage of the wider AP spacing this enables.

Other aspects of a greenfield deployment include rogue detection and intrusion prevention. These
functions will work successfully for all types of 802.11a/b/g/n devices.

29 802.11n Theory and Practice Aruba Networks

A further consideration will be whether to cover one or both of the 2.4 GHz and 5 GHz bands. We believe
all 802.11n clients are likely to be 5 GHz-capable, so for these clients, coverage of 5 GHz only will be
sufficient. However, the likelihood of requirements, albeit infrequent, for 2.4 GHz-only client support
means that we consider it prudent for most enterprise deployments of this type to include dual-radio
access points providing coverage of the 2.4 GHz band, even if this is legacy coverage and it becomes
intermittent at the edge of the cell (due to 802.11n-driven access point spacing).

4.1.2 Summary of ‘Greenfield’ recommendations:

• Confirm that 802.11n-capable clients are available for all current and foreseeable enterprise
requirements.
• RF planning can proceed with 802.11n range assumptions, allowing fewer APs. Use increased AP
spacing only if all clients are capable of at least 2 spatial streams. Restrict existing APs to a subset
of the available 5 GHz channels, and use the remainder for 802.11n.
• Consider power to the APs, edge switch ports, etc.
• Place APs to avoid line-of-sight for best MIMO performance.
• Disable co-existence mechanisms such as PCO that support mixed-mode operation.
• Enable good-neighbour mechanisms such as dual-CTS if there are other 802.11a APs in the same
building, on the same channel.
• Ignore ‘40 MHz-intolerant’ requests.
• Use dual-radio APs where the primary radio(s) is 802.11n at 5 GHz, while the second radio (can be
802.11a/b/g) runs in the 2.4 GHz band.
• Remember that this (scheme above) does not guarantee complete 2.4 GHz coverage for non-
802.11n clients: there may be areas at the edge of cells where signal strength will not support an
802.11a/b/g signal.

Note that for a lightly-loaded greenfield network, using 802.11n access points with 802.11a spacing,
and enabling PCO for a mixed client base presents a perfectly acceptable solution: throughput will
initially be low, probably lower than an 802.11a network until the number of legacy clients is reduced,
but the network can be installed once and grow with the client base, without any required
reconfiguration.

4.1.3 AP-overlay
Here, the 802.11n network would be planned as if there were no existing network, for optimum placement
of the new APs. The old and new network will be able to operate in parallel if care is taken with RF
channel allocation, and when the new one is complete and all clients are 802.11n-capable, the old APs
can be de-installed or abandoned.

Aruba Networks 802.11n Theory and Practice 30

 802.11n Access

802.11a/g Access

802.11n coverage

802.11a/g coverage

Since 802.11n makes best use of the 5GHz band (with 40MHz channels), it makes an attractive overlay
on an 802.11b/g network operating at 2.4GHz: the original network can operate in the 2.4 GHz band (or
even 2.4 GHz and several channels of the 5 GHz band) while the 802.11n network can use the remainder
of the 5 GHz band. In this scenario, there will be no need for legacy clients to connect to the 802.11n
network, so it can operate in HT mode. (It may be useful in the early days of such a network to use it in
PCO mode if there are very few 802.11n-capable clients, but this will negate much of the throughput
advantages of 802.11n.) Against this, the 802.11n network is additive, so additional cabling, power and
edge switch ports will probably be required: it is likely these upgrades would eventually be needed,
anyway.

4.1.4 Summary of ‘Overlay’ recommendations:

• Decide on the client migration strategy: it may be better to wait until 802.11n (or ‘draft-n’) clients
are available before investing in 802.11n-capable infrastructure.
• RF planning can proceed with 802.11n range assumptions, allowing fewer APs. 802.11n clients
supporting only one spatial stream may need to switch to 802.11a mode if they are at the edge of
802.11n coverage.
• Switch PCO off, use dual-CTS and disable ’40 MHz intolerant’.
• Consider power to the APs, edge switch ports, etc.
• Place APs to avoid line-of-sight for best MIMO performance.
• Use single-radio APs if there is no future requirement for 2.4 GHz 802.11n coverage. If there is an
ambition to remove the older infrastructure, it may be better to use dual-radio access points for the
overlay so the second radio can be switched on later (becoming similar to the greenfield model
over time, as the original infrastructure disappears).

31 802.11n Theory and Practice Aruba Networks

4.1.5 AP substitution
In this model, existing APs are swapped one-for-one with new 802.11n APs. If every AP is replaced with
802.11n, the resulting network will have plenty of capacity – we estimate perhaps a three-fold
improvement, depending on the mix of 802.11a/g and 802.11n clients – but will have more APs that would
be strictly necessary. In most networks it will be possible to abandon many AP locations – either running
an RF planning tool, or using RF self-calibration tools on a network mid-way through migration will give an
indication of whether this is feasible.

802.11n Access
802.11a/g Access

802.11n
802.11a/g

A network of mixed 802.11n and 802.11a/g or even 802.11b APs can operate indefinitely: this is a
possible solution where 802.11n is important for identifying rogue APs but not for capacity or range
considerations. Note that even though APs can be swapped at a location, the new 802.11n AP may
require different power and Ethernet connections.

4.1.6 Summary of ‘AP substitution’ recommendations:

• Decide on the client migration strategy: it may be better to wait until 802.11n (or ‘draft-n’) clients
are available before investing in 802.11n-capable infrastructure.
• Run an RF planning tool to determine which APs should be upgraded to 802.11n.
• Enable PCO, dual-CTS and ’40 MHz-intolerant’.
• Consider power to the APs, edge switch ports, etc.
• Place APs to avoid line-of-sight for best MIMO performance.
• Use single-radio or dual-radio APs, depending on the existing infrastructure and plans for future
migration.

4.2 Other considerations when planning an upgrade

Points to bear in mind when planning an upgrade strategy:

• 802.11n APs will require Gigabit Ethernet connections. This may in turn require upgrading cables
to category 5e or category 6 cable (older category 5 cable is not rated for Gigabit Ethernet).
• Also, GE ports must be available on edge switches in wiring closets.
• Since 802.3af power-over-Ethernet is not sufficient for the new 802.11n APs (assuming at least 2
antennas and RF chains at 5 GHz, an AP will draw at least 15W), 802.3at must be supported by
edge switches or mid-span POE equipment, or local power bricks must be used.

Aruba Networks 802.11n Theory and Practice 32

 • AP mounting and positioning guidelines may differ. With earlier 802.11 APs, it was important to
provide as clear a line of sight as possible between the AP and the client, so hiding APs behind
metal pipes or air conditioning ducts was not recommended. With MIMO, such practices may
become optimum for 802.11n.
• Opportunities to mount APs ‘in the workspace’ - on walls, partitions and on furniture, but not in the
ceiling plenum space – increase with 802.11n, as the non-line-of-sight characteristics are
favourable. This may mitigate some of the installation issues mentioned above: if GE drops
already exist in some offices, 802.11n APs can be located to take advantage of these, and local
power for the AP via a wall brick may be a reasonable option.

5 Conclusion
This note has two goals: to explain the technology and features of the 802.11n standard, and to examine
its implications in terms of likely product development cycles, and for the design of enterprise Wi-Fi
networks.

It is clear that 802.11n represents a significant leap forward in technology and performance for enterprise
networks. Uniform ‘greenfield’ 802.11n networks will be able to offer higher capacity and longer range
than current WLANs, and there are potential savings in terms of fewer access points to cover a given
area.

However, this note has also identified several issues that will delay the adoption of 802.11n in enterprise
networks. These include the (relatively) slow progress of standards, infrastructure requirements such as
LAN edge switch ports, cabling and power, and the installed base of 802.11a/b/g clients and WLANs that
must be considered in any migration strategy. Taken together, these are likely to slow the 802.11n
adoption wave, so a number of years will pass before the full benefits are felt.

In the pages above, we have disclosed the facts of 802.11n, along with our best estimates of future
activity. There is a case in enterprise networks for tuning and extending existing 802.11a/g networks in
the short-term, while testing and running pilot trials on draft-n and eventually full 802.11n-compliant
products. When and how to migrate to 802.11n will be an important decision: the information here is
intended to assist the enterprise network manager in formulating an optimum upgrade strategy.

Despite our best efforts, developments will inevitably prove us wrong in some of the predictions above:
our advice to the reader is to check often with the technical press and vendors for current information.

802.11n offers very real and exciting benefits: it will eventually change the way we build and operate
enterprise wireless LANs.

33 802.11n Theory and Practice Aruba Networks

6 Appendix
6.1 Note on expected ‘real-world’ cell capacity with 802.11n
Whereas an 802.11a/g access point can support a top data rate of 54Mbps, 802.11n can stretch to
600Mbps. This is important, as a Wi-Fi cell (the area covered by a single AP) shares all bandwidth: it’s
very similar to the original wired Ethernet, where all stations shared a single segment of cable, and had to
co-ordinate transmissions to avoid collisions. Three effects limit the effective capacity of an 802.11 cell
(examples are taken from 802.11g, but should be equivalent form 802.11n):

• Actual data rates. While specifications quote maximum data rates, these are only achieved under
the best radio conditions. The distance from AP to client, RF obstructions such as walls, furniture,
people, and interfering RF transmissions all limit the achievable data rate. Thus, a client 20 metres
from the AP in an office environment might only support a data rate of 12Mbps rather than the
advertised peak of 54Mbps. If all clients connect at this rate, the raw capacity of the cell becomes
12Mbps.
• Contention loss. Since the wireless medium is shared, stations wishing to send data must contend
for temporary control of the medium. This contention takes time that is then not available for data
traffic, lowering the capacity of the cell. Contention depends on the number of clients and the
length of frames sent: the more clients and the shorter the frame, the less the effective capacity of
the cell. For instance, while an 802.11g may achieve throughput of 36Mbps with 1500Byte frames,
capacity drops to around 16Mbps with 256Byte frames. The MAC aggregation function in 802.11n
should reduce losses due to contention, but it will not be effective for all types of traffic.
• Legacy support. All 802.11 systems are designed to support older 802.11 clients. Thus an
802.11g AP will support 802.11b clients. However, there is a cost associated with this. When
even a single legacy client joins a cell, all other clients and the AP must indicate traffic is present,
using data rates the older client can understand. For 802.11g/b compatibility, this means using
RTS/CTS at slower rates, considerably increasing overhead and decreasing cell capacity. And of
course, when the legacy station transmits or receives, it does so at lower data rates, reducing the
effective capacity of the cell.
• The 400 nsec guard interval is not a realistic option in an enterprise network, where there will be
many multipath reflections, some with long delay spreads.
• Client design for 802.11n is challenging, as it is difficult on many devices to find the space to mount
extra antennas, the extra RF chains and processing require more board area and power, and of
course 802.11n silicon will command a price premium for several years over 802.11a/g silicon. Not
least, NIC cards and clients such as Wi-Fi phones are complex to design: until recently, Wi-Fi
phones were limited to 802.11b.

All these effects serve to reduce the effective capacity of a cell. Thus, while 802.11n advertises rates to
600Mbps, the expected capacity of an 802.11n cell is between 100 and 200Mbs, and it could certainly be
less if clients connect over long distances, transmit short frames, or there are legacy 802.11a/b/g clients
present. However, this is still an increase of 5x over 802.11a/g technology.

Here are our estimates of ‘reasonable’ expectations for data rates in an enterprise 802.11n deployment
(this assumes ‘greenfield’ deployment with no legacy 802.11a/b/g clients or APs). Interpret this table as a
comparison of the achievable data rate of 802.11n with that of 802.11a or 802.11g at the same distance
from the access point. Alternatively, it can be read as the capacity of a cell (with all-802.11n clients) when
compared with a cell of the same radius.

Spatial Streams Channel width Guard Interval Range of PHY Performance

(MHz) (nsec) rates (MHz) relative to 802.11a/g
1 (2x1, 3x1) 20 800 7.2 – 72.2 1x – 1.5x

Aruba Networks 802.11n Theory and Practice 34

1 (2x1, 3x1) 40 800 13.5 – 135 1.5x – 2x
2 (2x2) 40 800 27 - 270 2x – 4x
3 (3x3) 40 800 40.5 – 405 4x – 6x

In addition to the PHY effects above, the many MAC enhancements in 802.11n will increase the
throughput, and hence the capacity of a cell. Analysis of these effects is challenging, as they are
extremely dependent on the data patterns. Some of the enhancements are aimed specifically at
streaming media such as video, but there has been speculation that they may actually reduce the
throughput of other types of traffic such as voice or file transfers.

6.2 Note on expected ‘real-world’ data rates with 802.11n

This section is not based on thorough testing. It is not based on rigorous simulation. When papers giving
figures through either of these methods become available, they will be both more accurate and more
useful than the section below. However, a since such information has not yet been published in digestible
form, this analysis, which is based on many simplifications, but which gives results Aruba judges to be
reasonable, may serve as a guide to the performance expected in practical 802.11n networks.

Methodology
This explanation is not necessary to an understanding of the results below, but is included so the
scientifically-inclined can understand the degree of simplification involved in deriving the results.

In deriving the expected data rates and how they vary with distance from the access point, the following
steps are employed:
• Calculate free space loss with distance.
• Apply a model (Rappaport et al) for typical in-building propagation loss.
• Calculate the signal to noise ratio (SNR) by distance from the access point.
• Apply some random statistics to the above to simulate a real-world network.
• Calculate the Shannon bandwidth of the channel, based on the SNR.
• Round to the nearest (lower) data rate available.

This necessarily includes many approximations. In addition to the methodology above, the following
assumptions were used:
• Propagation loss at 5.5GHz
• In-building propagation loss index of 2.7
• Assumed transmit power to the antenna of 100mW, and combined antenna gains of 6dBi
• 20MHz, 40MHz channel bandwidths for calculating thermal noise and Shannon bandwidth
• Variation of +/-6dB in signal strength due to noise and fading effects. This is applied as a random
factor uniformly distributed between +/-6dB of the mean calculated by the in-building propagation
formula.
• No allowance is made for beamforming, as this will not be incorporated in shipping silicon until at
least 2008.
• Only up to 3 spatial streams are analysed, rather than the maximum of 4 in the standard: early
products are expected to extend only 3 antennas.

Free Space propagation loss, SNR and channel capacity

The general equation for calculating the loss in power (P) at a distance (d), where P is measured in dB, is
given by:

P = 20log10(d) + 20log10(f) + A

35 802.11n Theory and Practice Aruba Networks

Where f is the frequency (for our purposes, 5.5 GHz as the centre of the 5GHz band), and A is a constant
where we sweep a lot of base assumptions. The calculation used in the spreadsheet is:

P = 20log10(d) + 47.26

This is adjusted to give a figure for power at the receive antenna connector based a power level to the
transmit connector of 100mW and combined antenna gains of 6dBi.

In-building propagation loss

Many academics, notably Theodore Rappaport, have researched how radio waves propagate in-building.
It is not too harsh to say that the conclusion is that it is nearly impossible to predict performance in any
particular case, but much work has been done on characterizing general performance when averaged
over many buildings, clients or measurements. Of the many different models proposed, we use one of the
simplest:

P =γ x10log10(d) + 47.26

Where γ is a parameter showing increased loss over distance compared with free space.

SNR and channel capacity

From an equation defining how receive power varies with distance, a few operations derive a usable data
rate.

First, define a noise floor so a signal to noise ratio (SNR) can be calculated. ‘Thermal’ noise depends on
Boltzmann’s constant (kB) and absolute temperature (T), and it calculated by:
B

P = kBT(Δf) which simplifies to PdBm = -174 + 10log(Δf)

Where P is power in dBm, and Δf is the channel bandwidth in Hz. In our case, the channel is either 20 or
40MHz. Thermal noise in a 20MHz channel is -101dBm; and in a 40MHZ channel, -98dBm.

Channel capacity is calculated from SNR by Shannon’s formula:

C = Blog2(1 + SNR)

Where C is the capacity in bits/sec, SNR the signal to noise ratio expressed as a ratio, not in dB, and B is
a constant depending on the channel bandwidth. Since 802.11n uses different coding schemes to deal
with differing levels of SNR, this equation is taken as a general guide but is adjusted to give results more
closely modelling 802.11n.

Aruba Networks 802.11n Theory and Practice 36

 Free space and in-building propagation loss model

80
Maximum data rate (Mbps)

70
Free
60 Space

50

40

30
In-
20 buildin
g
10

0
1
2
4
5
6
7
9
10
11
13
14
15
17
18
19
21
22
23
24
26
27
28
30
31
32
34
35
Distance (metres)

The graph above shows the results of the free-space and in-building propagation models. The ‘free-
space’ line assumes uninterrupted line-of-sight behaviour, and is obviously the upper-bound of what could
be expected in practice: it will not be possible to improve on these figures in practical networks. The in-
building plot shows a much steeper decline of data rate with distance, as one might expect: generally it
assumes non-line-of-sight, and many multipath reflections. The general form of this curve is well-
accepted: the slope varies depending on the type of environment. Our assumption is for a propagation
factor γ of 2.7 (estimates vary from 2.0 for free-space to 3.5 in extremely challenging environments), and
we judge this reasonable for a building with few ‘thick’ internal walls, but with desks, cubicle partitions and
‘thin’ office walls and doors. A modern office building might be typical of this, whereas buildings
constructed with internal walls of brick or concrete, or containing large metal objects would warrant a
higher γ factor.

Applying noise and calculating expected data rates for 802.11n

To the raw calculations above, we add a factor to simulate random noise in the environment. This is a
gross simplification of a typical enterprise environment, where there will be sources on non-random noise,
and other Wi-Fi devices, which will in effect interfere with the channel under consideration, but it serves to
give an indication of expected performance.

The methodology, explained above, is to apply a random variation in the range +/- 6dB to the mean values
for in-building propagation above. This gives the figure for a single spatial stream in the plots below.

To model multiple spatial streams (SS), a number of independent random events are calculated, one for
each SS in the system. These are then added using a variety of weightings, as it is not reasonable to
expect perfect RF isolation of the spatial paths: adding a second SS will cause some interference and
degradation of the first. Subtracting from this effect, however, the transmitter can use feedback from the
receiver to characterize the combination of channels and adjust it’s antenna weightings. Overall we
expect these effects to balance: the calculations above expect that the second and third spatial streams
are directly additive to the overall data rates.

37 802.11n Theory and Practice Aruba Networks

 Data rates varying with distance Free
Space
250
In-
200 building
Data rate (Mbps)

150 1 SS,
20MHz
with
100 noise
2 SS,
20MHz
50 with
noise
3 SS,
0 20 MHz
with
1
3
5
6
8
10
12
14
15
17
19
21
23
24
26
28
30
32
33
noise
Distance (metres)

The diagram above also includes a simplification, in that it assumes non-line-of-sight, but with significant
multipath: good conditions for MIMO with SDM. Our practical experience in our own office building
suggests that actual data rates can be lower at short distances from the access point, because there is
more likely to be line-of-sight, especially with ceiling-mounted antennas. This means that at short range,
line-of-sight negates the MIMO effect, and data rates are similar to single spatial stream figures. As
distance increases, the dominant line-of-sight/near-line-of-sight signal is attenuated, promoting multiple,
more isolated RF paths, and overall data rates increase.

Now the results must be matched to actual 802.11 data rates, depending on different coding
combinations. The plot below shows this for 802.11a, which can be used as a benchmark for comparison
with the 802.11n results.

Aruba Networks 802.11n Theory and Practice 38

 Data rates varying with distance - 802.11a Free
Space
80

70

60 In-
Data rate (Mbps)

building
50

40

30 802.11a
raw
data
20
rate
10
802.11a
0 actual
data
1
3
4
6
7
9
11
12
14
15
17
19
20
22
23
25
27
28
30
31
33
35
rate
Distance (metres)

Using the same methodology we model 802.11n data rates for different combinations of spatial streams
and channel bandwidth.

Free
802.11n data rates varying with distance Space

80 In-
building
70 1 SS,
20MHz
60
Data rate (Mbps)

2 SS,
50 20MHz
3 SS,
40 20MHz
30 1 SS,
40MHz
20
2 SS,
10 40MHz
3 SS,
0 40MHz
1
3
5
6
8
10
12
14
15
17
19
21
23
24
26
28
30
32
33

802.11a
Distance (metres) (20MHz)

The results demonstrate that even with one spatial stream, 802.11n gains range over 802.11a/g. This can
be ascribed to a series of minor PHY-layer improvements (beamforming is not included in this model
because it is not part of the first release of 802.11n, and it is not incorporated in 2007-releases of silicon).
The extent of this improvement of 802.11n range over 802.11a is perhaps 5%.

39 802.11n Theory and Practice Aruba Networks

Also, the effective range of 802.11n (the distance at which the lowest data rate becomes unreliable)
increases slightly with the number of spatial streams: maybe 20% with the second spatial stream. While
this may seem to be a small increase, it will allow an increase in the spacing of access points, and a
reduction in their number of perhaps 30-50%.

Note that as this model deals only with range and data rates at the PHY layer, the achievable throughput
will be less than indicated above – sometimes considerably less, especially when older 802.11a/b/g clients
are present in the cell.

The table below shows the estimated contribution of different 802.11n features in extending range (range
here is defined as the distance at which the error rate at the lowest defined data rate becomes excessive:
the range is compared to 802.11a).

Feature Estimated Notes

Improvement
Increased channel bandwidth, 5% More subcarriers are used, but the improvement is directed
20 MHz channel, modulation to a higher data rate rather than increased range.
& coding improvements
Increased channel bandwidth, 5% As above.
40MHz channel, modulation &
coding improvements
Diversity effects from multiple 7% Transmitting identical signal from 2 or 3 antennas at the
antennas, 2x1/3x1 AP, and receiving on one (relies on the AP using receive
processing to support the reverse link). This improvement
may not be realized in 2007.
Diversity effects from multiple 15% As above, except the client can combine received signals
antennas, 2x2, 3x2 from both antennas, and transmit identical signals on both.
This improvement may not be realized in 2007.
Transmit beamforming 2x1 10% This will not be implemented in silicon (at least with inter-
vendor interoperability) in 2007. (Not additive with
diversity).
Two spatial streams 20% (This is not additive with diversity effects or beamforming.)
Three spatial streams 30% (This is not additive with diversity effects or beamforming.)

The practical solutions available in 2007 should benefit from diversity effects and the two-spatial-stream
opportunity, giving the range estimates (based on two spatial streams) used through this paper for a 2007
greenfield 802.11n network.

Note that MAC enhancements do not affect the maximum range of an 802.11n signal, although they do
serve to increase the overall throughput at any particular AP to client distance.

Aruba Networks 802.11n Theory and Practice 40

6.3 Glossary of terms used in this note
802.11 is replete with acronyms, and 802.11n has contributed more than its share to the lexicon. The
following is a subset of terms used in the standard, but should cover all terms used above.

A-MPDU Aggregate MAC protocol data unit

A-MSDI Aggregate MAC service data unit
AP Access point
CTS Clear to send
DTT Downlink transmission time (Power-save multi-poll)
HT High throughput
HT-GF-STS High throughput greenfield short training symbol
HT-STF High throughput long training field
HT-SIG High throughput signal field
HT-STF High throughput short training field
LDPC Low density parity check
L-LTF Non-high throughput (= legacy) long training field
L-SIG Non-high throughput (= legacy) signal field
L-STF Non-high throughput (= legacy) short training field
LTF Long training field
MC Mobility Controller (an Aruba product controlling access points)
MCS Modulation coding scheme
MIMO Multiple input, multiple output
MISO Multiple input, single output
MRQ Modulation coding scheme request
MTBA Multiple traffic ID block acknowledgement
NAV Network Allocation Value
PCO Phased coexistence operation
PSMP Power save multi-poll
RF Radio frequency
RIFS Reduced inter-frame spacing
RTS Request to send
SISO Single input, single output
SM Spatial multiplexing
SNR Signal to noise ratio
STBC Space time block code
STBC/SM Space time block code / spatial multiplexing
TRQ Training request
TxBF Transmit beamforming
UTT Uplink transmission time (Power save multi-poll)
WLAN Wireless Local Area Network

References
The IEEE is the fount of 802.11 standards. While it does not publish standards until final ratification, much
useful information may be found here: http://grouper.ieee.org/groups/802/11/ .In particular, there is a
useful summary of each forthcoming amendment (currently under ‘WG Info / 802.11 Quick Guide’) and a
current estimate of schedules (use the search box for ‘timeline’).

The information in this paper was taken from and checked against P802.11n draft 2.0.

The Wi-Fi Alliance is working on certifications for 802.11n. The WFA web site is here: http://wi-fi.org/

41 802.11n Theory and Practice Aruba Networks

About Aruba Networks, Inc.
Aruba Networks provides an enterprise mobility solution that enables secure access to data, voice and
video applications across wireless and wireline enterprise networks. The Aruba Mobile Edge Architecture
allows end-users to roam to different locations within an enterprise campus or office building, as well as
to remote locations such as branch and home offices, while maintaining secure and consistent access to
all of their network resources. Using the Aruba Mobile Edge Architecture, IT departments can manage
user-based network access and enforce application delivery policies from a single integrated point of
control in a consistent manner. Aruba’s user-centric enterprise mobility solution integrates the ArubaOS
operating system, optional value-added software modules, a centralized mobility management system,
high-performance programmable mobility controllers, and wired and wireless access points. Based in
Sunnyvale, California, Aruba has operations in the United States, Europe, the Middle East and Asia
Pacific, and employs staff around the world. To learn more, visit Aruba at http://www.arubanetworks.com.

© 2007 Aruba Networks, Inc. All rights reserved. Aruba Networks and Aruba Mobile Edge Architecture are
trademarks of Aruba Networks, Inc. All other trademarks or registered trademarks are the property of their
respective holders. Specifications are subject to change without notice.

AN000WP80211n-2.7

Aruba Networks 802.11n Theory and Practice 42