|
Functional MIMO Testing
For 802.11n
by Richard Lu and Jose Graziani, Azimuth Systems
Multiple-input multiple-output (MIMO) technology, the heart of the upcoming IEEE
802.11n standard, is the foundation for the next generation of Wi-Fi products.
With the promise of greater throughput and range capabilities, 802.11n will
enable new voice, video, and data applications that demand greater performance.
It will be up to manufacturers to ensure that their MIMO
products deliver the robust interoperability, performance, and quality of
service required by these new applications. For that reason, proper test and
measurement of device and network capabilities are critical to ensure the
success of this growing market.
802.11n Test Challenges and Solutions
802.11n, with the presence of multiple transmit (Tx) and receive (Rx) chains,
significantly increases the complexity of the physical (PHY) layer over the
802.11a/b/g standards supporting a much simpler single-input single-output
(SISO) architecture. For example, in 802.11n, there are more than 300 modulation
coding schemes (MCS), which are functions of channel bandwidth, number of
spatial streams, and types of modulations. In contrast, 802.11g has 16 different
data rates. The complexity of the 802.11n media access control (MAC) layer also
has increased substantially compared to that of 802.11a/b/g technologies.
Ensuring that the complex MAC and PHY layers operate properly in an 802.11n
system requires thorough functional testing of device-to-device and
device-to-network configurations.
For real-world deployment, delivering
robust interoperability between 802.11n MIMO and legacy 802.11a/b/g SISO devices
is critical to maintaining the success of the Wi-Fi market. Manufacturers are
designing MIMO devices to address varying performance requirements.
Devices can be tuned under different environmental conditions
to provide maximum throughput, maximum range, or a compromise between the two.
This consideration of environmental variability, which does not exist in
development of SISO products, adds tremendous complexity to the development and
testing of MIMO devices. For that reason, thorough characterization of MIMO
devices in a variety of environments is required to verify that their
performance adequately meets market requirements.
Functional device-to-device testing and device-to-network
testing in a controlled static channel environment provide the necessary
conditions for rapid development and troubleshooting of layers 2 and 3 of an
802.11n system. This testing is required to verify the functionality,
conformance, and interoperability of 802.11n devices and networks that are
independent of layer 1 influences. Examples are basic association handshakes and
security negotiations.
Functional testing includes the measurement of higher-level
performance parameters that also are not dependant on RF channel
characteristics. One example of functional performance testing is the
measurement of system throughput with different data-encryption types.
While functional performance testing does not provide a
complete characterization of real-world performance, it is a critical step to
verifying proper system operation. Measuring the real-world performance of MIMO
systems in multipath and fading environments requires dynamic channel emulation
using a purpose-built channel emulator.
Functional testing validates operation, conformance, and
interoperability of MIMO devices; dynamic channel emulation addresses the
performance. The conduct of both test methodologies is required to verify that
802.11n devices function properly, provide robust interoperability, and deliver
the performance necessary to satisfy consumer demands.
Functional MIMO Testing
Functional testing of MIMO-enabled devices provides an accurate assessment of
hardware and software functionality, standards conformance, and interoperability
between MIMO and SISO devices as well as basic operation of MIMO performance
algorithms. It involves four critical components:
Protocol Testing: Verifies that device operation conforms to standard
specifications across communications states. This type of testing requires
extensive packet capture, filter, and 802.11n decode analysis capabilities.
Interoperability Testing: Verifies that access points (APs) and client devices
designed by different manufacturers supporting next-generation 802.11n as well
as legacy 802.11a/b/g technologies work well together.
Mobility Testing: Verifies operation of 802.11n client devices in mobile
operating conditions. Examples of such tests include throughput vs. range and
AP-to-AP roaming.
Application-Level Feature Testing: Verifies that the device delivers a quality
user experience. Test examples would be basic data throughput and voice call
quality measurement.
Effective functional testing of MIMO devices is conducted in
an RF-controlled, static channel environment, ensured by placing DUTs in
enclosures that isolate them from unwanted RF interference. Between the
enclosures, RF cables connect all the Tx and Rx ports of the devices. A static
channel simulator can be used in the RF path to create a stable environment in
which the channel maintains consistent output power and undergoes limited to no
fading.
While a static channel is not the best representation of a
real-world environment where the channel conditions are constantly changing, it
provides a basic test environment to focus on the complexity of the 802.11n MAC
and protocol layers. Testing devices in the simplified and stable RF conditions
provided by a static channel allows engineers to concentrate on features and
functions that are not affected by environmental variability.
Abnormalities in device behavior revealed through testing in a
static channel environment can be attributed directly to the operation of the
device itself and not to the ever-changing environmental conditions. Since these
abnormalities are not impacted by RF channel conditions, identifying and
diagnosing them are greatly simplified.
Contrasting Approaches to Testing
Two common approaches to functional MIMO testing are do-it-yourself and
standardized automated.
Do-it-Yourself Testing
Many do-it-yourself (DIY) test solutions are conducted over-the-air since they
provide the quickest way to start testing. While these basic solutions may be
low cost, the value of testing also is low because testing in an uncontrolled RF
environment produces inaccurate results that cannot be repeated.
More advanced DIY test solutions will do conducted testing.
They involve the design and implementation of custom test setups using disparate
hardware and software components including RF attenuators, switches, and packet
capture. However, the increased complexity of DIY solutions that connect to and
use a properly controlled RF environment significantly increases installation
and support costs.
Since the expertise of Wi-Fi device manufacturers is in the
development of products and not necessarily in the testing of them, DIY
solutions often start as ad hoc tests. As such, the solutions are not
architected to provide the scalability necessary to cost-effectively evolve from
basic to advanced setups capable of extensive functional testing of future
products and services. So while the initial expenditure of DIY test setups may
be low, the cost of a robust test setup will quickly exceed that of standard,
automated test platforms when hardware expenses and engineering resources are
accurately accounted for.
DIY methods often rely on manual test execution since
management and automation software that controls nonstandard test components
from multiple vendors generally is not commercially available. Furthermore, such
software would be very difficult and costly to develop in-house.
In the development of 802.11a/b/g products, conducting
thorough functional test coverage using manual DIY setups dramatically increases
the time to develop a quality product. The large increase in the operational and
performance complexity of 802.11n devices has significantly increased the number
and complexity of functional tests required to verify the proper operation of
802.11n systems.
In addition, DIY setups lack standard test data management and
instead must rely on custom data capture and analysis tools. These tools often
dont have the detailed information analysis capabilities required for robust
problem identification and debug, which reduce the effectiveness of internal
engineering efforts.
 Purchasing
and supporting expensive DIY test setups with custom data management
capabilities reduce the likelihood of supply-chain partners supporting the DIY
test approach and joint product engineering analysis. This inability to actively
engage supply chain partners in product test, problem identification, and debug
prevents a vendor from taking advantage of significant engineering efficiencies.
The result is an increase in the overall time to market, cost of the product,
and support cost when problems are found in the field.
As the complexity in the functionality of wireless devices and
networks continues to increase, development and testing using DIY test methods
are not scalable in terms of both time and money.
Standardized, Automated Testing
Standard test platforms are purpose-built to offer functional test capabilities
at lower cost and complexity than alternative DIY solutions. Effective
standardized test platforms will integrate seamlessly with RF isolation chambers
to ensure that accurate and repeatable test results are produced.
The use of test automation and standard performance test
scripts significantly reduces the time required to test 802.11n devices.
Furthermore, with the increased functional and performance complexity of 802.11n
devices, automated test is the only way to achieve the test coverage necessary
to produce high-quality 802.11n products.
What You Can Learn
A core test of an 802.11n network is the analysis of an association process
between an 802.11n AP and an 802.11n station (STA). A packet capture of the
association process is used to verify that the high-throughput (HT) capabilities
parameters are set correctly on the AP or STA. A functional test would involve:
1. Initiating an 802.11n packet capture
prior to beginning AP and STA association.
2. Initiating AP and STA association.
3. Comparing the trace file generated through step 2 and parsing it against the
required filter string to check for the existence of the HT capabilities
parameters.
Figure 1 presents a packet capture trace from an 802.11n
association process.
Figure 1. Packet Capture of an 802.11n
Association Process
Click here to see
larger image |
During an association process of Wi-Fi devices, their
capabilities are negotiated between the AP and STA. It is very important that
each device advertise the correct capabilities to ensure interoperability with
the other device. Packet analysis allows engineers to pinpoint protocol issues
when a given function does not work properly. Figure 1 shows a successful
association process.
Functional testing also can measure higher-level performance
such as throughput and roaming. In the case of roaming performance, the
functional test will measure the time that it takes a client to roam between
multiple APs. This provides a quick way to assess whether or not roam times fall
within the guidelines established by IEEE 802.11r.
Figure 2 presents the results of a smooth roaming
benchmark test conducted on an 802.11 Draft-n STA using the Azimuth ADEPT-n MIMO
Functional Test Platform. The plot shows the distribution of more than a
thousand roams by a station between two APs. This is a very normal distribution
where 50% of the roams occur in fewer than 35 ms and more than 97% of the roams
occur in fewer than 50 ms, a critical roam time specified in 802.11r.
|
Figure 2. Plot of 802.11n Client Smooth
Roam Test Results
|
Summary
The increased complexity of MIMO-enabled 802.11n products is driving the
requirement for extensive testing. Functional testing and its complement,
dynamic channel emulation, are critical components of a rigorous 802.11n test
strategy. Functional device-to-device testing and device-to-network testing in a
controlled static channel environment enable repeatable checking of layer 2 and
layer 3 of an 802.11n system to accurately assess basic device hardware and
software functionality, standards conformance and interoperability between MIMO
and SISO devices, and basic operation of MIMO performance algorithms.
Manual DIY approaches, encumbered by high implementation and
support costs, increased test time, and limited engineering efficiencies, will
not address the MIMO functional test requirements and business challenges that
engineering teams face. In contrast, standardized, automated 802.11n MIMO
functional test methods leveraging increased test coverage, reduced test time,
and engineering efficiencies enable engineering teams to deliver higher quality,
higher performance 802.11n products to market faster with less cost.
About the Authors
Richard Lu is the WiFi product line manager at
Azimuth Systems. Before joining the company in 2003, Mr. Lu started the design
verification lab at Coppercom and was an applications engineer at Zarak Systems,
later acquired by Spirent Communications. The 10-year veteran of the
telecommunications and networking industries earned a B.S. from University of
California. e-mail:
Richard_lu@azimuthsystems.com
Jose Graziani is an engineering manager
for Azimuth Systems. He held senior software engineering roles at several
companies including Avaya, 3Com, Quantum, and IFS International prior to joining
Azimuth five years ago. Mr. Graziani received a B.S. in electrical engineering
from Rensselaer Polytechnic Institute. e-mail:
jose_graziani@azimuthsystems.com
Azimuth Systems, 31 Nagog Park, Acton, MA 01720, 978-263-6610
FOR MORE INFORMATION
on 802.11n compliance
requirements
www.rsleads.com/708ee-198
|