Wireless
Last Updated: Feb 1st, 2008 - 10:12:17
|
See this article in our Digital Edition Download a pdf of this article (right click to save the pdf)
Internet service providers are quickly adapting to the
needs of the mobile subscriber that must have access to the Internet
and/or their email outside of their home or office. WiFi “hot-spots”
were a good interim solution (particularly if you happened to own a
Starbucks franchise!), but this solution no longer satisfies the need
of the subscriber who wishes to be connected all the time and anywhere.
3G service providers, in an effort to fill this need, adapted their
networks to provide Internet connectivity to mobile handsets. For some,
this was a step up. However, bandwidth limitations and the small size
of the devices make navigating the Internet or sending/receiving emails
slow and cumbersome.
Municipal wireless networks are starting to pop up in many cities
around the world, such as Singapore, Milan, London, as well as in
cities right here in the United States, such as Minneapolis, Portland
and Concord, CA. These mobile area networks (MANs) are beginning to
provide the “on all the time, anywhere (in the city, anyway)”
connection subscribers need. In addition, they provide the true
broadband experience that the mobile subscriber demands.
The complex nature of these systems and devices pose a unique challenge
to testing labs. In particular, the network and distribution equipment
is usually comprised of several OEM devices with proprietary equipment
that work together but serve different functions. The test lab must
consider a multitude of test standards and apply them in such a way as
to ensure compliance with all of them, while at the same time taking
care to not over-test. Because of the unique application and
installation of these devices, they are in some cases required to
comply with standards that they are otherwise not designed to meet.
We’ll explore some of those situations in this article.
What is Muni Wireless?
At risk of sounding over simplistic, municipal wireless local area
networks are more robust and widespread versions of the local area
networks (LANs) that we are all accustomed to in our homes and offices.
A network consisting of nodes or access points installed at various
locations around a city allow users to connect to them either for a fee
or, in some cases, at no charge. These new wireless ecosystems free the
user from a single “hot-spot” and allow them to connect anywhere at
anytime.
There are various technologies that now exist, such as traditional
802.11 WiFi and 3G, as well as new technologies such as WiMax slated
for rollout in the U.S. by Sprint/Nextel in late 2007. Each poses its
own unique set of issues for the testing lab; however generally
speaking they must be treated the same.
For this article, we’ll limit the discussion to WiFi muni access points
(also referred to as a WiFi node) operated and powered from a cable
head end for installation in the U.S. and Europe. These devices are
comprised of nodes that usually provide traditional wired cable TV,
high-speed Internet and VOiP to subscribers in a neighborhood or an
entire city. In the case of muni wireless LANs, these devices may be
modified to accept a standard wireless access point within the
ruggedized and hardened enclosure of the node, thereby providing a
wireless solution to a mobile subscriber.
What Are the Testing Challenges?
Testing is not difficult, since the tests are routine. However, the
difficulty is in ensuring compliance. This challenge is three fold.
First, it is essential to carefully review the functions of the
equipment, the applications and the installation to properly identify
all of the product standards and test methods that may apply. Since
this is a complex and composite device, you are dealing not only with
cable network equipment that must comply with EMC and product safety
requirements unique to these devices, but also a wireless device that
has its own set of requirements that include EMC, product safety and
wireless standards.
Second, the composite device has to be tested to all of the
requirements, which in some cases may mean that the wireless component
will have to be evaluated as network equipment and the network
components tested to wireless EMC requirements. These components were
likely not designed with the other standards in mind; therefore
compliance to them can prove to be difficult.
The last challenge is to develop the most efficient test plan that
takes into consideration all of the applicable standards, such that all
of the requirements are satisfied while not over-testing the device.
Here in the U.S., a WiFi muni node is simply a digital device that, in
the eyes of the Federal Communications Commission (FCC), is subject to
FCC Part 15 Subpart B, Class B verification (if in a residential
environment). It is also an intentional radiator and must comply with
FCC Part 15 Subpart C. If each component of the device was designed
properly to meet its own set of requirements, the device as a whole
should not have a problem complying with the requirements, so it would
seem that there isn’t a challenge here. The challenges however are
self-imposed by the manufacturer, as discussed below.
FCC Testing Challenges
A WiFi node shouldn’t have a problem meeting FCC requirements if each
component is designed to meet the specific requirements that apply to
it. The EMC challenge here is that compliance to FCC Part 76
(specifically 76.606 (a)(12) with respect to 76.609(h)) is required for
installed cable equipment and is obligatory to the cable provider and
not the manufacturer of the equipment. This is because potential
interference is due to RF leakage at RF connections in the system as a
result of poor installation and degradation of the equipment and
connections over time. The cable provider must make routine in-situ
measurements to ensure that its installed cable equipment continues to
comply with 76.605(a)(12).
To ensure that cable providers initially receive a quality product,
they will usually require evidence from the manufacturer that the
equipment was designed to meet the leakage requirements of
76.606(a)(12). Therefore, when the manufacturer contacts a test house
to request FCC Part 15 verification services, they will usually also
mention the Part 76 requirements as part of the request.
There are two problems here for the test house and the manufacturer.
First, the test methods given in 76.609(h) do not lend themselves to a
traditional testing site. The test methods and set up differ greatly
from those of the typical FCC Part 15 Subpart B test. How is that
handled by the test house? Keep in mind that the obligation of
compliance to this rule part is not obligatory to the manufacturer and
that all the cable provider is looking for is assurance that the device
is likely to comply. Therefore, this allows for some latitude in how
the test can be performed in order to provide that level of assurance.
The second problem is that some components of the composite system were
not designed to meet the strict field strength requirements of
76.606(a)(12).
Modifying the stated test method such that it is more conducive to a
test house is a reasonable way to deal with the first problem and gain
assurance that the device will comply when tested per the proper
in-situ test methods. FCC Part 76.609 (h) specifies the use of dipole
antennas positioned 3 meters from the equipment under test (EUT) and
directly below it, while maintaining a 3 meter separation from the
dipole antenna to the ground.
As part of developing the test plan, the FCC Part 15 Subpart B
requirement should also be considered to reduce the amount of testing.
There are at least two options for modifying the test plan. The first
is to use dipole antennas affixed to an antenna mast on a typical open
area test site, with the height set such that the antenna is 3m above
the EUT and at a 3m diagonal distance away from the EUT. This is the
closest way to simulate the test method on a standard test range, since
the antennas are positioned orthogonally in about the same position as
called for in the specified test method of 76.609(h), while also
reducing the +6dB effect of the ground reference plane. However, this
is very cumbersome if there are many discrete frequencies to measure,
since the dipoles will have to be adjusted for each data point. Also,
the requirements of FCC Part 15 Subpart B would be very difficult to
consider as part of this test method.
The most efficient way to provide assurance that the EUT can meet the
requirements of Part 76, while ensuring that it does comply with the
mandated FCC Part 15 Subpart B radiated emissions requirements of
15.109, is to create a composite set of limits and perform the test as
you would normally run an FCC Class B radiated emissions sweep at 3m.
Figure 1 is the standard FCC Part 15.109 Class B sweep against the
class B limit line. Figure 2 is the same sweep with the Part 76 limit
applied and Figure 3 is again the same sweep with a limit line that
represents the worst case between Parts 15 and 76.
Figure 1: FCC Part 15.109 Class B Limit Line
Figure 2: FCC Part 76.606(a)(12) Limit Line
Figure 3: FCC Part 76/15B Composite Limit Line
As shown in both Figures 2 and 3, the band between 54MHz and 216MHz is
very tight and difficult to meet, even for equipment that was designed
with Part 76 in mind. Testing to and complying with this composite
limit would represent a worst case scenario and ensures compliance to
FCC Part 15.109as well as assuring the cable carrier that the equipment
will meet the Part 76 requirements when provided to the carrier for
installation.
However, even with the +6dB effect of the ground plane, devices that
are required to meet the leakage requirements of FCC Part 76 usually
will meet these limits under the above test conditions. This is because
theses devices are RF and environmentally hardened to a much greater
extent than commercial grade equipment. If failures are encountered,
they are usually a result of poor RF connections and/or poor cables.
European Union Testing Challenges
The composite nature of a WiFi node adds additional testing challenges
for the European market. Care should be taken to evaluate the intended
installation and functionality of the device in determining the proper
route to conformity.
Because this device is both cable network equipment and an intentional
radiator and is usually rated at >50 Vac or 75 Vdc, three separate
Directives apply. The EMC Directive (89/336/EEC, soon to be
2004/108/EC), the R&TTE Directive (99/5/EC) and the Low Voltage
Directive (2006/95/EC, formerly 73/23/EEC) all apply for these devices.
Complying with the essential provisions of these directives involves a
multitude of product standards that must be considered to cover EMC,
radio spectrum and product safety requirements.
EMC
For EMC, the following product standards apply, assuming the WiFi node is 802.11a/b & g:
-
EN 50083-2:2001 - Cable networks for television signals, sound signals
and interactive services - Part 2: Electromagnetic compatibility for
equipment;
-
EN 301 489-1: Electromagnetic compatibility and Radio spectrum Matters
(ERM); ElectroMagnetic Compatibility (EMC) standard for radio equipment
and services; Part 1: Common technical requirements;
-
EN 301 489-17: Electromagnetic compatibility and Radio spectrum Matters
(ERM); ElectroMagnetic Compatibility (EMC) standard for radio equipment
and services; Part 17: Specific conditions for 2,4 GHz wideband
transmission systems and 5 GHz high performance RLAN equipment.
Different or additional requirements may exist for different wireless protocols. These should be carefully considered.
The testing challenges in EMC are in applying test methods, levels and
limits to a composite device consisting of components that were not
originally designed to meet those requirements. For example, the
wireless component (OEM access point) was designed to meet a wireless
EMC standard (i.e. EN 301 489-17) and not that of network cable
equipment. The same can be said for the cable component(s). It was
designed to meet the cable network equipment standard EN 50083-2. These
standards are very different in how radiated emissions (or
disturbances) are measured and how susceptibility is determined. EN
50083-2 requires that radiated disturbances on all RF ports, mains
leads and any single or multiple wire leads be measured using “coupling
devices” and not the standard radiated field strength method as called
for by EN 55022 (per EN 301 489-1). Absorbing clamps and EM injection
clamps are commonly used to measure the potential disturbances.
For traditional cable network equipment that has been properly
designed, this isn’t a problem. However, the access point component(s)
of the device was not designed to meet this requirement. These tests
require a very good RF enclosure and good quality hard-line or cables.
The RF enclosure (as the name implies) is designed to suppress RF
egress and leakage resulting from the cable network transmission
frequencies and data rates, but not usually the high data rates
associated with Ethernet traffic from a wireless access point type
device. This can prove to be a very challenging test to pass.
Conversely though, the composite device as a whole would have a better
chance of meeting the requirements of EN 301 489-1 & -17 if
designed to meet EN 50083-2.
Product Safety
In addition to the wireless and EMC considerations for the EU’s CE
mark, similar convergence of standards occurs in the product safety
area of compliance. IEC/EN 60950-1 is generally the preferred standard,
although some manufacturers choose to use IEC/EN 60065 7th Ed. Because
cable network powered equipment transfers data as well as audio and
video signals, either 60950-1 or 60065 may be applied, at the choice of
the manufacturer. IEC Guide 112, Guide on the safety of multimedia
equipment, states:
“The following fundamental principle was adopted as the basis for this guide:
Equipment complying with the requirements of IEC 60065, as well as
equipment complying with the requirements of IEC 60950, is considered
to be safe when used on its own.
Such equipment when interconnected in multimedia systems in accordance
with the installation instructions is also considered to be safe.”
In addition, IEC/EN 60728-11 applies as well. Clause 8.1.2 requires
that network-powered equipment comply with either 60950-1 or 60065.
This standard also limits the maximum voltage of the network power to
65 Vac or 120 Vdc.
This standard technically applies to the network and not individual
equipment, but the equipment manufacturer should have their products
evaluated to this standard so that the cable operators buying the
equipment can be assured that it will comply when installed into their
network, much in the same way they would for FCC Part 76.
The EN equivalent of IEC 60728-11 is EN 50083-1+A2; the cessation date
of this standard is 1/04/2008, when it is replaced with EN
60728-11:2005.
Radio
Testing for compliance to the appropriate radio spectrum product
standard (EN 300 328 for 802.11b&g, and EN 301 893 for 802.11a, in
the case of this example) is rather straightforward. The RF hardened
enclosure is a superior enclosure to those of standard access points;
therefore cabinet radiation is of little concern even at the
measurement frequency extremes. Failures, if encountered, can be
mitigated as you would normally do for any other wireless access point.
Test Plan
The test plan complexity lies predominantly in the EMC requirements.
The safety standards are mostly an either/or decision at the
manufacturers discretion, except for the application of IEC 60728-11.
This is also the case for the wireless test plan. There is no real
overlap that allows for cost and time saving measures between EN 300
328 and EN 301 893. The device must be evaluated in whole to both
product standards.
As noted above, one of the most complex pieces to this compliance puzzle is to build an EMC test plan that:
-
Ensures presumption of conformity across all directives and product standards;
-
Maximizes resources to prevent duplicate testing, keeping costs to a minimum for the client;
-
Schedule resources to prevent unnecessary setup and teardown of the system, and to promote efficient movement through the lab.
It helps to build a matrix to identify all of the product standards and
list all of the basic test methods below them to identify overlap. From
this, commonalities between product standards can be identified. By
using logic and sound engineering judgment, the test plan can be
reduced to take into account these commonalities. (See Table 1 for an
example.)
Summary
Ironically, it is the wireless certification element of the test plan
that is the simplest part of the regulatory compliance test plan for
the WiFi muni node. But when the entire ecosystem of the municipal
wireless local area network (systems and devices) is examined, the
complex nature of these products poses a unique challenge for
regulatory test labs. Testing the network and distribution equipment
that is usually comprised of multiple OEM devices with proprietary
equipment that works together but serves different functions is a
challenge for the lab management as they attempt to apply a host of
standards with overlapping test requirements.
As the expert, the test lab should take on a certain responsibility as
they guide their clients through this maze of different standards. The
test lab must consider the complexity of test standards and apply them
in such a way as to ensure compliance with all of them, yet taking care
to not test the product beyond the combined requirement of all of the
applicable standards.
Because of the unique application and installation of these devices,
manufacturers are, in some cases, required to meet standards that their
products are not otherwise designed to meet. It is up to the test lab
to properly identify all of the requirements with which these composite
systems must comply. It is up the manufacturer’s design team to ensure
that they comply. As in most cases, it is advantageous to have a
representative from a testing organization involved early in the design
process, and regularly during the cycle leading up to the point of
testing.
This regulatory complexity becomes even more pronounced when multiple
markets are served by the same product, such as North America and
Europe (as discussed here). If the same product is intended for Asia or
South and Central America, an even greater level of complexity is
experienced. Different countries apply varying requirements. Some
require in-country testing, some accept FCC or CE. Some interpret VoIP
as telecom (requiring additional regulatory compliance), and some do
not. Some have very strict wireless requirements, and some are very
lax.
In today’s environment of ever-increasing bandwidth demand from the
consumer through the cable providers to the manufacturers, these
regulatory challenges will become more and more prevalent. The new
economy (and the emergence of telecom infrastructure requirements in
the developing economies) ensures that manufacturers will continue to
develop products that will fall under multiple standards for their
intended markets. The regulatory professional must be able to
understand the concurrent application of all of these standards. n
Sam Wismer is the Technical Director of Advanced Compliance Solutions, Inc., and can be reached at swismer@acstestlab.com.
© 2007 Conformity
Top of Page
|