Testing CDMA and EV-DO Mobile Terminal Devices: Beyond Conformance
This white paper provides an overview of the methods used to test mobile devices deployed on commercial CDMA/EV-DO networks. Topics include several key areas of testing, both as mandated by industry standards bodies and implemented by commercial network operators and mobile device manufacturers.
White Paper 122
Testing CDMA and EV-DO Mobile Terminal Devices:
Beyond Conformance
September 2009
Testing CDMA and EV-DO Mobile Terminal Devices
White Paper 122
2 Spirent White Paper
SPIRENT
541 Industrial Way West
Eatontown, NJ 07724 USA
Email: sales@spirent.com
Web: http://www.spirent.com
Page Part Number: 79-002163, Version A0
AMERICAS 1-800-SPIRENT • +1-818-676-2683 • sales@spirent.com
EUROPE AND THE MIDDLE EAST +44 (0) 1293 767979 • emeainfo@spirent.com
ASIA AND THE PACIFIC +86-10-8518-2539 • salesasia@spirent.com
© 2009 Spirent. All Rights Reserved.
All of the company names and/or brand names and/or product names referred to in this document, in particular, the
name “Spirent” and its logo device, are either registered trademarks or trademarks of Spirent plc and its
subsidiaries, pending registration in accordance with relevant national laws. All other registered trademarks or
trademarks are the property of their respective owners.
The information contained in this document is subject to change without notice and does not represent a
commitment on the part of Spirent. The information in this document is believed to be accurate and reliable;
however, Spirent assumes no responsibility or liability for any errors or inaccuracies that may appear in the
document.
Testing CDMA and EV-DO Terminal Devices
Table of Contents
Spirent White Paper 3
Table of Contents
Scope ........................................................................................................ 4
Introduction............................................................................................... 5
CDMA/ EV-DO Conformance Test Specifications......................................... 6
Minimum Performance Testing ..................................................................... 6
Receive Diversity Testing .................................................................................... 7
RF Testing Specific to EV-DO Rev B ......................................................................8
Protocol and Interoperability Testing ............................................................ 9
More on Stage II testing...................................................................................... 9
Signaling Testing for EV-DO Rev B..................................................................... 12
Field Testing .............................................................................................. 12
Testing for the data-centric wireless world .............................................. 13
Location-Based Services (LBS) testing ........................................................ 13
Data Service Testing .................................................................................. 17
Performance Testing with Conformance Test Cases ................................. 19
Summary and Conclusion ........................................................................ 22
Acronyms................................................................................................. 23
Testing CDMA and EV-DO Mobile Terminal Devices
White Paper 122
4 Spirent White Paper
SCOPE
This white paper provides an overview of the methods used to test mobile devices
deployed on commercial CDMA/EV-DO networks. Topics include several key areas of
testing, both as mandated by industry standards bodies and implemented by
commercial network operators and mobile device manufacturers.
Testing CDMA and EV-DO Mobile Terminal Devices
Introduction
Spirent White Paper 5
INTRODUCTION
The internal complexity of a modern cellular device represents an incredible
achievement on the part of the wireless engineering community. It is a testament to this
that most cellular subscribers will never know everything involved in the little device
thrown in a pocket and taken for granted.
In the spirit of competition, the cellular industry long ago moved away from the specter
of proprietary systems by enacting technical specifications for cellular devices. Anyone
who wanted to build a cell phone could do it, as long as the device complied with the
specifications. As part and parcel of this initiative, test specifications were developed so
that proof of conformance could be made available.
Ownership of these specifications belongs to standards bodies made up of industry
participants. While the standards body is chartered to ensure that the technology works,
participants can have very different ideas on who should bear the more egregious
burdens of this insurance. In a strange twist of necessity, erstwhile competitors
suddenly become allies in deciding what technical traits should and should not be
strictly enforced; mobile device manufacturers band together to sway things to their
advantage, while the outnumbered network operators attempt to do the same. The result
is a compromise that serves its intended purpose, but does not necessarily represent
the best interests of any individual entity, including the subscriber.
Testing CDMA and EV-DO Mobile Terminal Devices
White Paper 122
6 Spirent White Paper
CDMA/ EV-DO CONFORMANCE TEST SPECIFICATIONS
MINIMUM PERFORMANCE TESTING
A few years ago, the test methodologies driven by the 3GPP2 were thought of in terms of
stages. “Stage I” testing, documented in the IS-98 specification (later called TIA-98) was
meant to ensure “minimum performance”. This vague phrase described testing receiver
and transmitter performance under specific conditions and quantifying results to
guarantee a specific level of performance.
For example, a receiver can be tested by measuring its ability to demodulate a given
signal. Disciplined testing demands a test that is both repeatable and realistic, which is
much more complicated than it sounds at first.
In addition to electrical noise that is always present, wireless signals are subject to a
phenomenon known as “fading”. To describe a very complicated topic in just a few
words, fading is the effect of reflected RF signals arriving at a receiver after having
traversed different paths. Because the signals arrive at different phase angles (based on
the distance traveled), the received signal is affected by constructive and destructive
interference. To make matters even more complex, this scenario is constantly changing.
For testing, the changes are accurately random while still being repeatable.
Table 1: Minimum Performance Test Specifications
Revision
Numbering
Numbering Document Diversity Testing Reference
3GPP2 TIA Qualcomm
App Note
Title
1x
C.S0011 TIA-98 Recommended
Minimum
Performance
Standards for
Dual-Mode
Spread Spectrum
Mobile Stations
80-V9141-1B cdma2000® Min.
Perf. For Rx
Diversity-Capable
Mobile Stations
Application Note
EV-DO Rev. 0
C.S0033 TIA-866 Recommended
Minimum
Performance
Standards for
cdma2000 High
Rate Packet Data
Access Terminal
80-V8941-1A Recommended
Minimum
Performance for
CDMA2000® 1xEV-
DO Rx Diversity-
Capable Access
Terminals
EV-DO Rev. A
C.S0033A TIA-866A Recommended
Minimum
Performance
Standards for
cdma2000 High
Rate Packet Data
Access Terminal
80-VF496-
11A
Recommended
Minimum
Performance for
CDMA2000® 1xEV-
DO Revision A Rx
Diversity-Capable
Access Terminals
Testing CDMA and EV-DO Mobile Terminal Devices
Test Specifications
Spirent White Paper 7
Revision
Numbering
Numbering Document Diversity Testing Reference
EV-DO Rev. B
C.S0033B TIA-866B Recommended
Minimum
Performance
Standards for
cdma2000 High
Rate Packet Data
Access Terminal
Receive Diversity Testing
In an effort to squeeze the most efficiency out of a given wireless channel, the cellular
world has begun implementing the Receive Diversity feature (sometimes abbreviated to
“RXDiv”). Wireless channels are never static, so the fading process described above
always affects any wireless path between a radio transmitter and a radio receiver.
Diversity is one way to combat this deleterious effect. In a Receive Diversity system, the
receiver has two RX antennas, creating two separate signal paths from the transmitter.
This is not as complex as more advanced antenna techniques, such as Multiple-In-
Multiple-Out (MIMO) or beamforming systems, but it does provide a means of signal
redundancy.
In Figure 1, the mobile device is equipped with RXDiv capability. It shows that even when
one of the two paths undergoes a seriously deep fade, the other is probably alright. In
the worst case, the RXDiv system provides a demodulated signal that is at least as good
as the better of the two transmitted paths. Note that RXDiv does not require any
additional capabilities on the transmitter side of the RAN, which makes it a desirable
“first step” in eventually moving towards more advanced antenna techniques. Note that
this is not presently a part of the CDMA/EV-DO roadmap.
Testing CDMA and EV-DO Mobile Terminal Devices
White Paper 122
8 Spirent White Paper
Figure 1: Receive Diversity System
To accurately test RXDiv devices, the test system must include a way to precisely control
the emulated effects on the two RF paths. This is not as simple as controlling two
independent paths, because the relationship between the paths is important. Simply
put, a concept called “correlation” quantifies the similarity between two paths as to the
effects of the RF environment. If two paths react to RF environmental changes in exactly
the same way at the same time, the inter-path correlation is 1. If one path bears no
relationship to the other, the correlation is 0. Let it suffice to say that the wireless
channel emulator used for RXDiv testing must have complete control over both the time-
varying fading and the correlation between the two paths.
Note that this is a very basic topical overview of minimum performance testing. The
knowledge required to successfully implement this testing can (and does) fill hundreds
of books on realistic RF environmental effects and the math required to model it.
RF Testing Specific to EV-DO Rev B
Conceptually, the radio link of RF EV-DO Rev B is little more than multiple EV-DO RevA
carriers being used in parallel. At the radio layer, Rev B adds seven test cases to the
industry standard Minimum Performance Conformance Tests. As with any other
conformance testing performed by a professional automated system, the conformance
test cases can be used again to automate device performance testing.
The Minimum Performance test cases specific to Rev B are modified versions of traffic
demodulation tests and pure RF testing (for instance, receiver sensitivity, dynamic range,
spurious emission testing, etc).
Testing CDMA and EV-DO Mobile Terminal Devices
Test Specifications
Spirent White Paper 9
PROTOCOL AND INTEROPERABILITY TESTING
“Stage II” testing was roughly defined as “signaling testing” and its test cases were
documented in specification number IS-898 (later named TIA-898). This included both
signaling and interoperability testing, and did not discriminate between RAN signaling
and inter-network-element signaling.
More on Stage II testing
A few years ago, the 3GPP2 realized that the mass of test cases contained in the TIA-898
specification was confusing to the intended audience. While the spec contained any test
case roughly based on protocol, cases were not delineated in terms of the intent of the
tests.
As an example, one test case was meant to ensure the proper working of the system’s
Call Forwarding function. This was intended to be an interoperability test, meant to
guarantee that a particular device and a particular network could successfully work with
each other. But the same document was being used to test mobile device designs to
ensure proper signaling. When a tester or engineer used TIA-898 to test cellular device
performance, they would come to the Call Forwarding test, dutifully connect the device to
the network or network emulator being used for testing, and step through the process.
In the best case, the device would display the requisite “Call Forwarded” message and
testing would continue. However, because the mechanism behind call forwarding sat
entirely within the network, this test merely proved that the MS received and handled the
Flash With Information message properly; a function that had already been tested
elsewhere.
In other cases, the engineer would begin to spend time “debugging” a problem that did
not exist on the device under test. To prevent this, the 3GPP2 moved to a system wherein
device signaling conformance and interoperability tests are kept separate. In addition,
specific test areas, such as SMS and MEID are called out in separate documents.
This breakdown is illustrated in Figure 2 and summarized in Table 2.
Testing CDMA and EV-DO Mobile Terminal Devices
White Paper 122
10 Spirent White Paper
Figure 2: Evolution of TIA-898 into Separate Specifications
Testing CDMA and EV-DO Mobile Terminal Devices
Test Specifications
Spirent White Paper 11
Table 2: Signaling Conformance Test Specifications
3GPP2
Numbering
TIA Numbering Document
C.S0031 TIA-898 Signaling Conformance Tests for
cdma2000 Spread Spectrum Systems
C.S0043 TIA-1035 Signaling Conformance Test
Specification for cdma2000 Spread
Spectrum Systems
C.S0060 TIA-1044 Signaling Conformance Test
Specification for Over-the-Air Service
Provisioning
C.S0061 TIA-1045 Signaling Conformance Test
Specification for Short Message
Service
C.S0073 TIA-1084 Signaling Test Specification for
Mobile Station Equipment Identifier
(MEID) Support for cdma2000 Spread
Spectrum Systems
C.S0044 TIA-1036 Interoperability Specification for
cdma2000 Air Interface
1x
C.S0062 TIA-1046 Signaling Conformance Test
Specification for cdma2000 Data
Services
Hybrid Mode
C.S0094 TIA-1157 Signaling Conformance Test
Specification for Interworking of
cdma2000 1x and High Rate Packet
Data Systems
EV-DO Rev. 0
C.S0038 TIA-919 Signaling Conformance Specification
for High Rate Packet Data Air
Interface
C.S0038A TIA-919A Signaling Conformance Specification
for High Rate Packet Data Air
Interface
EV-DO Rev. A
C.S0073-A TIA-1084-A Signaling Test Specification for
Mobile Station Equipment Identifier
(MEID) Support for cdma2000 Spread
Spectrum Systems
C.S0038B TIA-919B Signaling Conformance Specification
for High Rate Packet Data Air
Interface
EV-DO Rev. B
C.S0073-B TIA-1084-B Signaling Test Specification for
Mobile Station Equipment Identifier
(MEID) Support for cdma2000 Spread
Spectrum Systems
eHRPD/LTE
C.S0095 <PROPOSED> Signaling Test Specification for E-
UTRAN – cdma2000 Connectivity and
Interworking
Testing CDMA and EV-DO Mobile Terminal Devices
White Paper 122
12 Spirent White Paper
Signaling Testing for EV-DO Rev B
Industry standard testing for EV-DO Rev B devices requires some additional testing
beyond the test cases used for EV-DO Rev A. Specifically, there are new test functions to
verify correct routing update (basic service connection and handover functionality) and
for the MAC layer.
FIELD TESTING
“Stage III” testing is field testing. Although it is indispensible, field testing is neither
controllable nor repeatable. Field testing is an overview, meant to ensure that the
previous testing steps were performed correctly. When lab-based testing is approached
with rigor and forethought, field testing is merely a “sanity check”. When lab-based
testing is not done correctly, prolonged field testing becomes a very expensive
proposition.
Testing CDMA and EV-DO Mobile Terminal Devices
Testing in a Data-centric Wireless World
Spirent White Paper 13
TESTING FOR THE DATA-CENTRIC WIRELESS WORLD
LOCATION-BASED SERVICES (LBS) TESTING
Advances in technology brought further testing requirements, whether they were driven
by industry bodies or not. Because most CDMA/EV-DO devices incorporate GPS
receivers, the industry added cases specifically to test them. The TIA-916 standard
outlines this mechanism.
However, the original drivers for A-GPS testing were emergency-services location
capabilities mandated in some countries. In most of the world, the real driver for A-GPS
functionality is the promise of revenue based on commercial location-based services.
While commercial and emergency location services usually use different methodologies
(IP-based user-plane communication rather than standard-driven control plane
communication), the testing methodology is sound for both types of services. GPS
testing is far from a stochastically exact science, and the statistical testing methods
require a lot of data gathered over a long period of time. The methods used in TIA-916
are very useful in ensuring the quality of commercial services, and commercial testing
systems use similar methods to quantify device performance for commercial LBS.
Table 3: Location-Based-Services Test Specifications
3GPP2
Numbering
TIA Numbering Document
1x C.S0036 TIA-916 Recommended Minimum Performance
Specification for C.S0022-0 Spread
Spectrum Mobile Stations
1x User Plane User Plane <PROPRIETARY>
EV-DO Rev. 0 User Plane User Plane <PROPRIETARY>
EV-DO Rev. A User Plane User Plane <PROPRIETARY>
Autonomous Mode <PROPRIETARY>
Over-The-Air (OTA) <CTIA PROPRIETARY>
Understanding the issues in testing A-GPS-based LBS requires some understanding of
the concepts that make LBS work. The following is a very brief summary of a very
complex topic.
Testing CDMA and EV-DO Mobile Terminal Devices
White Paper 122
14 Spirent White Paper
A GPS satellite is an accurate clock that continually transits timing information. Any GPS
receiver (including cellular devices) uses this timing information to calculate its distance
from each satellite. Theoretically, each distance can be thought of as the surface of a
sphere. When the distances to three satellites are known, the device has enough
information to pinpoint its position at an intersection of the three spheres, as shown in
Figure 3.
Figure 3: GPS Positioning Based on Satellite Distances
In reality, the inherent inaccuracies can be thought of as “thicknesses” in the surfaces of
the spheres. These “thicknesses” are statistical, not discrete in nature, and are based on
a large number of factors out of the scope of this paper. The meaningful point is that A-
GPS testing is iterative, and the results analysis is statistical.
LBS testing provides an extreme example of the inherent dangers of relying on field
testing. As an example, let us look at what sounds like a reasonable field testing
scenario. Suppose a handset manufacturer has upgraded supporting firmware in a
device, and wants to find out if the firmware provides any real improvements. The
manufacturer has already invested a lot of time and money into collecting data in the
field.
In the first place, there is no such thing as a small set of A-GPS test data. Because of the
intrinsic variability of GPS testing, the approach is statistical rather than stochastic. In
most cases, the device must make a large number of calls. For each call, the true
position of the device is compared to the position as reported.
Testing CDMA and EV-DO Mobile Terminal Devices
Testing in a Data-centric Wireless World
Spirent White Paper 15
After data is gathered, each type of test normally requires two statistical calculations.
Typically, the requirements look something like this:
%90}67.0)(.{.
1
>><σεPLC
and
%90}95.0)(.{.
2
>><σεPLC
For example, the first equation can be read as follows: “There is a sufficient amount of
data gathered to say with a confidence level greater than 90% that the probability of a
distance error being less than σ
1
is 67% percent, or that the probability of a distance
error being greater than σ
1
is less than or equal to (1 – 0.67)” For the data in Figure 4, σ
1
is 25 m. It seems apparent that more than 67% of the data points fall within the required
distance, so this test probably passes, provided a sufficient number of calls were made
to calculate a confidence interval of more than 90%. In reality, this data would likely be
analyzed a second time to ensure that 95% or more of the data points fall within a
greater distance range (σ
2
).
Figure 4: Graphical Representation of a Data Set for A-GPS Testing
Testing an A-GPS receiver requires a GPS satellite field (a constellation of GPS satellites)
as well as a supporting network that includes the “number-crunching” element called a
Position Determination Entity (PDE). While some mobile devices are able to
autonomously calculate latitudes and longitudes, others offload the number-crunching
to the network-based PDE. In the latter cases, the mobile devices capture satellite
readings, send the captured data to the PDE, and receive latitude/longitude information
in return.
Testing CDMA and EV-DO Mobile Terminal Devices
White Paper 122
16 Spirent White Paper
There are several problems with trying to gather this kind of information in the field:
• A-GPS (or any GPS-based) testing usually takes place in a sparsely-populated area,
away from skyscrapers and mountains. Most busy labs are far from these isolated
areas, and travel to these areas is never cheap.
• Satellites will have changed position by the time a test is iterated. Contrary to
popular belief, GPS satellites are not geo-stationary, they are “geo-synchronous”.
This is a reference to the fact that they orbit the Earth roughly once every twelve
hours. In addition, their paths are not as predictable as one may assume. In other
words, one cannot develop a repeatable GPS field test by testing in the same place
at the same time of day.
• The CDMA environment may also have changed.
• The most efficient way to perform these tests is to calculate confidence intervals as
data is gathered. Otherwise, there is the risk of wasting time gathering more data
than is necessary.
The solution is an automated testing system that includes not only a complete emulated
CDMA network (including PDE), but a GPS satellite simulator as well. A commercial GPS
satellite emulator can accurately emulate multiple GPS satellites either artificially (by
providing control over individually satellites) or based on realistic testing scenarios (by
replicating the satellite field at a known place and time). Figure 5 shows a simulator’s
satellite field (or satellite constellation) in action. The outer ring of the diagram
represents the horizon, the inner ring represents all points on the circle seen at a
45°angle, and the center point represents the point directly overhead. The right side of
Figure 5 shows a close-up view of the apparent satellite movement (movement relative
to a fixed position on Earth) being tracked over time.
Figure 5: GPS Simulator Satellite Field View with Relative Movement
Testing CDMA and EV-DO Mobile Terminal Devices
Testing in a Data-centric Wireless World
Spirent White Paper 17
DATA SERVICE TESTING
Data service testing is an area where some of the best practices evolved without benefit
of industry standards. To be clear, the TIA-918 specification does call out a set of tests
meant to ensure IP functionality. For example, PPP negotiation and registration tests are
called out in this spec. Once again, standards bodies have ensured that the technology
works, but operators have taken it upon themselves to develop refined testing
standards. The reason is apparent; if data communications are not implemented well on
devices, operators bear the costs of 1) maintaining a network that spends much of its
time re-transmitting data and 2) supporting subscribers who have paid for data service.
It is obvious that this data service testing involves throughput measurements using
common protocols (for instance, FTP, UDP, HTTP, etc). What may not be as obvious is
that channel setup/teardown times, RLP operation, round-trip message delays, and
more can have expensive effects if they are not implemented well. Over the years,
operators have added requirements based on other lessons learned in deployment.
A prime example is what is called “data retry testing”. Data retry testing was born when
operators realized that they could not maintain complete control (the “walled garden”)
over the applications that would eventually be run on their networks. After IP capabilities
were available over the air, consumers demanded an increasingly varied array of
services, some of which would be delivered by third parties. The question that cannot be
answered without data retry testing is: “What happens when a popular wireless-based
service suddenly becomes unavailable?”
In the hard-wired networking world, bandwidth is relatively cheap, so if an application
begins repeatedly requesting a data connection, not much is lost. This is not the case in
the wireless world. Imagine a popular client being used by tens of thousands of wireless
users spread out across a country. What happens when the base stations suddenly
cannot support the required service option? Or when the AAA server becomes too busy
to service subscribers, or when the application server itself goes down? Network
operators need to know that a simple server glitch will not cause tens of thousands of
devices to simultaneously begin repeatedly inundating the network with connection
requests; bringing the network to its knees.
Another example is the concept of “throttling”. Operators sometimes find that a new
device or service catches on with the public, instigating an immediate increase in data
usage. Because the operator needs to distribute service across all its subscribers, the
operator needs a “valve on the data pipe”. This is called throttling, and its functionality
must be tested on mobile devices. In this case it was the CDMA Certification Forum™
(CCF) who stepped up and reacted to the need, after some operators began to develop
their own applicable test plans.
Testing CDMA and EV-DO Mobile Terminal Devices
White Paper 122
18 Spirent White Paper
Table 4: Data Services Testing
3GPP2
Numbering
TIA Numbering Document
C.S0037 TIA-918 Signaling Conformance Specification for Wireless IP
Networks Version 1.0
1x, EV-DO Rev 0 and
EV-DO Rev A
01-06-09-TG-Data Session
Throttling Test Case Specification-
1.0.0 (Contribution)
CDMA Certification Forum™ (CCF) Data Session
Throttling Test Case Specification
Testing CDMA and EV-DO Mobile Terminal Devices
Performance Testing
Spirent White Paper 19
PERFORMANCE TESTING WITH CONFORMANCE TEST CASES
While the industry relies on specific pass/fail limits to ensure the viability of a
technology, existing test cases can be used to gather data that is useful throughout the
lifecycle of a product. For example, by adding a few key functions to an automated
system’s test executive software, an automated “pass/fail” testing system can be turned
into a diagnostic tool, a debugging station, a place to gather data for technical strategy
planning, and more. Network operators can run comparative analyses to determine the
devices that will drive positive user experiences, as well as those that will unduly burden
network resources; two handsets can present the same user experience, while one eats
up a disproportionate share of transmit power and/or network data bandwidth. A smart
operator wants this information before tens of thousands of units start appearing on the
network. Along similar lines, a handset designer can use the same set of tests to prove
out incremental improvements based on firmware changes.
For example, “swept parameter” analysis is a very effective way to pinpoint surprising
differences between commercially available (or approved) phones. For example, Figure 6
gives an example of a swept-parameter test suite, ready to run on the automated
systems. Figure 7 shows a typical set of resulting responses for three similar handset
devices when subjected to an increasingly significant impairment.
All three of the devices described in Figure 7 may meet the standard requirements, but
one of them (in green) demands a greater share of network resources as the network
continually re-transmits information. Owners of this device eventually become
unsatisfied, affecting revenues and support costs for both the manufacturer and the
network operator.
Figure 6: "Swept-parameter" Test Suite on an Automated Testing System
Testing CDMA and EV-DO Mobile Terminal Devices
White Paper 122
20 Spirent White Paper
Figure 7: Throughput/Impairment Curves for Three Similar Devices
Another example would be the predictive analysis of what will happen to the LBS in the
near future. As stated above, the orbits of GPS satellites are not absolutely predictable,
but near-future time frames can be calculated with reasonable accuracy.
Suppose an operator notices it is three weeks away from seeing a sub-optimal satellite
configuration in a heavy-usage urban area.
Referring once again to Figure 3 and its accompanying text, GPS-based location
determination is based on distances thought of as spherical surfaces. The inherent
errors can be thought of as “thicknesses” of these surfaces. Ideally, GPS satellites are
spread out across the satellite field.
But what happens when the satellites eventually align in one way or another? The term
Geometric Dilution of Precision (GDOP) is a measure of how optimally satellites are
spread in the constellation. To visualize this, imagine the thick-shelled spheres
generated from the satellite field shown in the right-hand side of Figure 8. Even when the
“spheres” are known for all five of the satellites, their intersections are relatively large
volumes. This can be interpreted to mean that GPS-based measurements will be fairly
inaccurate at a specific time and place.
Testing CDMA and EV-DO Mobile Terminal Devices
Performance Testing
Spirent White Paper 21
Figure 8: Satellite Fields with Good (left) and Bad (right) GDOPs
To quantify this effect, engineers can use TIA-916-based test cases with simulated
satellite fields set to mimic an upcoming satellite configuration. Ideally, this is as simple
as entering time and place information into software running on the automated system,
as shown in Figure 9.
Figure 9 - Controlling GPS Satellite Fields with Time/Place Information
Testing CDMA and EV-DO Mobile Terminal Devices
White Paper 122
22 Spirent White Paper
SUMMARY AND CONCLUSION
A standards body such as 3GPP2 takes on the monumental responsibility of ensuring
that a technology works according to a set of documented requirements. Yet the charter
of a standards body seems limited when compared to the charter of an operator or cell-
phone maker, both of whom must maximize profits.
This paper has presented an extremely high-level overview of the basic elements of
testing a CDMA/EV-DO mobile device. Any single topic presented here can easily be the
subject of years of detailed study. Fortunately for device makers and operators alike, this
study has been completed by manufacturers of professional automated testing systems.
The paper discussed conformance and performance testing in several key areas:
• Minimum Performance (transmitter and receiver) Testing: Including a discussion of
RX Diversity testing.
• Signaling Conformance Testing: Including a discussion of how the industry standard
specifications have evolved in this area.
• Data service Testing: Including discussions of throughput testing and “safe-for-
network” Data Retry testing.
• LBS Testing: Including both government-mandated emergency service systems and
commercial location-based services.
Thoroughly testing a mobile device is a daunting task. For CDMA and EV-DO devices,
thousands of pages are devoted to the test specifications alone. Each test case is
traceable to a specific requirement called out in the technical specs. Very few engineers
will ever read all of the documentation that applies to device testing.
It is tempting to assume that device testing is complete after the required conformance
tests are finished. However, the needs of a wireless network operator or device
manufacturer are very rarely in line with the charter of the typical standards body. For
this reason, the industry has adopted a de facto standard automated testing system that
is flexible enough to support a wide range of controllable, repeatable performance
testing.
Testing CDMA and EV-DO Mobile Terminal Devices
Summary and Conclusion
Spirent White Paper 23
ACRONYMS
3GPP2 3
rd
Generation Partnership Project (2)
AAA Authentication, Authorization, and Accounting.
A-GPS Assisted GPS
CDMA Code Domain Multiple Access
EV-DO EVolution, Data Only
GDOP Geometric Dilution Of Precision
GPS Global Positioning System
IP Internet Protocol
IS Industry Standard
LBS Location-Based Services
MEID Mobile Equipment Identification
OTA Over The Air
PDE Position Determination Entity
PPP Point-to-Point Protocol
RAN Radio Access Network
SMS Short Message Service
TIA Telecommunications Industry Association
