Professional Documents
Culture Documents
20 ANTENNA GPS
11 SURVEY WORLD
CLICK HERE TO
DOWNLOAD THE 2011
ANTENNA SURVEY
www.gpsworld.com/antennasurvey
March 2011
Vol. 22, Number 3
gpsworld.com
» cover storY
transportation
Road Talk 36
Interoperability Considerations for V2X GNSS Receivers
The presence of different types of devices, spanning multiple GNSS
receiver types, configurations, hardware, software, and consequent
widely varying capabilites, among a user mix of vehicles, cyclists, and
pedestrians, poses several engineering challenges for a V2X scheme
in which all road users share data with each other and with the road
infrastructure.
By Chaminda Basnayake
on the edge
the business 17 Application Challenge
ST-Ericsson, Broadcom GPS +
GLONASS Receivers; Delorme Iraq on the Map 49
Downloadable World Base Map; By Anas Malkawi
Navteq LocationPoint Ad Network;
Trimble Railway Track Maintenance
and Modernization; NovAtel’s
SMART-MR15 for Agriculture;
Department of Transportation
Vehicle-Connectivity Challenge;
Events; Online Webinars
G
PS has become a key This change amounts to a de facto This guest editorial addresses a
component of the U.S. reallocation of Lightsquared’s spectrum subject of paramount importance
national infrastructure, the use from space to terrestrial wireless. to the GNSS industry, to the U.S.
driver of a significant part of the In fact, the new broadband service is national infrastructure, and to the
civilian economies of the world, and planned to operate in urban areas, and global GNSS community.
the enabler of millions of professional the space service will operate outside I urge you to take immediate
precision uses and consumer benefits. these areas. action by contacting U.S.
The viability of the GPS signal is now The LightSquared terrestrial government representatives,
threatened — ironically by what appears broadband signal is about 1 billion times indicated at the end of this article.
to be a misguided attempt to increase the received power of the GPS signal — Alan Cameron, editor-in-chief
accessibility to broadband by creating a on Earth. Members of the GPS industry
needless zero-sum result for customers have been conducting experiments
who want both services. and analyses, and these figures come Department of Defense’s statements that
The threat is real and immediate. from those very early studies. Soon, we the service would be free and not subject
The U.S. Federal Communications may experience GPS interference — to any restrictions in one’s ability to
Commission (FCC) has issued a jamming — on an almost unimaginable receive and use the broadcast signals.
conditional waiver to LightSquared, a scale and to a geographical extent that This uncertainty was due primarily
company engaged in developing could create widespread havoc. to the implementation of Selective
Availability (SA), which intentionally
For many GPS users, theoretical fixes to the interference degraded the available accuracy of
the GPS signal. SA was permanently
are not likely to prove viable. removed in 2000 by President Clinton’s
1996 Presidential Decision Directive.
Many factors have enabled users and
4G-LTE (long-term evolution) cellular Threats. The GPS system works so potential users to see GPS as a reliable,
networks for wholesale-only basis well that we often forget the complexity consistent technology that provides
commerce with its business partners. behind it and take for granted the service significant increases in productivity,
LightSquared Scheme. LightSquared we use daily. One reason GPS works efficiency, precision, continuing
acquired a company providing a so well and is seldom defeated is that innovation, and many other benefits.
combined space-based and ancillary the signals broadcast by the satellites These factors include the reliability
land-based service using the L-band can be received under a wide variety of the overall GPS technology,
radiofrequency. The FCC conditional of conditions on Earth. Historically, improvements in receivers and in
waiver, granted to LightSquared the FCC and the International successive next-generation satellites,
on January 26 of this year, allows Telecommunications Union, advances in differential and relative
it to broadcast a new terrestrial understanding potential interference positioning, dynamic applications, and
broadband service from 1,500-watt issues, intentionally planned uses of real-time kinematic solutions. And, just
terrestrial transmitters — 40,000 of adjacent swaths of the L-band so that as importantly: stable, predictable U.S.
which will eventually be installed satellite-based transmissions, relatively policy.
by LightSquared — in the portion low-power, would be natural neighbors, Investments. Now, by virtue of this
of L Band (1525 MHz–1559 MHz) so as to cause as little disturbance as unusual FCC action, uncertainty has
immediately adjacent to the 1559–1610 possible to radio-navigation uses. This been thrown into the viability of the
MHz band used by GPS. dedicated purposing of the bands and hundreds of millions of GPS receivers
Instead of offering dual-mode the resulting environment of negligible in use today. Much research and
handsets exclusively as required by their interference is one reason that GPS has development work is being done on
FCC license, retailers purchasing this become reliable and its use ubiquitous. improving receiver performance and
combined service can choose to offer Long-time observers of the GPS taking advantage of improvements
terrestrial mobile phones only, which scene will remember how civilians, and planned for the satellites. The most
was the change in license terms that especially potential international users, dollars go towards devising new
LightSquared was seeking via wavier. initially had uncertainty about the U.S. applications, products, and services
4 GPS World | March 2011 www.gpsworld.com
out in front
that improve the quality life of millions making process for reallocation of 1 billion times stronger than GPS. There
of Americans, create new companies, spectrum use. The standard regulatory is more to the research, all done with
markets, and jobs. These dollars are also approach allows sufficient time for GPS simulators and signal generators
being spent by government agencies, robust public comment by all potentially (see news story on page 14 and www.
not just the Department of Defense, but affected parties, including the conduct of gpsworld.com/data for test results).
very visibly by Agriculture, Commerce, interference studies and the introduction Clearly the jamming level will vary
Interior, Energy, Homeland Security of comments on interference results in with geography. We don’t yet know
and Transportation. More than likely, the public record. Instead, the FCC order LightSquared’s broadcast-tower siting
the remaining departments either granting the waver to LightSquared has plan. But it is clear that if LightSquared
have active programs that are using or mandated what appears to be fast-track is allowed to broadcast terrestrially
considering using GPS or are positively GPS interference research.
affected by others’ use of GPS. Currently, the proposed LightSquared
That’s just the executive branch. terrestrial broadband service does not High-Precision Users
Other parts of the federal government, have an installed user base. In contrast, High-performance L1 receivers (sub-
as well as state and local governments, the installed GPS user base represents meter) have a wide-bandwidth RF
do research on GPS technology and a broad and diverse range of use front-end to improve performance,
applications and actively use GPS to representing hundreds of millions of about 20 MHz, compared to a
improve the lives of citizens, increasing users established over 30 years. consumer receiver that typically has a
work and recreation, efficiency, and The final Working Group report is front-end bandwidth of 2 MHz.
safety. In many local government due to the FCC on June 15, 2011. The GPS World contributing editor
settings, there is active cooperation to FCC order requires the GPS community for survey and GIS Eric Gakstatter
improve delivery of services by having to participate “in good faith” in this discusses this aspect of the issue in his
governmental and non-governmental study effort. In response, the U.S. GPS recent e-mail newsletter column at
organizations collaborate around the Industry Council and others are working www.gpsworld.com/l2high.
simple fact of accurate position being on this interference study to protect GPS
available through GPS, with significant operations under these extraordinary
cost savings in current lean budgets. regulatory conditions. on the mobile satellite system (MSS)
It is inexplicable that another part of A further problem created by band, dedicated until now to signals
the government would be so cavalier in the FCC conditional waiver is that compatible with satellite transmissions,
rapidly and uncharacteristically granting LightSquared is able to move ahead there is a substantial danger that millions
a waiver that clearly endangers the with its infrastructure development, of GPS receivers will be adversely
whole system. And only after granting assuming that viable solutions to the affected.
the waiver, which must act at least jamming issue will be found. For many Some obvious impacts are loss of
as a yellow light for LightSquared’s GPS users, theoretical fixes are not operational viability of businesses
mobilization plans, comes the likely to prove viable. involved in aviation, surveying,
agriculture, engineering and
Obvious impacts are loss of operational viability of businesses construction, vehicle navigation,
mariners, transportation, public safety
involved in aviation, surveying, agriculture, engineering and and homeland security, disaster
management, utilities, mapping, and
construction, vehicle navigation, mariners, transportation, public scientific research. Several of these
safety and homeland security, disaster management, utilities, involve safety-of -life issues, which are
at risk of being jammed.
mapping, and scientific research. Keep in mind that GPS was
envisioned as a system for space
and time. Its longest life as a useful
requirement for a study — to be headed Negative Impacts. Preliminary research contributor to society has been as a time
by LightSquared — to determine done by member companies of the standard. Countless networks, whether
impacts and mitigation of interference USGIC already has been reported in for computing, broadcasting, power
with the GPS signals. GPS World. The research indicates that generation — even, ironically, cell
Why Fast Track? The FCC grant of a LightSquared’s 1,500-watt terrestrial phones — are synchronized using the
reallocation of spectrum use from space transmitters will result in a signal 90 dB most precise signal practically available.
to terrestrial on a fast-track waiver stronger than GPS over the coverage Fixed GPS receiving stations for time
did not follow the standard FCC rule- area; this amounts to signal strength reference may be able to be designed
www.gpsworld.com March 2011 | GPS World 5
out in front
to withstand some interference from the recent FCC Order, NTIA must
high-power broadcasting on adjacent review the report on results of the
frequencies, but nobody has tried so far. FCC-mandated interference study.
Any hypothetical fixes for GPS beg ◾ Specifically, ask Congress to de- www.gpsworld.com
a more fundamental question: Why mand that the FCC include specific editorial
should Lightsquared, a new entrant with language to protect GPS use in the Editor in Chief Alan Cameron | acameron@questex.com
Managing Editor Tracy Cozzens | tcozzens@questex.com
no existing business, be allowed to shift final FCC Order to LightSquared Art Director RJ Pooch | rpooch@questex.com
the burden of mitigating interference after the interference study is com- EDITORIAL OFFICES
created by its operations to millions of pleted. 201 Sandpointe Avenue, Suite 500,
Santa Ana, CA 92707-8716 USA
consumers, government agencies, and ◾ Ask the Secretary of Commerce 714-338-6700 | Fax 714-338-6717
businesses who have invested in GPS and the White House Office of Sci- www.gpsworld.com | info@gpsworld.com
over the last 30 years? ence and Technology Policy (OSTP) COntRibuting editors
Keep in mind that other users of the to inform the NTIA Administrator Innovation Richard Langley | lang@unb.ca
MSS band will also be affected. Many to urge the FCC chairman to take Defense PNT Don Jewell | djewell@questex.com
LBS Insider Kevin Dennehy | kdennehy@questex.com
commercial and governmental uses of this same action to protect GPS in Professional OEM Tony Murfin | tmurfin@questex.com
the very band that LightSquared will the final FCC Order. Survey Eric Gakstatter | egakstatter@questex.com
Transportation Bill Thompson | bthompson@questex.com
occupy with its terrestrial transmitters ◾ Contact the FCC chairman directly Wireless Pulse Janice Partyka | jpartyka@questex.com
may also be jeopardized. and urge this same action.
advertising
We must also remember that the ◾ Finally, help develop user and
Publisher George Casey
FCC has its own agenda, to implement beneficiary awareness of the grave george.casey@questex.com | 216-706-3752
its National Broadband Plan. What is danger being posed to GPS and Eastern Region/International Account Manager
truly difficult to comprehend is that make your elected and congres- Michael Eager | meager@questex.com | 913-254-9622
broadband and GPS will serve the same sional representatives aware of the Western/Northern Regional Account Manager
Jeannie Hortness | jhortness@questex.com | 714-338-6732
impact that interference with GPS Marketing Manager Sarah Joy Obaña
Large-scale disruption of the would have on your work. sobana@questex.com | 714-338-6763
The large-scale disruption of the GPS Vice President, Industrial & Specialty Group
GPS service must not be on our service mustn’t be on our hands due to Kevin Stoltman | kstoltman@questex.com | 216-706-3740
GATE Transmit Station 1 GATE Virtual Satellite Mode GATE Test Vehicle
Fully compliant to Operational Galileo test range with eight “virtual” Galileo satellites
Galileo OS ICD
Transmitted signals are fully compliant to Galileo OS SIS ICD
A
s cell phones move into the to go beyond the native capabilities networks as are present in GSM,
next generation called Long- of GNSSs to achieve tangible UTRA, and CDMA networks. This
Term Evolution (LTE), also accuracy, time-to-first-fix (TTFF), meant that there was a need to define
sometimes called 4G, and the methods and availability improvements. assisted-GNSS positioning as well
of wireless transmission change, so Contributors proposed introducing as introduce positioning methods,
too must the methods of providing local ionosphere and troposphere such as enhanced cell ID (ECID) and
location information over those new models as well as carrier-phase-based hyperbolic time-difference-of-arrival
wireless interfaces. LTE Positioning relative positioning — in cell phones! (TDOA) methods for non-GNSS
Protocol (LPP) and Secure User Plane However, legacy implementations, devices, hybrid use, and for GNSS-
Location (SUPL) 2.0 and 3.0 are the architecture limitations, and the denied environments. The underlying
key players in this new picture. lack of a business case hindered driver of all this work was the U.S.
Cellular industry location standards this development. In the end, RRLP Federal Communications Commision
first appeared in the late 1990s, with support was limited more or less to (FCC) Wireless E911 mandate.
the 3rd Generation Partnership (3GPP) the native assistance data types such
Radio Resource Location Services as global Klobuchar and NeQuick LTE Location Architecture
Protocol (RRLP) Technical Specifica- models for the ionosphere. The same LTE location architecture is shown in
tion (TS) 44.031 positioning protocol approach was also mapped to 3GPP Figure 1. The evolved serving mobile
TS 25.331 Radio Resource Control location center (E-SMLC) is the server
(RRC) protocol, which defines the component in charge of positioning ac-
positioning procedures and assistance tivities. The mobility management entity
data delivery for Universal Mobile (MME) gives the positioning request to
Telecommunications System E-SMLC, which then controls the user
Terrestrial Radio Access (UTRA) — equipment (UE, the LTE device to be
that is, wideband code-division positioned) and, possibly, LTE base sta-
multiple-access (WCDMA) and tions (eNodeBs), to perform positioning.
time-division synchronous CDMA LTE Positioning Protocol. The actual
networks. positioning and assistance protocol
LTE Positioning Protocol between E-SMLC and UE is called
Long-Term Evolution Networks LTE Positioning Protocol (LPP).
extensions in the A-GNSS A fresh push for location services in In overview, LPP consists of three
3GPP started in 2009 for LTE Release 9 independent elementary procedures:
domain include local technologies. LTE is sometimes called capability exchange, assistance data
atmospheric models, for more 4G, but to be precise only a further evo- exchange, and location information
lution of LTE, called LTE Advanced exchange, which refers to both
accurate delay compensation. (LTE-A), will be 4G, together with measurement and position. The
WiMAX evolution 802.16m. associated messaging is shown in
for GSM networks. Today RRLP is the The starting point for LTE location Figure 2. In addition to the six message
de facto standardized protocol to carry, services work was to enable similar types shown, there are LPP Error
for instance, GNSS assistance data to positioning capabilities in the LTE and LPP Abort messages to handle
GNSS-enabled mobile devices.
A major update of RRLP began in
2007, when RRLP Release 7 added
support for assisted-Galileo, and
Release 8 for the rest of the GNSS
including GLONASS, modernized
GPS, QZSS, and the various SBASs.
RRLP Releases 7/8 set high
expectations in terms of performance
improvements. The initial idea was ▲ FIGURE 1 Long-Term Evolution (LTE) location architecture.
▲ FIGURE 5 OMA SUPL 2.0 and 3.0 protocol stacks. TIA-801 is the
3GPP2-defined positioning protocol for the CDMA networks. Note
that ULP 1.0 (not shown) supports RRLP, RRC, and TIA-801.
▲ Doppler, SNR, and “code-minus-phase” graphs for satellite 211, collected ▲ THIS Log file reported up to 26 satellites locked simultaneously (col-
during its pass. For additional satellite graphs, visit gpsworld.com/javad. lected on the TR_G3TH board in Moscow).
» consumer oem
» location-based services
Navteq LocationPoint Ad Network Expands with New Publishers Worldwide
Global map provider Navteq has an- location content provider for handset ity to actively engage consumers near
nounced it has dramatically increased and auto makers — as a leading ag- points of purchase and route them to
the reach of its LocationPoint Advertis- gregator of ad inventory that spans advertised merchants, and metrics that
ing Network by forming partnerships devices, apps, and global regions. measure return on investment, the
with RIM and Samsung, in addition to Navteq LocationPoint is a location- company said. Navteq has conducted
an ongoing collaboration with Nokia. based mobile advertising network that trial and commercial campaigns on its
Also joining the Navteq LocationPoint delivers ads to smartphone users when LocationPoint Ad Network through
ad network are location-based applica- they’re in the proximity to go, buy, and 2010. A European effort for McDonald’s
tion providers Appello, CoPilot Live, shop at advertised merchants. The net- garnered 7 percent consumer click-
Navigon, NDrive, Poynt, and Telmap. work expansion paves the way for the through rates and won an EMMA for
According to the company, the ex- network to provide brand advertisers Europe’s most-effective location-based
pansion positions Navteq — map and with large-scale audience reach, abil- mobile ad campaign.
» transportation/rail
Trimble Introduces Railway Track Maintenance and Modernization Solution
Trimble has introduced a new addition and modernization. The system and data needed for railway lifecycle man-
to its Railway Solutions portfolio — software were acquired from Sinning agement — from the planning phase,
the Trimble GEDO CE Trolley System. Vermessungsbedarf GmbH of Wiesen- through design and construction, and
The new trolley system and software theid, Bavaria, Germany. now, with the addition of the Trimble
provide as-built survey and documen- Trimble says it offers users products GEDO CE Trolley System, the subse-
tation for railway track maintenance that quickly and accurately capture quent maintenance phase.
18 GPS World | March 2011 www.gpsworld.com
the business
» machiine control/agriculture
» transportation
Department of Transportation Hosts Vehicle-Connectivity Challenge
The U.S. Department of Transportation’s communications, which will be the munications can potentially address
Research and Innovative Technology basis for a future system of connected 81 percent of all unimpaired vehicle
Administration (RITA) has announced vehicles that will communicate with crashes. Selected prize recipients will
the Connected Vehicle Technology each other as well as the surrounding be fully funded to present their ideas
Challenge. The new competition seeks infrastructure, such as traffic signals, for connected vehicle technology.
industry ideas for using wireless con- work zones, and toll booths. The competition, which runs
nectivity between vehicles. According to a National Highway through May 1, is open to all compa-
RITA, through the competition, is Traffic Safety Administration report, nies, not just those involved in trans-
soliciting ideas for products or applica- wireless vehicle-to-vehicle (V2V) and portation. More information can be
tions that use dedicated short-range vehicle-to-infrastructure (V2I) com- found at www.challenge.gov.
A
s roads become more congest- a GNSS receiver in an onboard unit usage, creating a financial incentive for
ed, governments and regional (OBU) provides a convenient and flex- changing driving behavior. This might
authorities seek better ways ible means to support automated fee include lower overall use of private
to manage their existing networks. collection. For GNSS positioning to cars and, in particular, reducing peak-
Road-user charging (RUC) is increas- be used as the basis for billing driv- time travel levels in urban areas by
ingly promoted to tackle the conges- ers, however, it must meet stringent effectively spreading out the morning
tion challenge: charging drivers a fee, reliability and availability require- and evening rush hour. RUC can also
perhaps on a monthly billing basis, ments, and at the same time be based encourage commuters to use alterna-
derived from the distance their vehi- on low-cost equipment. tive forms of public transport.
cles have traveled, time of travel, and We have developed a prototype to To automate the process of collect-
whether congested roads were used. provide both the positioning avail- ing charges, electronic fee-collection
Recording trip information with ability and integrity required for this (EFC) systems have been developed
28 GPS World | March 2011 www.gpsworld.com
Road | Transportation
based largely on dedicated short-range satellite signals blocked and reflected, based solution for flexible road-user
communications (DSRC). In a DSRC are well documented. In some cases, charging that can provide the required
solution, a simple tag on the vehicle not enough signals are available to high integrity and high availability in a
receives a signal when it passes a road- determine a position, and when there cost-effective and scalable manner.
side beacon and a charge is computed are enough satellites, the ranges will This robust, high-availability, high-
accordingly. Cameras with automatic be prone to errors and the geometry integrity solution is delivered firstly
number-plate recognition (ANPR) is likely to be poor. Signals are more through providing reliable assistance
technology are also widely used, main- likely to be available in the longitu- (A-GNSS) data from EDAS to opti-
ly as an enforcement tool. dinal direction of the street, but with mize receiver acquisition and track-
Both technologies rely on fixed few or no satellites on either side of the ing capabilities and reduce TTFF, and
roadside infrastructure. As charging vehicle. Signal blockage is a particular secondly through implementation of
schemes to date have focused on spe- problem when the GNSS receiver is embedded GNSS reliability algorithms
cific areas (individual cities) or road started up, and it attempts to decode into an OBU, providing assurance of
infrastructure (major motorways, tun- satellite ephemeris data. This requires positioning information (Figure 1).
nels, and bridges) this type of technol- around 30 seconds of uninterrupted These features are intended to make
ogy provides an adequate solution. tracking with a relatively strong signal a positive contribution in terms of the
To meet future policy goals, howev- for each satellite, and in an obstructed key RUC performance criteria, as de-
er, this is not feasible. More extensive urban environment it may take many fined by the GNSS Metering Associa-
charging schemes covering greater ar- minutes to determine the first receiver tion for Road User Charging:
eas, more road types, more classes of position. ◾ Accuracy: right cost per trip
vehicle, and which will vary charges Charging schemes typically aim to ◾ Integrity: probability and amount of
depending on location and time of day charge for at least 99 percent of road overcharging
require a far more flexible solution. usage. If a typical journey length is 30 ◾ Availability: amount of charged us-
Flexible schemes require the position- minutes, this means that only 18 sec- age.
ing element to be onboard the vehicle. onds with no usable position solution Assistance Server. An assistance ser-
GNSS-based devices, possibly aug- can be tolerated and hence time to first vice supplying suitably equipped
mented with other sensors, have been fix (TTFF) must be below 18 seconds, OBUs is one means to maintain rapid
identified as the best option to achieve and ideally much lower. TTFF and meet the requirement for
this. When positions can be determined, high positioning availability. The most
Using GNSS, the OBU tracks the they must be sufficiently accurate to significant contribution assistance
location of the vehicle, and this is identify the correct road segment that can make to TTFF is to provide the
matched against the road network to the vehicle was on, and they must be ephemeris data, which takes around 30
calculate a charge. A GNSS solution reliable. Reliability, or integrity, be- seconds to download from a satellite
can support many different charging comes critical if road users are to be signal. Assistance data can also reduce
strategies including time distance and charged on the basis of GNSS-derived the frequency search space when a re-
place (TDP) based charging for road positions. Users must have confidence ceiver is acquiring signals, as the ex-
sections, geographic areas, and cordon that they are being charged correctly pected Doppler frequency can be com-
schemes. While GNSS offers great for schemes to be effective and to gain puted from the approximate receiver
potential, several operational and per- public acceptance. and satellite positions.
formance limitations have prevented Whilst GNSS-based solutions are The assistance server in SIGNA-
more widespread adoption. Operation- entering the market, for example in TURE is based on EDAS, currently
ally, OBUs are relatively expensive, Germany and Slovakia for heavy goods available as a beta version. EDAS al-
fraud prevention is potentially com- vehicles, barriers to wider adoption re- lows a user to plug into EGNOS to
plex, and charging schemes must also main. Many countries are considering receive the data collected by all the
accommodate infrequent users. GNSS GNSS-based road pricing, and they all current EGNOS Ranging and Integ-
performance is limited in terms of the face similar challenges in ensuring the rity Monitoring Stations (RIMS). This
ability to provide sufficiently accurate accuracy, integrity, and availability of makes it possible to access EGNOS
positions with high availability and in- a GNSS-based solution for nationwide data when there is no clear sight to
tegrity in all operating conditions. deployment. the EGNOS geostationary satellites,
To be fully flexible and to target which can often be the case in urban
congested areas, OBUs must work SIGNATURE Solution areas, particularly at higher latitudes.
in all environments including urban The principal objective of the SIGNA- As well as supplying EGNOS mes-
areas. Urban-canyon problems, with TURE project is to prototype a GNSS- sages, EDAS also provides GPS obser-
www.gpsworld.com March 2011 | GPS World 29
Transporation | Road
Mean TTFF(s)
Open Sky Dense Urban
SIGNATURE 1 3
COTS - Warm 28 136
COTS - Hot 1 2
COTS - Assisted 8 9
▲▲ Table 1 Time-to-first-fix results. ▲▲ FIGURE 7 Position error and protection level, Nottingham city center.
ments.
Further tests were also conducted in the center of London,
in a worst-case obstruction environment. In this area the
current solution falls just short of the requirements defined
for this project. In such cases, better performance could be
obtained using a hybrid solution making use of additional
▲▲ FIGURE 6 RAIM Impact (red = no RAIM, yellow = RAIM). sensor inputs, but this will increase equipment costs and po-
tentially installation costs, too. A more practical approach
ing a signal generator to control the strength of the received may be to simplify charging schemes in the densest urban
signal to assess acquisition and tracking sensitivity. At –145 environments, using zones and cordons rather than using
dBm, the SIGNATURE receiver takes an average of 1.1 sec- more detailed approaches that require a continuous high-
onds to acquire 4 satellites and determine a first fix, and 5.1 performance positioning solution to be maintained in all
seconds to acquire 12 satellites. conditions.
Positioning Algorithms. A variety of configurations have Benefits of EGNOS Data. The SIGNATURE solution has the
been investigated in the positioning algorithms, including ability to provide EGNOS data to positioning algorithms on
applying outlier-detection routines at different stages of the the OBU and to vary the rate at which this information is up-
solution and comparing snapshot and filtered approaches. dated and used. Field tests have assessed the potential ben-
Figure 6 shows a simple example of how the RAIM algorithm efits of this source of data in various environments, starting
has been effective in detecting and excluding outlying mea- from the case in which EGNOS messages are continuously
surements contaminated by multipath. By removing these available for the positioning solution and then investigating
meaurements and re-computing the OBU location, better how any beneficial effects lessen as the data is provided less
position estimates are obtained. frequently. The greatest benefit from EGNOS was derived
Figure 7 shows the accuracy and integrity of the SIGNA- by applying corrections prior to performing the RAIM FDE
TURE solution assessed using a high-grade GNSS/INS ref- algorithm. This led to more consistent measurements and
erence in Nottingham city center. In this case, the horizontal produced lower HPL values. Figure 8 shows a comparison for
accuracy is 4.4 meters (95 percent), and the computed pro- a Nottingham test in which a GPS-only solution is compared
tection level is shown to bound the actual position error with against an EGNOS solution in which a full set of corrections
the required confidence. is provided.
In rural, semi-urban, and urban (Nottingham) conditions, This reduction in HPL values through the application of
a positioning solution has been demonstrated that supports EGNOS corrections is clearer when the distribution of HPL
all charging accuracy, integrity, and availability require- values falling into discrete bins is assessed (Figures 9 and
34 GPS World | March 2011 www.gpsworld.com
Road | Transportation
Manufacturers
The SIGNATURE receiver is based on the Terasic Altera DE3
System with a high-density Stratix III FPGA (EP3S260),
and on the Rakon GRM8652 high-performance front end.
Acknowledgments 7(:$PHULFD
The SIGNATURE project has received funding from the =DQNHU5RDG
European Community’s Framework Programme (FP7/2007- 6DQ-RVH&$
2013) under Grant Agreement No. 228237 and is supervised
)$;
the European GNSS Supervisory Authority (GSA). Full de-
tails of the project can be found at www.gnsssignature.org. Any ZZZWHZDPHULFDFRP
views expressed here are entirely those of the authors and do
www.gpsworld.com March 2011 | GPS World 35
V2X system users.
▲▲ FIGURE 1
Road Talk
Interoperability Considerations for V2X GNSS Receivers
The presence of different types of devices, spanning multiple GNSS receiver types, configurations, hardware, software, and conse-
quent widely varying capabilites, among a user mix of vehicles, cyclists, and pedestrians, poses several engineering challenges for
a V2X scheme in which all road users share data with each other and with the road infrastructure.
Chaminda Basnayake, General Motors Global Research and Development
T
he use of location awareness for transportation the GNSS technology further develops, some vehicles may
safety, efficiency, and security — a major area of have multiple-frequency GNSS capability compared to
research and development for academics, automotive legacy single-frequency devices. In essence, all compatible
manufacturers, and organizations such as the U.S. V2X devices will have to be carefully designed to ensure
Department of Transportation — has focused attention on their interoperability with the rest of the system.
enabling communication between vehicles and other road This article investigates positioning challenges arising
user entities in a concept know as V2X, a term encompassing from multiple GNSS receiver types, configurations,
both vehicle-to-vehicle (V2V) and vehicle-to-infrastructure hardware, and software in a V2X operational environment.
(V2I) systems, so that they can share location and other This produces a clear need to have minimum performance
status information. As a result, any road user entity may see standards for V2X-capable GNSS receivers. The article
all others around it. This capability is almost always built on further investigates the implications of land-based visibility
GNSS technology. obstructions on relative positioning, and implications on
Future V2X systems will be able to include all road user standalone position accuracy both as a result of limited
entities, ranging from vehicles to cyclists to pedestrians, in GNSS satellite visibility and WAAS satellite visibility.
this information-sharing system. While it sounds natural for
everyone to talk to each other and share data for collective V2X Background
benefit, the presence of different types of devices among V2X systems rely on two critical enabling technologies:
this user mix poses several engineering challenges. As an communications and positioning. Organizations and
example, a V2X device in a vehicle may have a built-in industry collaborations have developed and demonstrated
GNSS receiver with a roof-mounted antenna and another various V2X systems over the last few years. These efforts
vehicle may have a retrofitted V2X device with a passive have produced interoperable prototype V2V and V2I
antenna and relatively limited accuracy capabilities. As systems and over-the-air (OTA) messaging standards.
36 GPS World | March 2011 www.gpsworld.com
Road | Transportation
is a high-quality L1/L2 receiver using signal was split into multiple outputs satellites used by each receiver in the
a geodetic-grade antenna, used with such that multiple receivers can be same environments as in Figure 6.
WAAS enabled. The GPS/INS system tested at the same time. For each Overall, Type A receiver tracks and
was connected to the geodetic-grade replay of the RF data, a benchmark uses on average 2–3 satellites more
antenna. The RF recording system receiver was also used to verify that compared to the Type B receiver, likely
was also connected to the automotive- there is no run-to-run difference as a due to its high-sensitivity capability.
grade GPS L1 antenna. result of the RF replay. Type A C1 and C2 receivers also track
The data was collected on a test Outputs from each GPS receiver and use 2–3 satellites fewer compared
route in Detroit, Michigan that from field and replay runs were logged to the all-in-view Type A receiver.
included durations of urban and deep to PCs using receiver specific binary Freeway Data. The vehicle heading in
urban canyon (40 miles per hour (mph) formats. The recorded output from this segment was predominantly north
or less), freeway (55–70 mph), and each receiver included its position, or northwest. The sky view can be
suburban/local (30 mph) driving. position error estimate, velocity, considered a combination of urban and
The RF data were subsequently satellite-specific measurements and open sky conditions. As highlighted in
replayed to GNSS receivers that were indicators such as pseudorange, carrier Figure 7, all-in-view Type A receiver
not a part of the field set-up. RF data phase, and signals-to-noise ratio. was able to use up to 11 GPS satellites
was also replayed to receivers with with an average of around 9 satellites.
forced sky-visibility obstructions Data Processing and Analysis Type A C1 and C2 receivers used, on
and various WAAS settings. For The data was first decoded from the average, about 3 satellites fewer than
limited sky-visibility tests, certain receiver-specific formats to a common the all-in-view receiver. All three
satellites were removed from each format, then corrected for antenna receivers show satellite count drops
receiver’s view by receiver-specific offsets. To simplify the process, the down to 4 at certain times in this
configuration software. The satellite reference system position solution was segment.
selection and restriction was done to translated to the position of the test The satellite count of the Type
mimic typical sky-view obstructions antenna using the known between- B receiver shows the limitations of
encountered in normal driving. antenna distance and orientation of the not using the high-sensitive tracking
Type A receiver was chosen to vehicle as measured by the reference capability. The satellite count shows
illustrate the impact of visibility system. As a result, all the receivers frequent drops below 4 satellites and
differences. A total of 13 satellites and the reference system are reporting
were visible in the entire data set (Figure the location of the test antenna.
4). To create obstructed sky-view Then, data fields such as position and
scenarios, two Type A receivers were velocity for each receiver were time-
configured to not use certain satellites matched with the reference solutions,
in their position solutions. These and the actual error was calculated.
configurations were: For a limited dataset, additional
◾ Configuration 1 (C1): PRNs 7, 10, measurement-level differential
and 13 blocked processing was done to show the
◾ Configuration 2 (C2): PRNs 6, 16, difference between a DPOS and an
21, and 31 blocked RTK or a code-based differential
C1 mimics a vehicle/receiver with relative position solution.
no visibility in the Northwestern Figure 6 shows a plot of the 2D
▲FIGURE 4 Sky view during the test.
part of the sky, whereas C2 mimics position error observed from each
a receiver without visibility in the receiver during the test as a function
East/Northeastern part of the sky. of driving environment. Overall, Type
Sky visibility restrictions do not B receiver shows better accuracy as
vary with the heading changes of the expected from a dual frequency high
vehicle. For example, for C1 receiver, quality receiver. However, it shows
Northwestern sky is always obstructed spikes of large error increases at
regardless of the vehicle orientation. times, mostly observed in the freeway
Figure 5 shows an example RF data scenario with large error excursions.
replay setup. The record and replay With Type A receivers, relatively
system was controlled through a PC larger errors are observed with the
and the recorded data was also stored limited-constellation receivers.
▲FIGURE 5 RF data replay set-up.
in the controller PC. The output RF Figure 7 shows the number of
38 GPS World | March 2011 www.gpsworld.com
Road | Transportation
▲▲ FIGURE 6 Position
error (2D) of each receiver as a function of driving
environment. ▲FIGURE 7 GPS satellites used by receivers.
▲▲ FIGURE 8 Freeway positioning accuracy. ▲▲ FIGURE 10 Local (east) positioning accuracy. ▲▲ FIGURE 12 Local (North) accuracy.
a relatively accurate position solution whenever it has B receiver were 58 percent of total driving time in urban
sufficient satellites visible. In the case of less than optimal canyon, 60 percent in the freeway scenario, 95 percent and
satellites availability, Type B receivers tend to show rapidly 99 percent in the local road scenarios.
degrading positioning accuracy, which may be reliably Velocity Domain Performance. For each test segment, velocity
detected using its quality indicators. estimates from each receiver were logged at the default data
rate of 4 Hz. For analysis purposes, North and East velocity
Position Error Distributions readings from each receiver were converted to 2D speed
Position error probability distribution functions were estimates. These were used with reference system speed
generated for the first three road segments using the estimates to generate 2D speed error statistics (Table 1).
time series data above. Figures 16–18 show these functions Based on Table 1, no significant biases or errors were
for Freeway, Local (East), and Local (North) segments, observed from any particular receiver or configuration.
respectively. They lead to these general conclusions: The only exception was the increased errors in the Urban
◾ Based on the mean and the spread of the distributions, Canyon segment, particular for C1 and C2. This is expected
Type B receiver has consistently provided the most
unbiased and accurate positioning performance out of all
the receivers considered. Overall, the output appears to
be unbiased, as should be the case for a high quality dual
frequency receiver with WAAS capability.
◾ Type A all-in-view receiver shows the next best overall
accuracy statistics with some biases in certain cases. These
biases can be time-of-day-dependent and may differ for
different times of the day or if observed over a longer time.
◾ Type A C1 and C2 receivers show very significant
vehicle-heading-dependent biases/errors. This is with very
limited sky view obstructions (that is, C1 only restricts less
than 1/8 of the entire sky view whereas C2 covers around
1/4) and with the same type of the receiver.
◾ Although enabling WAAS should theoretically help
minimize the biases observed in these tests, the availability ▲▲ FIGURE 9 Freeway positioning accuracy time series.
(open line-of-sight) of WAAS satellites for automotive
applications in these environments must be taken into
consideration for WAAS accuracy benefits to be applicable.
For these datasets, WAAS signals availabilities for a Type
▲▲ FIGURE 14 Urban canyon accuracy. ▲▲ FIGURe 11 Local (east) positioning accuracy time series.
▲▲ FIGURE 13 Local (North) positioning accuracy time series. ▲▲ FIGURE 15 Urban canyon positioning accuracy time series.
.to be a result of limited satellite availability in a challenging processing software in post-mission. The methods were:
environment with additional forced satellite eliminations. ◾ Differenced Positions (DPOS). Latitude and longitude reported
Virtual Two-Vehicle Analysis. Assume that Type A and Type by each vehicle were time-matched; distance between
A C1 receivers were located in two vehicles. Ideally, both the two points was calculated.
receivers should report the same location, as they were both ◾ Code and Carrier. Single frequency (L1) GPS RTK
connected to the same antenna on a single vehicle, creating positioning with float ambiguity estimation.
a zero-baseline scenario. However, as shown in the previous ◾ Code Only. GPS code measurements generated a relative
section, a meter-level separation was observed between the position solution.
two solutions. The 2D receiver separation results of this processing are
In this virtual two-vehicle scenario, relative position shown in Figure 19 as three subplots for freeway (top), local/
of one receiver (Type A) with respect to the other (Type east (middle), and local/north (bottom) scenarios. The 2D
A C2) was estimated by three methods, using GNSS data separation results for local scenarios show clear performance
▲▲ FIGURE 16 ▲▲ FIGURE 17Local road (east) position error ▲▲ FIGURE 18 Local road (north) position error
Freeway position error
distribution. distribution. distribution.
Realistic Randomization V
irtually all GNSS receiver test-
ing can be classified into one
of two broad categories: syn-
A New Way to Test GNSS Receivers thetic or realistic. The former typically
involves simulator-based trials, using
Alexander Mitelman a pre-defined collection of satellite
orbits, receiver positions, and signal
GNSS receiver testing should not be left to chance. Or should it? There are two propagation models (ionosphere, mul-
common approaches to testing GNSS receivers: synthetic and realistic. In syn- tipath, and so on). Examples of this
thetic testing, a signal simulator is programmed with specific satellite orbits, type of testing include the 3rd Genera-
receiver positions, and signal propagation conditions such as atmospheric tion Partnership Project (3GPP) mobile
effects, signal blockage, and multipath. A disadvantage of such testing is that phone performance specifications for
the models used to generate the synthetic signals are not always consistent assisted GPS, as well as the “apples-to-
with the behavior of receivers processing real GNSS signals. Realistic test- apples” methodology described in an
ing, on the other hand, endeavors to assess earlier GPS World article (see Further
receiver performance directly using the Reading).
signals actually transmitted by satellites. The The primary advantage of synthetic
signals may be recorded digitally and played testing is that it is tightly controllable
back to receivers any number of times. While and completely repeatable; where a
no modeling is used, the testing is specific high degree of statistical confidence is
to the particular observing scenario under required, the same scenario can be run
Innovation Insights which the data was recorded including the many times until sufficient data has
with Richard Langley satellite geometry, atmospheric conditions, been collected. Also, this type of test-
multipath behavior, and so on. To fully ing is inherently self-contained, and
The method can be examine the performance of a receiver using thus amenable to testing facilities with
data collected under a wide variety of sce- modest equipment and resources.
applied to any simulator narios would likely be prohibitive. So, neither Synthetic approaches have signifi-
cant limitations, however, particularly
scenario. testing approach is ideal. Is there a practical
alternative? The roulette tables in Monte when it comes to predicting receiver
Carlo suggest an answer. performance in challenging real-world
Both of the commonly used testing procedures lack a certain characteristic environments. Experience shows that
that would better assess receiver performance: randomness. What is needed tests in which signal levels are fixed
is an approach that would easily provide a random selection of realistic at predetermined levels are not always
observing conditions. Scientists and engineers often use repeated random predictive of actual receiver behavior.
samples when studying systems with a large number of inputs especially For example, a receiver’s coherent in-
when those inputs have a high degree of uncertainty or variability. And math- tegration time could in principle be
ematicians use such methods to obtain solutions when it is impossible or tuned to optimize acquisition at those
difficult to calculate an exact result as in the integration of some complicated levels, resulting in a device that passes
functions. The approach is called the Monte Carlo method after the principali- the required tests but whose perfor-
ty’s famous casino. Although the method had been used earlier, its name was mance may degrade in other cases.
introduced by physicists studying random neutron diffusion in fissile material More generally, it is useful to observe
at the Los Alamos National Laboratory during the Second World War. that the real world is full of random-
In this month’s article, we look at an approach to GNSS receiver testing ness, whereas apart from intentional
that uses realistic randomization of signal amplitudes based on histograms variations in receiver initialization, the
of carrier-to-noise-density ratios observed in real-world environments. It can primary source of randomness in most
be applied to any simulator scenario, independent of scenario details (posi- synthetic tests is simply thermal noise.
tion, date, time, motion trajectory, and so on), making it possible to control By comparison, most realistic test-
relevant parameters such as the number of satellites in view and the resulting ing approaches are designed to mea-
dilution of precision independent of signal-strength distribution. The method sure real-world performance directly.
is amenable to standardization and could help the industry to improve the Examples include conventional drive
testing methodology for positioning devices — to one that is more meaning- testing and so-called “RF playback”
fully related to real-world performance and user experience. systems, both of which have also been
described in recent literature (see Fur-
www.gpsworld.com March 2011 | GPS World 43
innovation | Signal Processing
0.15
bridge the gap between the two approaches described above.
It describes a novel method for generating synthetic sce-
0.10
narios in which the distribution of signal levels closely ap-
proximates that observed in real-world data sets, but with an
element of randomness that can be leveraged to significantly
0.05 expand testing coverage through Monte Carlo methods.
Also, the test setup requires only modest data storage and is
easily implemented on existing, widely deployed hardware,
0
B 15 20 25 30 35 40 45 50 making it attractive as a potential candidate for standardiza-
C/N0 (dB-Hz)
tion.
The approach consists of several steps. First, signal data
Center of mass = 22.48 dB-Hz
is gathered in an environment of interest and used to gen-
0.20 erate a histogram of carrier-to-noise-density (C/N0) ratios
as reported by a reference receiver, paying particular at-
tention to satellite masking to ensure that the probability
Normalized bin count
0.15
of signal blockage is calculated accurately. The histogram
is then combined with a randomized timing model to cre-
0.10
ate a synthetic scenario for a conventional GNSS simulator,
whose output is fed into the receiver(s) under test (RUTs).
The performance of the RUTs in response to live and simu-
0.05 lated signals is compared in order to validate the fidelity and
usefulness of the histogram-based simulation. This hybrid
approach combines the benefits of synthetic testing (repeat-
0
B 15 20 25 30 35 40 45 50 ability, full control, and compactness) with those of live test-
C/N0 (dB-Hz)
ing (realistic, non-static distribution of signal levels), while
▲▲Figure 1 Example histograms (top: a real-world urban canyon, the avoiding many of the drawbacks of each.
San Francisco financial district; bottom: 3GPP TS 34.171 “coarse-
time assistance” test case). Histograms
The method explored in this article relies on cumulative his-
ther Reading). Here, no modeling or approximation is in- tograms of C/N0 values reported by a receiver in a homoge-
volved; the receiver or recording instrument is physically neous signal environment. This representation is compact
operated within the signal environment of interest, and its and easy to implement with existing simulator-based test
performance in that environment is observed directly. The setups, and provides information that can be particularly
accuracy and fidelity of such tests come with a price, how- useful in tuning acquisition algorithms.
ever. All measurements of this type are inherently literal: Motivation and Theoretical Considerations. To motivate the
the results of a given test are inseparably linked to the spe- proposed approach, consider an example histogram con-
cific multipath profile, satellite geometry, atmospheric con- structed from real-world data, gathered in an environment
ditions, and antenna profile under which the raw data was (urban canyon) where A-GPS would typically be required.
gathered. In this respect, the direct approach resembles the This is shown in FIGURE 1, together with a representative his-
synthetic methods outlined above — little randomness ex- togram of a standard “coarse-time assistance” test case (as
ists within the test setup to fully explore a given receiver’s described in the 3GPP Technical Standard 34.171, Section
performance space. 5.2.1) for comparison. (Note that the x-axis is actually discon-
Designing a practical alternative to the existing GNSS tinuous toward the left side of each plot: the “B” column des-
tests, particularly one intended to be easy to standardize, ignates blocked signals, and thus corresponds to C/N0 = –∞.)
represents a challenging balancing act. If a proposed test is From the standpoint of signal distributions, it is evident that
too simple, it can be easily standardized, but it may fall well existing test standards may not always model the real world
short of capturing the complexities of real-world signals. On very accurately.
the other hand, a test laden with many special corner cases, The histogram is useful in other ways as well. Since the
or one that requires users to deploy significant additional data set is normalized (the sum of all bin heights is 1.0), it
44 GPS World | March 2011 www.gpsworld.com
Signal Processing | innovation
0.06
siderations. While the raw data for this study was obtained at
0.05 walking speeds (1 to 2 meters per second), and thus unlikely
0.04
to significantly misrepresent rapid C/N0 fading, studies done
at higher speeds (such as test drives) may require a refer-
0.03
ence receiver capable of producing C/N0 measurements at
0.02 a higher rate.
0.01 A third limitation is the sensitivity of the reference re-
0
ceiver. Ideally, the reference device would be able to track
B 10 20 30 40 50
C/N0 (dB-Hz) all signals present during data gathering regardless of signal
▲▲Figure 3 Data collection, Gamla Stan (Old Town), Stockholm (top:
strength, and would instantaneously reacquire any blocked
route and street view; bottom: cumulative histogram). signals as soon as they became visible again. Such a receiver
would fully explore the space of all available signals present
between successive starts). in the test environment. Unfortunately, no receiver is infi-
Once the data collection is complete, the reference data nitely sensitive, so a conventional commercial-grade high
set is processed with a current almanac and an assumed el- sensitivity receiver was used in this context. Thus the result-
evation angle mask (typically 5 degrees) to produce an in- ing histogram is, at best, a reasonable but imperfect approxi-
dividual histogram for each satellite in view, along with a mation of the true signal environment.
cumulative histogram for the entire set, as shown in Figure Finally, a potentially significant error source may be in-
3 (bottom). The masking calculation is particularly impor- troduced if the net effects of the reference receiver’s noise
tant in properly classifying which non-reported C/N0 values figure plus implementation loss (NF+IL) are not properly
should be ignored because the satellite in question is below accounted for in preparing the histograms. (If an active an-
the elevation angle mask at that location and time, and which tenna is used, the NF of the antenna’s low-noise amplifier es-
should be counted as blocked signals. sentially determines the first term.) The effect of incorrectly
In addition to proper accounting for satellite masking, the modeling these losses is that the entire histogram, with the
raw source data should also be manually trimmed to ensure exception of the “blocked” column, is shifted sideways by a
that all data points used to build the histogram are taken ho- constant offset.
mogeneously from the environment in question. Thus the The correction applied to the histogram to account for this
file used to generate the histogram in Figure 3 was truncated effect must be verified prior to further acquisition testing.
to exclude the section of “open sky” conditions between the This can be done by generating a simulator scenario from
start of the file and the southeast corner of the test area, and the histogram of interest, as described below, and recording
similarly between the exit from the test area and the end of a sufficiently long continuous data set using this scenario
the file. and the reference receiver. A corresponding histogram is
Finally, the resulting histogram is combined with a ran- then built from the reference receiver’s output, as before,
46 GPS World | March 2011 www.gpsworld.com
Signal Processing | innovation
tsim
t1 t2 t3 t4
sample) variance of individual C/N0 Constructing Scenarios
Adjust PRNs 03, 09, 27
tn ~ exp (λ)
readings observed during data collec- Once a target histogram is available, it
Adjust PRNs 08, 11, 12, 27, 29, 30, 32 tion as a measure of the stability of the is necessary to generate random signal
Adjust PRN 16
readings. While this is more physically amplitudes for use with a simulator
realistic than an arbitrarily chosen scenario. This is done by means of a
Adjust PRNs 03, 08, 09, 11, 12,
16, 21, 24, 27, 29, 30, 32
interval as described above, it is still technique known as the probability in-
a fixed interval. These observations tegral transform (PIT). This approach
PRN 03
suggest that a timing model including uses the cumulative distribution func-
tni = exp (N λ) some measure of randomness may rep- tion (or, in the discrete case considered
resent a more realistic approach. here, a modified formulation based
PRN 08
receiver for the two timing-thread the proposed method to be amenable λ-1 (s) D D
models described above and several to standardization, it is obviously nec- Indiv SV Multi SV
values of the Poisson/exponential pa- essary to gather data from many ad- 1.0 0.048 0.024
rameter, λ. The reference cumulative ditional environments. Indeed, it ap- 3.0 0.034 0.047
mass function (CMF) for each test was pears likely that no one histogram will
derived from the histogram generated encapsulate all environments of a par- 5.0 0.049 0.033
for the raw (empirically collected) data ticular type (such as urban canyons), so ▲▲TABLE 1 K-S test statistics (NF+IL = 4 dB).
set. significant additional experimentation NF+IL (dB) D
These results illustrate good agree- and data collection will be required
ment (D < 0.05) between the overall here. Also, as mentioned at the begin- 0 0.214
signal distribution profile in the empir- ning of the article, the proposed meth- 4 0.024
ical data set and that in each of the six od will need to be tested with multiple 8 0.106
simulated fluctuating scenarios. receivers to verify that a particular re- ▲▲ TABLE 2 Two-sided K-S test statistics for the
As a further check, TABLE 2 shows the sult is not unique to any specific brand “Multi SV” model (λ-1 = 1.0 sec).
same K-S statistic for the histogram or architecture. Finally, higher rate C/
generated from the “Multi SV” timing N0 source data may also be necessary on the paper “Fluctuation: A Novel
model as a function of several NF+IL to capture the rapid fades that may be Approach to GNSS Receiver Testing”
values. As before, the reference CMF encountered in dynamic scenarios, presented at ION GNSS 2010.
comes from the raw (empirically col- such as drive tests, and the fluctuation
lected) data set, and the same reference timing models will need to be revisited ALEXANDER MITELMAN is the GNSS research
manager at Cambridge Silicon Radio, head-
receiver was used to generate data once such data becomes available. quartered in Cambridge, U.K. He earned his
from the simulator scenario. Evidently, S.B. degree from the Massachusetts Institute of
an NF+IL value of 4 dB gives good Acknowledgments Technology and M.S. and Ph.D. degrees from
Stanford University, all in electrical engineer-
agreement between empirical and sim- The author gratefully acknowledges ing. His research interests include signal-quality
ulated data sets. the assistance of Jakob Almqvist, Da- monitoring and the development of algorithms
Validation of Receiver Performance. Fi- vid Karlsson, James Tidd, and Christer and testing methodologies for GNSS.
nally, TTFF tests with the simulated Weinigel in conducting the experi-
MORE ONLINE
scenarios described above are con- ments described in this article. Thanks
ducted with the same receiver(s) used also to Ronald Walken for valuable Further Reading
in the original data gathering session. insights on the accurate treatment of For references related to this article, go to gpsworld.com and click
on Richard Langley’s Innovation under Inside GPS
Here, the K-S test is used to compare the source environment in calculating World in the left-hand navigation bar.
the live and simulated TTFF results target histograms. This article is based
rather than signal distributions. An
1.0
example result, illustrating cumula-
tive distribution functions of TTFF, 0.9
is shown in FIGURE 5 for the live data
set collected during the original data 0.8
Cumulative probability of successful fix
A
s a geodetic surveyor, I served in the U.S Army for
10 years. During that time, my team and I developed
a nationwide GPS infrastructure system called the
Iraqi Geospatial Reference System (IGRS). We installed
Continuously Operating Reference Stations (CORS) and High
Accuracy Reference Network Stations (HARNS), the first
Iraqi owned and maintained system of its type.
As a native Arabic speaker, my role was to train the Iraqi
engineers to install additional CORS, as well as update and
maintain the IGRS as a part of the International GNSS Service
(IGS) network to sustain the accuracy of engineering and map-
ping projects. The IGRS was critical to other major infrastruc-
ture projects in the effort of rebuilding the battered nation, such
▲▲ Malkawi discussesinstallation
of Iraqi operated and ▲▲ The resultof traveling in
▲▲ Soldiers establish geodetic control for an airport aeronautical maintained CORS with military vehicles over roads
survey. Al-Hassani. infested with IED.
Copyright 2011 QUESTEX MEDIA GROUP llc All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical including by photocopy, recording, or information storage and retrieval without permission in writing from the publisher,
Questex Media Group LLC. Authorization to photocopy items for internal/educational or personal use, or the internal/educational or personal use of specific clients is granted by Questex Media Group LLC for libraries and other users registered with the Copyright Clearance Center, 222 Rosewood Dr., Danvers, MA 01923, 978-750-8400
fax 978-750-4470. For uses beyond those listed above, please direct your written request to questexpermissions@theygsgroup.com or 800-494-9051, Ext. 100.
privacy notice: Questex Media Group provides certain customer contact data (such as customers’ names, addresses, phone numbers and e-mail addresses) to third parties who wish to promote relevant products, services and other opportunities which may be of interest to you. If you do not want Questex Media Group to
make your contact information available to third parties for marketing purposes, simply call toll-free 866-344-1315 or 1-847-763-9594 (Outside the US) between the hours of 8:30 am and 5 pm (CT) and a customer service representative will assist to remove your name from Questex’s lists.
GPS WORLD (ISSN 1048-5104) is published monthly (12 issues/yr) by Questex Media Group LLC, 306 W Michigan St, Ste 200, Duluth, MN 55802. SUBSCRIPTION RATES: U.S. and possessions — 1 year (12 issues), $73.70; 2 years (24 issues), $128.70. Canada and Mexico — $95.70, 1 year; $150.70, 2 years. All other
countries — $139.70, 1 year; $255.20, 2 years. International pricing includes air-expedited service. Single copies (prepaid only): $7 in the United States, $9 all other countries. Back issues, if available, are $19 in the U.S. and possessions, $23 all other countries. Include $6.50 per order plus $2 per additional copy for U.S. postage
and handling. Periodicals postage paid at Duluth, MN 55806, and additional mailing offices. POSTMASTER: Please send address changes to GPS WORLD,
P.O. Box 1270, Skokie, IL 60076-8270, USA. Canadian G.S.T. number: 840 033 278 RT0001, Publications Mail Agreement Number 40017597. Printed in the U.S.A.