GNSS 20090301 Mar 2009

Previous Page | Contents | Zoom in | Zoom out | Front Cover | Search Issue | Next Page q
q
M
M
q
q
M
M
qM
Qmags
THE WORLDS NEWSSTAND
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_____________________________________________________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
INSIDE GNSS DIGITAL
Explore our new digital edition
You have told us that Inside GNSS in print is perfect for the plane.
Our new digital edition is even more perfect for the times you would like to travel light, or share articles
with colleagues across the ocean, or archive your favorite searchable issues when the bookshelves
are full.
Give it a try!
N A V I G A T I O N K E Y
Previous/Next Page:
Contents:
Zoom in/out:
Return to the front cover:
Search:
Drag:

Browser version: click on the page to zoom in, and again to zoom out.
PDF: use the zoom in/zoom out buttons in the top or bottom tool bar.

Browser version: click the double arrow button in the Navbar
PDF: click on the Front Cover in the Tool bar.
Click on the search button to perform a full keyword or phrase search. You can
search a single issue, or your complete archive.
When you have zoomed in on a page in the browser, a left click and hold allows you to
drag the page around, into, and out of view.
mouse click or keyboard on left /right arrows to turn page
takes you to the Table of Contents page
Subscribe to Inside GNSS
Share with a colleague
March-April 2009
www.insidegnss.com
GNSS SOLUTIONS
Removing the Bias
from GLONASS Frequencies
GPS SPOOF DETECTOR:
Using Multiple Antennas
to Spot a Phony Signal
GNSS WORLD:
What Race?
What Competition?
On the Trails
of the Inuit
Contents | Zoom in | Zoom out Search Issue | Next Page For navigation instructions please click here For navigation instructions please click here
Contents | Zoom in | Zoom out Search Issue | Next Page For navigation instructions please click here For navigation instructions please click here
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_________________
_________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
___________________
_______________
_________________________
_______________
_______________
4
InsideGNSS MA RC H/A P RI L 200 9 www.insidegnss.com
TECHNICAL ARTICLES
40 AMulti-Antenna
Defense
Receiver-Autonomous GPS Spoong Detection
Paul Y. Montgomery, Todd E. Humphreys, and
Brent M. Ledvina
Spoong GNSS is no joke and the risks are
real. But solutions exist to thwart the use of
intentionally spurious signals, as this case study
shows in the use of multiple antenna receivers to
detect spoong attempts.
TABLE OF CONTENTS
MARCH/APRIL 2009
VOLUME 4
NUMBER 2
TOC BY THE
NUMBERS
8
Thinking Aloud
10
360 Degrees
10
GNSS Hotspots
ARTICLES
16
GNSS World
18
Human Engineeringt
24
GNSS Solutions
30
On the Trails of the Inuit
40
GPS Spoong Detection
48
EASY Suite 2
DEPARTMENTS
53
Industry View
54
Advertisers Index
54
GNSS Timeline
COVER STORY
30 OntheTrailsofthe
Inuit
Adapting GPS for Nunavut Hunters and
Environmental Research
Kyle OKeefe, Shari Gearheard, Gary Aipellee,
Apiusie Apak, Jayko Enuaraq, David Iqaqrialu,
Laimikie Palluq, Jacopie Panipak, Amosie
Sivugat, Brandon Culling, Josiah Lau, Tina Moss-
tajiri, Trevor Phillips, Michael Brand, Ryan Enns,
and Edward Wingate
Keeping track of the location, conditions, and
uses of Arctic trails over time and space uncov-
ers a great deal about the environment and
relationship of people to their land. The Igliniit
(Trails) Project is making change visible with its
integrated GPS system for expert Inuit hunters
from Nunavut, who log thousands of kilometers
each year on dogsleds and snowmobiles.
P
h
o
t
o
:

E
d
w
a
r
d

W
i
n
g
a
t
e
48 GPSEASYSuiteII
A Matlab Companion
Kai Borre
GNSS practitioners are familiar with Matlab, a
computing language and interactive environ-
ment for algorithm development, data visualiza-
tion, data analysis, and numeric computation.
In this new series, a prominent Danish GNSS
researcher uses Matlab to illustrate and explain
some common GPS issues.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
ENGINEERING SOLUTIONS FROM THE GLOBAL NAVIGATION SATELLITE SYSTEM COMMUNITY
March/April 2009 Volume 4/Number 2
EDITORIAL
Editor & Publisher Glen Gibbons glen@insidegnss.com
Art Director Tim Jordan
Graphic Artist Gwen Rhoads
Circulation Director Peggie Kegel
Contributing Editor for Working Papers: Gnter Hein
Guenter.Hein@unibw-muenchen.de
Contributing Editor for GNSS Solutions:
Mark Petovello mark.petovello@ucalgary.ca
Technical Editor Hans J. Kunze klaglobal@earthlink.net
Contributing Writers/Copyeditors Eliza Schmidkunz,
Melody Ward Leslie
Web Designer/Developer Mike Lee
Web Editor Sierra Robinson
IT Technical Support Elijah Buck
Circulation Assistant Anna Liv Gibbons
MARKETING AND PUBLIC RELATIONS
Director/Partner Eliza Schmidkunz eliza@insidegnss.com
ADVERTISING
sales@insidegnss.com
Telephone: 408-216-7561 Fax: 408-216-7525
PUBLISHED BY GIBBONS MEDIA & RESEARCH
1574 Coburg Road No. 233
Eugene, Oregon, 97401-4802 USA
Telephone: 408-216-7561
Fax: 408-216-7525
Copyright 2009 Gibbons Media & Research 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
Gibbons Media & Research. Authorization is granted to photocopy items, with attribution, for
internal/educational or personal non-commercial use. For all other uses, contact Glen Gibbons.
INSIDE GNSS (ISSN 1559-503X) is a controlled circulation magazine, published six times a year. Inside
GNSS is a registered trademark of Gibbons Media and Research LLC. Postage paid at Lebanon Junction
MPO, KY 40150-9998, Mail Permit #473. INSIDE GNSS does not verify any claims or other information
in any of the advertisements or technical articles contained in the publication and cannot take
responsibility for any losses or other damages incurred by readers in reliance on such content.
Editorial Advisory Council
VIDAL ASHKENAZI
Nottingham Scientic Ltd., Nottingham, United Kingdom
JOHN BETZ
MITRE Corporation, Bedford, Massachusetts, USA
PASCAL CAMPAGNE
France Developpement Conseil, Vincennes, France
MARIO CAPORALE
Italian Space Agency, Rome, Italy
PER ENGE
Stanford University, Palo Alto, California, USA
MARCO FALCONE
European Space Agency, Noordwijk, The Netherlands
SERGIO GRECO
Thales Alenia Space, Rome, Italy
JEAN-LUC ISSLER
CNES, Toulouse, France
CHANGDON KEE
Seoul National University, Seoul, Korea
MIKHAIL KRASILSHCHIKOV
Moscow Aviation Institute, Moscow, Russia
SANG JEONG LEE
ChungnamNational University, Daejon, Korea
JULES MCNEFF
Overlook Systems Technologies, Inc., Vienna, Virginia, USA
PRATAP MISRA
MITRE Corporation, Cambridge, Massachusetts, USA
BRAD PARKINSON
Stanford University, Palo Alto, California, USA
TONY PRATT
Professor and Consultant, United Kingdom
SERGEY G. REVNIVYKH
Federal Space Agency, Korolyov, Russian Federation
MARTIN RIPPLE
Thales ATM, Melbourne, Australia
CHRIS RIZOS
University of New South Wales, Sydney, Australia
TOM STANSELL
Stansell Consulting, Rancho Palos Verdes, California, USA
RAYMOND J. SWIDER
Ofce of the Assistant Secretary of Defense for Network and
Information Integration , Washington D.C. USA
A.J. VAN DIERENDONCK
AJ Systems, Los Altos, California, USA
JRN TJADEN
European Space Agency, Korou, French Guiana
FRANTISEK VEJRAZKA
Czech Technical University, Prague, Czech Republic
PHIL WARD
Navward Consulting. Garland, Texas, USA
CHRISTOPHER WILSON
Tele Atlas, Redwood City, California, USA
LINYUAN XIA
Sun Yat-Sen University, Guangzhou, China
AKIO YASUDA
Tokyo University of Marine Science and Technology, Tokyo, Japan
SUBSCRIBE ONLINE
FREE one-year subscriptions are available to qualied readers who work in
GNSS-related companies, organizations, research institutes, government
agencies and the military. Subscribe online at www.insidegnss.com/subscribe.
You may also change your information, renew your subscription, unsubscribe,
and nd out how to buy individual copies, past issues, and institutional and
library subscriptions at www.insidegnss.com/subscriptionservices
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_____________
______________
__________________
_______________
______________________
______________
www.insidegnss.com MA RC H/A P RI L 200 9 InsideGNSS 7
24GNSS Solutions
What About GLONASS Inter-Frequency
Biases?
Mark Petovello with Frank Takac
Working Papers will return in the
May/June issue.
8 Thinking Aloud
Whither Galileo, and Why?
Glen Gibbons
10 360 Degrees
News fromthe world of GNSS
More Problems with GPS IIF;
L5 Deadline Nears
GNSS Hotspots
FY10 Budget,2008 FRP
Iridium, Russian Space Collision
16 GNSS World
What Race? What Competition?
18 Human Engineering
Grace Xingxin Gao
53Industry View
54Advertisers Index
54GNSS Timeline
Calendar of Events
TABLE OF CONTENTS
COLUMN
DEPARTMENTS
Grace Gao,
Human Engineering
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
________________
_____________
8
T
aken on their own, the words
were chilling, Galileo may
simply be too late to matter.
And perhaps more chilling
in context.
Tey were spoken by Frank van
Diggelen, technical director and chief
navigation ofcer for semiconductor
manufacturer Broadcom, which is
combining GPS with Bluetooth and
other RF technologies to provide
location capability for consumer
products.
Van Diggelens comments were
directed to the topic of the panel
discussion I was moderating at the
recent Munich Satellite Navigation
Summit GNSS and mobile phone
navigation and the prospects for
Galileo getting on board with GPS.
Te news for Galileo wasnt much
better from Greg Turetzky, a panelist as
well as the marketing director at SiRF
Technology, another GNSS chipset
manufacturer that had long supported
Galileos development.
Galileo will be a requirement
either by the market or regulation,
but the value is dependent on time,
Turetzky said. And time is what
Galileo may be running out of.
Te panel comments have particular
resonance because the Galileo program
itself anticipates that mobile handsets
represent the majority of the future
GNSS market in unit volume, and that
Galileo chips will have a place there.
Te irony is that Galileos place in
the GNSS sun was its to lose.
Early in 1994, when serious
discussions began on a European
GNSS program and even more so
in 1999, when the program gained
its name Galileo appeared well-
positioned to become the number
two GNSS system. Te collapse of
the Soviet Union and the tumultuous
years under former President Boris
Yeltsin had lef Russias GLONASS
system on the brink of extinction.
But the task of getting many
separate European Union (EU) member
states to agree on a large infrastructure
project and the ultimately fruitless
efort to form a Public-Private
Partnership to implement it (as part
of the associated politics) bogged
the initiative down. Ambitious eforts
to leverage the civil GNSS as an asset
for foreign policy, attracting extra-
European participants such as China
and Israel to the program, further
complicated the process.
Te European program has shown
a remarkable predilection to snatching
defeat from the jaws of victory a
tendency borne out by the preliminary
observations of the European Court of
Auditors in a report on management of
the Galileo programs development and
validation phase.
One conclusion: Te programmes
governance was poor. Te division
of roles between the entities involved
in the development and validation
phase of the programme . . . was not
clearly defned. Particularly from 2005,
the [European] Commission failed
to provide adequate leadership in
developing and managing Galileo.
Te auditors comprise an indepen-
dent, professional investigatory agency
established under EU treaties to assess
whether the EU budget has been
implemented correctly.
In their preliminary report (about
which we will write more soon), the
auditors note, Te Galileo programme
organisation has changed markedly
since 2007.
Te continuing question, of course,
is whether it has changed sufciently
and soon enough to make a diference.
Adding to its self-inficted woes,
the Galileo program now faces
a profoundly diferent external
environment than it did 15 years ago.
Russia is well along in re-
establishing and modernizing
GLONASS. And China, in addition
to announcing an ambitious launch
schedule that would bring Compass
satellites online ahead of Galileo,
proposes to overlay signals on the
security-oriented Public Regulated
Service (PRS) from which the Galileo
program expects to gain one-third of
its revenues.
Ultimately, many of these
considerations will not matter, because
the main driver for building Galileo
has been European desire and,
arguably, need for an independent
GNSS under its control to secure
a critical infrastructure for its 27
member states.
If sovereignty is Galileos trump
card, it can be played at any time,
and when the system is here and fully
operational, that will be soon enough.
GLEN GIBBONS, JR.
Editor
The continuing question is whether management of Galileo has
changed sufciently and soon enough to make a difference.
THINKING ALOUD
Whither Galileo, and Why?
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_____________
____________
10
More Problems
with GPS IIF; L5
Deadline Nears
D
iscovery of a power anomaly
in signal generator of the frst
GPS Block IIF space vehicle
(SV) has thrown a new wrinkle
into the long-delayed follow-on genera-
tion of spacecraf.
Meanwhile, its all systems go (and
all fngers crossed) for launch of a mod-
ernized Block IIR satellite IIR-20(M)
with the demonstration L5 navigation
payload designed to secure priority GPS
access to the frequency.
In the words of a GPS Wing spokes-
man at the Space and Missile Systems
Center (SMC), Los Angeles Air Force
Base, California, In reviewing test data
from the fnal phase of SV1 thermal vacu-
um test, [government and Boeing mission
assurance teams] identifed a new concern
that a component in the L2 transmitter
may not have sufficient design margin
to operate at its highest required power
throughout the satellite lifetime.
The anomalies resulted in power
surges that reportedly afect all signals
on the L2 frequency and appear likely to
delay transport of the spacecraf to the
launch site in Florida.
Designed and built by the Boeing
Company, the IIF satellites carry a naviga-
tion payload with high-power amplifers
built like the payloads on all versions
of the GPS satellites to date by the ITT
Corporation in Clifon, New Jersey.
Boeing has identified multiple
options for addressing the concern and
is working parallel solutions to deliver
redesigned transmitters this summer,
said the GPS Wing spokesman.
The GPS Wing and Boeing are
reviewing the shipment plans in con-
junction with the corrective actions to
determine when to ship a satellite to the
360 DEGREES
Cape Canaveral launch site for contin-
ued testing. Te Wing may choose to
ship SV2 for testing while it completes
actions on SV1, said the spokesman.
SV2 would then return to Los Angeles
AFB for further work.
Launch of the first IIF had been
expected in October. Now the GPS
Wing predicts that launch only as late
2009 with a second launch not projected
for sooner than six months afer the frst
launch. SV1 is awaiting Final Integrated
System Test; SV2 is in Initial Integrated
System Test.
Meanwhile, the U.S. Air Force is in
a race against the clock to get the new
L5 signal on the air by August 26, 2009,
in order to meet an International Tele-
communications Union (ITU) deadline
IMAGE CREDITS: 1. CAOQUICKLOOKS; 2.OFFICE OF THE ALASKA LIUTENANT
GOVERNOR; 3. GPSSIRF; 5. GALILEO MUNICH POSTER.
Hawaii (the Big Island)
Aliens Invasive trees in
Hawaiian rainforest preserves
form dense thickets, shade
out native species and are
spreading fast. A strategy
that uses LIDAR, GPS,
and advanced imaging
spectrometers gave Carnegie
Airborne Observatory and the
U.S. Forest Service detailed
maps of Big Island ecosystem
changes in 850 square miles
that had been impossible
to survey before. (The pink
trees dont belong there.)
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_________________
_____________
for securing a preferential L5 frequency
allocation for GPS operations.
Te frst and probably only oppor-
tunity to meet the deadline: a modern-
ized GPS Block IIR-M satellite IIR-
20(M) with an experimental L5 signal
demonstration payload scheduled for a
March 24 launch.
IIR-20(M) had originally been set to
fy in June 2008, but detection of a sus-
pected faulty component in a Delta II
rocket during a pre-launch check led to
postponement of IIR-M launches until
the component could be replaced.
And if problems should arise with
the IIR(20)-M satellite or launch? Origi-
nally, the U.S. planned to meet the dead-
line with the frst IIF satellite, said the
spokesman. Te IIR-20 demo payload
was developed as the back-up plan.
Should the IIR(20)-Ms L5 signal not
get on the air, it could deprive GPS of
unconditional use of the frequency.
FY10 Budget,
2008 FRP Throw
Doubt on eLoran
P
resident Barack Obama appears
to have weighed in on the long-
running Loran/eLoran debate
on the side of terminating the
terrestrial radionavigation system and,
apparently, its enhanced version that had
been proposed as a backup to GPS.
In a February 26 message to U.S.
Coast Guard (USCG) members, Vice-
Admiral V. S. Crea, USCG Vice Com-
mandant and Chief Operating Ofcer,
said the Fiscal Year 2010 (FY10) budget
outlined in a document sent to Congress
calls for termination of Loran-C in the
coming year.
Specifcally, Crea cited a passage from
a section on potential savings across the
federal government that proposed the
termination of outdated systems such as
the terrestrial-based, long-range radio-
navigation (LORAN-C) operated by the
U.S. Coast Guard resulting in an ofset
of $36 million in 2010 and $190 million
over fve years.
During a question and answer period
following his State of the Coast Guard
address on March 3, the USCG Com-
360 DEGREES
GNSS Hotspots
Oyster Bay, NewYork
Chipset Champs ABI
Research ranked Broadcom
rst for innovation and tops
overall in its sophisticated
GPS Integrated Circuit
Manufacturer Vendor Matrix.
SIRF Technology is rst in
innovation and second overall
and Texas Instruments placed
third. The Oyster Bay, New
York company announced
results on March 12. See
article on page 53.
Austin, Texas
Hats off! A GPS-enabled
bowler hat arrived at the
South by Southwest 2009
trendsetting arts and media
festival in Austin, Texas,
courtesy of two British
gamesters and the Arts
Council of England. Rules
of the game: Find the hat,
get the hat, run fast. Magic
words: Pardon Me, I Believe
You Have My Hat. Players
followthe hat in real time
on Google maps and plot an
interception.
Anchorage to Nome, Alaska
Iditarod GPS The 2009 Alaskan
Iditarod mandated Iridium
communications and GPS-enabled
tracking devices on all 68 starting
dogsleds this year. Fans love to follow
the mushers in real time during the
1,500-mile race fromAnchorage to
Nome. When the sleepy race leader
missed the trail on March 13, everyone
knewabout it.
Munich, Germany

Systems Report Notes
fromthe March Munich
Satellite Navigation Summit:
GLONASSs rst CDMA signal
goes on the L3 frequency
overlapping one of Galileos,
Compass/Beidou aims for
regional capability in 2010
with full constellation 5 to
10 years later; tensions
between Europe and China
on frequency overlays on the
EUs security service; potential
Galileo cost overruns; GPS to
secure frequency with an L5
demonstration payload on
upcoming satellite launch.
See GNSS World article, page
16.
GLONASS
Compass
Military
Breaking
Policy
Signal Launch
Other Systems Galileo
GPS
Bright Idea
Technology
Commercial
Consumer
Satellite
History
Glitch
Conference
International Space Station
Oops. In 1993, GPS satellite
37 rode a Delta II rocket and
dutifully went into Middle
Earth Orbit. Meanwhile, a Yo
weight off the PAM-Dstage
joined thousands of other
micrometeoroids and orbital
debris in space. On March 12,
its changeable orbit brought
the errant debris into the 2.8
mile safety zone around the
International Space Station.
The crewevacuated briey to
their Soyuz lifeboat.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
12
Iridium, Russian
Space Collision
Raises GNSS Risks
T
he collision of an active Iridi-
um satellite and a defunct Rus-
sian military communications
satellite underlines a concern
raised recently at the International
Committee on GNSS (ICG) about the
need to include coordination of space
vehicle (SV) operations.
On March 4, Iridium announced
that it had completed the replacement
of the operational Iridium satellite lost
in the February 10 collision.
In comments during a session of
Working Group A (compatibility and
interoperability) during the ICGs
December meeting in Pasadena, Cali-
fornia, Professor Grigory Stupak of the
Russian Institute for Space Device Engi-
neering (RISDE), noted that Russias
position is that compatibility includes
more than just sig-
nal s, i ncl udi ng
among other things,
t he shar i ng of
orbits and disposal
of satellites.
Similar concerns
have been expressed
in recent years by
organizations that
have studied t he
possibility of GNSS
satellites, particu-
larly those that have been parked in
disposal or graveyard orbits, to drif and
possibly collide with one another.
More than 2,000 satellites, active and
decommissioned, are currently orbiting
our planet, and many times that amount
of sizable objects characterized as space
debris mostly within a few hundred
miles of Earth.
Iridium ofcials said the recent inci-
dent demonstrated the need for even
more aggressive action and called for
enhanced actions to increase the margin
of safety for space operations.
Tese include long-term investment
to improve Space Situational Awareness
(SSA) so that the space environment can
be better understood and characterized,
as well as improved information sharing
between industry and the U.S. govern-
ment to improve the timeliness and
accuracy of conjunction assessment and
warning.
More Room at MEO. To be sure, the low
earth orbits (LEO) of Iridium 33 and
Cosmos 2251, which collided at an alti-
tude of just under 500 miles, comprise a
much smaller space more heavily popu-
lated by SVs than the middle earth orbits
(MEO) in which GNSS satellites operate.
Te volume of a spherical space increas-
es exponentially with its radius or, in the
case of orbiting satellites, their distance
from the center of the Earth.
In comparison to LEOs, GPS satel-
lites orbit at an altitude of approximately
12,600 miles (20,200 kilometers); Russias
GLONASS SVs, at 11,842 miles (19,100
kilometers); Galileo 14,397 miles (23,222
kilometers); and Compass, 13,113 miles
(21,150 kilometers). Moreover, because
of their system design, GNSS spacecraf
transmit their orbital positions (epheme-
redes) as part of their navigation mes-
sages. Consequently, operational satel-
lites are readily trackable.
Decommissioned satel l ites are
switched of and placed into a disposal
orbit higher than the operational con-
stellation. Out-of-service GPS satellites,
for instance, are boosted about 500
kilometers (310 miles) further out into
space.
With implementation of a new launch,
anomaly resolution and disposal opera-
tions system (LADO) in 2007, operators
with the U.S. Air Force 2nd and 19th
Space Operations Squadrons at Schriever
Air Force Base, Colorado, took over that
responsibility for GPS. LADO success-
fully performed its frst disposal for
GPS SVN 29 in November 2007.
Nonetheless, satellites can drif out
Earth satellites and orbital
debris. NASA image
mandant, Admiral Thad Allen, was
asked about the budgetary proposal
that would leave the nation without a
primary backup to GPS.
In his reply, Allen made a distinc-
tion between operating LORAN and a
backup to GPS. While one could be the
other, it doesnt necessarily have to be,
he said. Te policy decision was taken
to terminate LORAN-C. Negotiations,
discussions and outreach to stakehold-
ers will continue for a requirement for
a backup for GPS. Should that backup
become eLORAN, that is something that
will be addressed in the future.
Noting that LORAN-C stations in
Alaska are operating with 1960s vacu-
um-tube technology, Allen called for an
up or down vote on the system. Any
requirements for a backup to GPS need
to be generated in the interagency [pro-
cess], Allen added, and the Depart-
ment of Homeland Security will have
the lead on that efort.
After efforts by several agencies to
end the program in recent years, despite
the recommendation of an Independent
Assessment Team (IAT) that eLoran
could serve as a back, the Department
of Homeland Security (DHS) appeared
to have embraced completing modern-
ization of the system in former President
Bushs FY09 budget.
Language in the 2008 Federal Radi-
onavigation Plan (FRP) released last
month, however, made the status of the
system appear a little ambiguous.
Although eLoran appeared to have
been confrmed early last year as a back-
up system for GPS with a funding com-
mitment from DHS in line with an
Independent Assessment Teams recom-
mendation, the 2008 FRP seems to leave
the systems status a little uncertain.
360 DEGREES
360 Degrees continued on page 15
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
of their disposal orbits over time.
Although initially circular, as with
those of operational SVs, disposed
satellites orbits eventually become
increasingly elliptical, mostly as the
result of sun-moon gravitational
efects on the spacecraf.
Disposal Orbit Studies. In studies
for the Air Force early in this decade,
Aerospace Corporation researcher
Chia-Chun (George) Chao and col-
leagues predicted that GPS satellites
placed in disposal orbits will even-
tually in 20 to 40 years drift
down into the operating constella-
tion. Besides intersecting the GPS
constellation, these satellites eventu-
ally could pose a threat to operational
satellites in low Earth and geosynchro-
nous orbits, Chao said.
GLONASS, which has about 100
failed satellites within its constellation,
is located about 1,000 kilometers (621
miles) lower than GPS and could pose a
collision problem in 40 years, the Aero-
space studies showed. Similar issues are
associated with the Galileo and Com-
pass constellations both active and
defunct spacecraf.
The tendency for circular orbits to
become elliptical is strongly dependent
on orbit inclination and altitude, accord-
ing to Chao and fellow Aerospace
researcher Anne Glick. The effect
becomes more pronounced for Galileo
orbits due to their higher altitude.
Chao suggested that the dete-
rioration of GNSS satellites orbits
could be ameliorated by reducing
the initial eccentricity of the disposal
orbit through precise satellite burn
maneuvers. Further, the growth of
orbit eccentricity could be reduced by
circularizing the initial disposal orbit
as much as possible and by achieving a
more favorable argument of perigee,
or orientation of the perigee relative to
the equatorial plane.
In a 2004 report on upper stage
disposal of satellites, the Inter-Agency
Space Debris Coordination Committee
(IADC) working group on mitigation
considered MEO disposal for naviga-
tion satellites, but felt that further work
is required to characterize the collision
risk in the region before recommenda-
tions can be formalized.
GNSS operating zone
GPS disposal orbit
Degraded GPS disposal
orbit after 2040 years
Degraded GPS disposal
orbit after 160180 years
LOW-EARTH ORBIT
EARTH
GEOSTATIONARY ORBIT
Graphic design based
on concept originally
presented by The Aerospace Corporation
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_______________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_____________________________________
______________________________________
_____________
In the policy section describing the
status of a backup system in case GPS
service is compromised or rendered
uncertain, however, the FRP says that
DoT has determined that sufcient alter-
native navigation aids currently exist for
aviation, commercial maritime, rail, and
highway in the event of a loss of GPS-
based services.
[T]herefore Loran currently is not
needed as a back-up navigation aid for
transportation safety-of-life users, th
plan concludes.
Te DoD has issued a similar state
ment regarding its user requirements.
Te FRP continues, DHS is deter-
mining whether alternative backups
or contingency plans exist across the
critical infrastructure and key resource
sectors identifed in the National Infra-
structure Protection Plan in the event of
a loss of GPS-based services. An initial
survey of the Federal critical infrastruc-
ture partners indicates wide variance in
backup system requirements. Terefore,
DHS is working with Federal partners to
clarify the operational requirements.
Despite its routine absence from
administration budget proposals since
1994, Loran has been rescued by con-
gressional advocates who have provided
$160 million since 1997 to enhance the
Loran system and conduct research to
make eLoran a viable backup to GPS.
Tose funds have helped modernize
and add to the network of Loran trans-
mitters that cover the United States,
equipping every station with three cesi-
um clocks, while encouraging miniatur-
ization of Loran receiver technology able
to use new all-in-view techniques that
provide positioning and timing accuracy
approaching that of GPS.
FRP 2008. Compared to its 2005 pre-
decessor, this foundational document
which refects the ofcial radionavi-
gation policy and planning for the fed-
eral government is about 2-1/2 times
as large: 183 pages versus 72 pages for
the previous version.
Addition of three new sections
roles and responsibilities, radio-
navigation system user requirements,
and extensive appendices covering geo-
detic datums, reference systems, system
parameters, and description account
for much of the bulked up qualities of
the new FRP. Plus the 2008 version ofers
16 fgures and 19 tables, compared to a
mere 3 fgures in the 2005 document.
As a result, the 2008 FRP provides
a much richer resource than previous
versions.
Issued under the signatures of rank-
ing ofcials at the U.S. departments of
transportation, defense, and homeland
security, preparation of the FRP is coor-
dinated with other civil federal agencies
by the Department of Transportation
(DoT) Research and Innovative Tech-
nology Administration (RITA).
360 Degrees continued from page 12
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_________
16
M
unichs high-level satnav summit the frst week in
March opened with a plenary titled, Te World-
wide Race in GNSS and closed with a panel, Te
Competition among the Big Four.
Despite the provocative session titles, few speakers were
willing to admit that either a race or a competition was under
way in the GNSS world. Visa problems reportedly kept a Chi-
nese spokesperson from joining the GNSS race session, and
the competition panel was opened by GPS pioneer Brad
Parkinson invoking the motto of GNSS interchangeability:
Any four [satellites from any system] will do.
Indeed, one way of looking at the Summits premise is
that the United States already won both the race and the
competition in late 1993 with a declaration of initial opera-
tional capability (IOC) and 24 operational GPS satellites on
orbit. Te Russian Federation came in second in 1995.
End of story.
But within the conferences dozen panel discussions and
inevitable hallway conversations lurked many indications
that the race continues and the competition is ferce.
China squeezed out a few additional details on its imple-
mentation plans, announcing that three more Compass
satellites would be launched this year, including one in the
frst half of 2009, and seven in 2010. Russia announced its
decision to put CDMA signals on the new GLONASS civil L3
band centered at 1208 MHz.
Galileo representatives put a brave face on a program that
continues to encounter adversity at home and abroad. As did
U.S. ofcials for a GPS program that has had nearly a year-
long halt in its launch schedule due to a questionable compo-
nent in the Delta II rocket, and now may have encountered
new problems in the next-generation Block IIF satellites.
Te general downplaying of a GNSS competitive race
might best have been refected in the observation of Mike
Shaw, director of the U.S. National Coordination Ofce for
Space-Based Positioning, Navigation, and Timing (PNT):
Te race should not be among the provider nations and the
services they ofer. Tey [GNSS providers] should focus on
the issues of compatibility and interoperability. Te race is
really in the [GNSS equipment and services] industry sector.
Despite the denials of competition, a race of sorts is being
run in the GNSS world. What kind of race? Arguably, its
a marathon. All of the programs have planning processes
under way that reach to 2020 or beyond.
Other aspects of the situation, however, give the impres-
sion of a sack race, with two or more GNSS providers running
in tandem under bilateral and multilateral accords, each with
one leg in the same sack. Or within a few years it could even be
likened to leapfrog as each round of system modernization pro-
pels a GNSS provider temporarily to the front of the pack.
GLONASS. In some ways, the GLONASS program afer
an allocation of more than 100 billion rubles (nearly $3 bil-
lion) in funding for its 20022011 modernization efort
has progressed most steadily in recent years.
With the three newest satellites from a launch last
December now in operation, GLONASS has a 20-bird con-
stellation including 19 modernized space vehicles (SVs),
the most in more than a decade. Some 17 of the spacecraf are
broadcasting a second full civil signal on the GLONASS L2
frequency, the only such GNSS system doing so.
Its signal-in-space user range error (URE) is down to 1.8
meters still high compared to GPSs 1-meter URE, but
GLEN GIBBONS
Despite the best efforts of a Munich conferences organizers to characterize the four
programs as a competitive race, representatives of the worlds GNSS operators insist
the its all about compatibility and interoperability well, almost all . . . .
GNSS WORLD
What Race?
What Competition?
Sergey Revnivykh,
Russian Space Agency
Col. Dave Buckman,
U.S. Air Force Space Command
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
phones playing a prominent role in down-
stream markets, he added, Galileo may
simply be too late to matter.
COMPASS. As for Compass, if China
executes its currently announced sched-
ule for satellite launches, it would mark a
substantial acceleration in that program.
Given the caution with which Chinese
ofcials have announced their plans, the
10 satellites in two years commitment
could well be met.
Indeed, a Chinese representative indi-
cated that the Compass program is under
pressure from Beijing to show progress
in bringing the planned fve civil and fve
restricted services online. Te schedule
also suggests that China has a lot of satel-
lites already built and ready to fy soon.
Autonomous positioning accuracy for
the open service is expected to be at least
10 meters, according to Jing Guifei, chief
of the international cooperation division
in the National Remote Sensing Center of China (NRSCC). A
wide area diferential service providing one-meter real-time
positioning and a short message service (SMS) is also part of
the Compass program, Jing said.
As the newcomer to the GNSS feld, in the words of
Yin Jun, director of the European Afairs Division of Chinas
Ministry of Science and Technology (MOST), Compass is
not in the same place at the start of the race.
Yin stressed that GNSS should not be a competitive
exercise. We need coordination among system providers,
he said. Although a regional capability is expected once the
frst 10 Compass satellites are in place, Yin said a global Com-
pass service would not arrive until between 2015 and 2020.
GPS. As the leading GNSS provider, the United States might
be thought to have the luxury of improving on a real and exist-
ing system with 31 operational SVs on orbit. In fact, the GPS
program is in the midst of a full-blown modernization phase.
Launch of a modernized GPS Block IIR satellite SV
IIR-20(M) is scheduled for March 24, the frst since dis-
covery of a faulty component in the Delta 2 booster last June
led to a suspension of launches. A demonstration payload
for the new L5 civil signal is on the IIR-20(M), and faces an
August 2009 deadline to meet an International Telecom-
munications Union requirement for securing primary GPS
access to the frequency.
Te last IIR-M should go up in August, according to Col.
Dave Buckman, PNT command lead for Air Force Space
Command at Peterson Air Force Base, Colorado.
Launch of the frst Block IIF spacecraf is scheduled for
October 2009, although anomalies discovered in the signal
generator of the second IIF now under construction has
introduced some uncertainty into the plan.
within the 3.7 meters called for in the GLONASS Interface
Control Document (ICD) and several times better than the
UREs of just year ago. By the end of last year, GLONASS was
typically providing a standalone receiver with fve-meter
positioning accuracy using pseudoranges.
Launches have taken place regularly as scheduled over
the past few years, and another six satellites are set to go up
in triple launches in October and December this year. If suc-
cessful, that should bring the GLONASS constellation to full
operational capability (FOC) with 24 satellites early in 2010.
But thats not all. Te next-generation GLONASS-K will
begin launching next year and include a CDMA (code divi-
sion multiple access) signal on L3, which will more closely
align with other GNSS systems that the systems legacy fre-
quency division multiple access (FDMA) design.
A decision about new GLONASS signals at the L1C and
L5 frequencies depends on negotiations by a U.S./Russia
working group, but could lead to additional CDMA signals,
said Sergey Revnivykh, deputy director of the Russian space
agencys Mission Control Center.
Te stable progress in rebuilding and modernizing
GLONASS has even drawn interest from players in the
mobile phone industry. Nokia has been investigating the use
of GLONASS for its handsets. And, at the Munich summit,
Frank van Diggelen, technical director and chief navigation
ofcer for Broadcom Corporation, a semiconductor company
that targets mobile handset manufacturers, appeared to com-
pare GLONASSs prospects favorably to Galileo.
If GLONASS, which almost has a complete constellation,
fnds its way onto consumer devices, then consumers will have
access to 65 satellites (GPS 31 + SBAS 7 + QZSS 3 + GLONASS
24 = 65), van Diggelen said. Tis may be enough. In a wor-
risome aside for Europes system, which is counting on mobile
Paul Verhoef, head of EC Gaileo Unit, and Yin Jun, China Ministry of Science and Technology
What Race? continued on page 52
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Amazing
Grace
A Hot Startfor
a Young Engineer
from Chinas
Ice City
MELODY WARD LESLIE
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
I
magine that your only light source is a 50-watt bulb.
Visualize it shining at you from 20,000 kilometers
(about 12,000 miles) away.
Tats about how weak the signals are from
the new Galileo and Compass satellites, and thats why
Grace Xingxin Gaos accomplishments in being the frst
to derive the code generators for both systems are so
amazing.
Now that GPS is no longer the only game in space, its
crucial to manage the mushrooming number of signals
that will beam down on us from new GNSS systems.
Gaos work already has driven development of new
receivers capable of acquiring and tracking these signals.
In the future, each satellite will generate multiple signals,
creating possibilities for new applications and products
weve yet to imagine.
A research associate at the Stanford GPS Laboratory,
Gao received her doctorate from Stanford University last
September, just four years afer arriving in the United States
from China.
Te 28-year-old electrical engineer was a stranger to
GNSS when she came to Palo Alto in 2004. However, her
breakthroughs since being
introduced to the feld
by GPS icon Per Enge,
professor of astronautics and
aeronautics and director of
Stanfords GPS Research
Laboratory, already have had
such impact that she received
the Institute of Navigations
Early Achievement Award
for 2008.
Just one year earlier, as
an ION GNSS sponsored
student, her paper received
both the Best Presentation
Award and the Student Paper Award.
Enge said Gao is among the most impressive people ever
to survive Stanfords tough rite of passage known as the
Ph.D. Qualifying Exam in electrical engineering, an ordeal
involving 10 one-on-one interviews with faculty members in
a single day.
I was stunned by her cool and comprehensive answer
to my question, Enge recalled. I later learned she had been
in the U.S. for only four months. She fnished in the top few
percent of students. Fortunately, she subsequently asked to
join the GPS Laboratory, and her colleagues here and in the
larger navigation community are the richer for it.
The More Signals, the Merrier
Gao remembers being struck by the warmth of Enges
personality. She said that, when he invited her to visit the lab,
what she saw captured her curiosity, and GNSS became her
calling. Te next thing she knew, she was immersed in eforts
to understand the signals of the new GNSS systems, which
may contribute to a total of about 120 satellites in the future.
When there are more satellites it can be the more the
merrier, Gao said. On the other hand, the satellites share
or partially share the same frequency bands, so its also more
crowded.
Greater numbers of satellites increase redundancy,
which is a good thing because if one fails, you still have the
others. However, much as AM radio stations with closely
neighboring frequencies interfere with each other in the
wrong conditions, the new satellite signals will drown each
other out unless they are managed.
Te satellite-transmitted power is less than 50 watts and
the satellites are about 20,000 kilometers above the Earth,
Gao explained. We already use a high-gain antenna but
HUMAN
ENGINEERING
Compass Points
Engineering Specialties
Signal processing
Her Compass Points
My parents in Harbin and my husband, David Varodayan
My mentors and colleagues in the Stanford GPS Lab and supportive
friends of all kinds
Stanford University, a place full of newideas and excitement
Favorite Equations
Fourier Transform:
f(x): time domain function; : frequency domain function
Inverse Fourier Transform:
The beauty of the Fourier transformand the inverse Fourier transform
is that they connect the time-domain representation of a signal with its
frequency domain representation, allowing us to viewone signal fromtwo
different perspectives.
Year She Began Working in GNSS
2005 Her teamstruggled to make a high gain GPS antenna in a lab class:
We made a lot of antennas of different types. None of themworked! I
learned to appreciate failures and to growfromthem.
Fell In Love With GNSS. . .
. . . when she started decoding GIOVE-A signals with Professor Per Enge at
Stanford.The signal structure was unknown and the received signal power
was weak. . . it was a great challenge.
Events That Most Signify GNSS Has Arrived
Launches of GIOVE-A, the rst Galileo test satellite, in 2005 and Chinas rst
Compass MEOsatellite in 2007.
Memorable Mentor
Professor Hu Zhiming, her calculus teacher at Chinas Tsinghua University:
He really cared about students. It was a big class, about 200 students. He
remembered every students name and IDnumber, and even the mistakes
the students made in their homework. He showed me the beauty of math
and the various applications in engineering. I learned about Fourier
transformfor the rst time in his class.
Inuences of Engineering Outside Work
I enjoy xing things at home, such as my vacuumcleaner.
Favorite Non-GNSS Activities
Tennis, snowboarding, traveling, and beading.
Special Interest
Gao seeks to encourage girls to pursue engineering by hosting workshops
on GPS in the Sally Ride Science Festival at NASA Ames where she uses
games to convey basic concepts such as trilateration. I feel sad that some
talented girls hide their math skills and give up on science/engineering,
because its considered uncool or not-for-girls.
Grace Gaos web page:
http://stanford.edu/~gracegao
BBC video of the Ice Disneyland in Harbin, China Gaos hometown:
<http://news.bbc.co.uk/1/hi/world/7813475.stm>
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
20
HUMAN ENGINEERING
the signal is still buried in noise. Te signal structure is so
complicated. For example, the satellites are moving so theres
the Doppler efect, and both the satellite and the receiver
clocks may drif.
Separating the code from all that noise required every
bit of the competitive intensity that helped Gao excel
academically in China and the United States. However,
she says the collaborative culture of the GNSS research
community has become far more important to her than the
drive to be frst.
To do anything, even just pure research, I need a lot
more skills than just getting a good score or being smart,
she said. I need help. I need people I can discuss the problem
with.
Three Code Breakthroughs in Three Years
Gao uncorked the code for Europes frst prototype satellite
for Galileo, named GIOVE-A, within weeks afer its signals
became active in January 2006. She derived the underlying
algorithms that generate the codes so that she could decode
its signals on all three frequency bands.
I was very excited, she recalled. I stayed up every night.
I even dreamed about the code, and I would try something
else when I woke up. Im very grateful that I had wonderful
mentors and colleagues to help me.
Te following year, in April 2007, China launched the frst
medium earth orbiting satellite belonging to their Compass
system.
Using her new techniques, Gao blitzed through
demodulation of the Compass-M1s civil codes in all three
frequency bands, proved that all Compass-M1 codes are
Gold codes, and derived their code generators as linear shif
feedback registers. Ten she applied the codes in a sofware
receiver for the Compass-M1.
Last year in April, when the Europeans launched GIOVE-
B, a pre-production prototype satellite for Galileo, Gao again
was the frst to discover the underlying code generators.
Growing Up in Ice City
Gao grew up in Harbin, an ancient Chinese city of about ten
million located within 300 kilometers of Russia and nearly
as close to Korea and Japan as well. Known as Ice City and
likened both to Moscow and Paris, Harbin ranks among the
worlds foremost centers of snow and ice culture.
Gaos coordinates:
37.428178, -122.172173
Gao visits the
Kennedy Space Center
With her
parents at the
Ice Pagoda in
Harbin, China.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
With Chinas best ski resort and a host of sparkling new
venues built to Olympic specifcations, Harbin has designs
on the 2018 Winter Olympics. Tis years 25th annual Harbin
Ice and Snow Sculpture Festival featured a life-size frozen
replica of Disneyland plus a 525-foot-wide sculpture of Santa
Clauss head. By comparison, the faces of U.S. presidents
carved into Mt. Rushmore are only 60 feet high.
As a child, Gao was inspired by some unusual feats of
engineering during winters with temperatures as frigid as
minus 35 degrees Celsius (-31F). She remembers making her
own ice creations while the adults cut enormous ice blocks
from the river in the center of Harbin. Tey created fanciful
structures four and fve stories high, wired for electricity, and
unbelievably colorful at night.
A Fathers Dream
Gaos father had dreamed of becoming an engineer from
boyhood, a dream crushed by the social and political
upheavals of Chinas Cultural Revolution in the mid-1960s
and 1970s. He was in middle school when all schools were
closed and students were sent to work on farms.
Although her father didnt pressure her to pursue the
career that he had missed out on, Gao says he encouraged
her interest in engineering at every turn. And it is normal for
Chinese girls to excel in math and science. In China, when I
grew up, smart kids preferred engineering, she said.
When she began her undergraduate studies at Beijings
Tsinghua University Chinas MIT Gao decided to
choose an English name because her American friends had
difculty saying her Chinese name, Xingxin (pronounced
shing-sheen). I chose Grace because of its meaning, she
explained.
In Chinese culture, people care mainly about what
names stand for, Gao adds. Te Chinese characters for
her name are . Te frst character, , is the
family name Gao, which means tall or high. Te last two
characters are her given name. (Xing) means awake or
sober and (Xin) means heart.
Afer earning her bachelors and masters degreesand
numerous awards, including recognition as the Exceptional
Graduate of Tsinghua and Exceptional Graduate of Beijing
Gao joined IBMs China Research Laboratory as a research
engineer.
California Dreaming
I realized I really like doing research, she said. I came to
the United States because I wanted to see the world and to
know diferent cultures. She had many ofers, but chose
Stanford because of its reputation, a terrifc scholarship, and
Californias balmy climate and cultural diversity.
Tree years ago she met her husband, David Varodayan,
a fellow Ph.D. candidate in electrical engineering, while
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_____________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
________
___________
_____________
_____________
_____________
______________________________
accompanying mutual friends on a road trip to Oregon.
As they took in the sights from Californias Mt. Lassen to
Oregons Crater Lake, she learned they had taken some of the
same classes. Teyd even been assigned to the same small
study group, but hadnt met in person because somehow both
never showed up in person at the same time.
We complement each other, Gao says of the
relationship, At the same time, we share many common
interests. Te kitchen is our favorite place to discuss
research.
While Varodayan completes his doctorate, Gao is keeping
her options open and looking forward to starting a family.
Meanwhile, she continues to collaborate with colleagues
around the world working on the GPS Evolutionary
Architecture Study. Her projects involve analyzing the
overall performance and the potential for interference among
multiple GNSS constellations. She also is helping build a
system to monitor GNSS in real time to detect issues such as
power failure and changes in signal transmission.
Tere are always new problems and challenges, she said.
Editors note: Tree articles refecting Grace Gaos work in
decoding GNSS signals have appeared in Inside GNSS with her
as a coauthor: GNSS Album: Images and Spectral Signatures
of the New GNSS Signals, May/June 2006, GNSS over China:
Te Compass MEO Satellite Codes, July/August 2007 and
GIOVE-B on the Air: Understanding Galileos New Signals,
May/June 2008.
HumanEngineeringis a regular feature that highlights some of the
personalities behind the technologies, products, and programs of the GNSS
community. We welcome readers recommendations for future proles.
Contact Glen Gibbons, glen@insidegnss.com.
Grace and her husband,
David Varodayan
HUMAN ENGINEERING
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
____________
_________
________
24
InsideGNSS MA RC H/A P RI L 200 9 www.insidegnss
GNSS
Solutions:
What are the
challenges
associatedwith
GLONASS(FDMA)
ambiguityresolution
andhoware they
addressed?
W
hile the GNSS industry
eagerly awaits the introduc-
tion of new global naviga-
tion satellite systems such
as Galileo and Compass, the Russian
Federation has been steadily modern-
izing its GLONASS system. At the time
of writing, 20 GLONASS satellites are
in orbit. With launches of six more
satellites scheduled this year, a fully
operational constellation of 24 satellites
could be realized by early 2010.
GLONASS, like GPS, broadcasts
carrier phase signals in the L1 and L2
frequency bands. Each GLONASS sat-
ellite broadcasts its signals on slightly
diferent frequencies within a given
L-band using a technique known as
frequency division multiple access
(FDMA). Conversely, the GPS signal
structure is based on code division
multiple access (CDMA) whereby all
satellites transmit on the same L1 and
L2 carrier frequencies. Each GLONASS
satellite can be identifed by its signal
frequency while individual GPS satel-
lites are distinguished by unique pseu-
dorandom noise (PRN) codes trans-
mitted with the navigation message.
Reliable resolution of the integer
ambiguities inherent in the carrier
phase measurements is the key to using
GPS and GLONASS for high-precision
(centimeter level) positioning applica-
tions, especially real-time applications.
However, the GLONASS ambiguity
resolution process is more compli-
cated compared to GPS because of the
FDMA signal structure. Te remain-
der of this discussion focuses on the
challenges associated with GLONASS
ambiguity resolution and how these
challenges are addressed.
GPS ambiguities related to double-
diference carrier phase observations
are usually resolved in GPS data pro-
cessing schemes. Te double-diference
technique efectively mitigates com-
mon errors introduced by the receiver
and satellite hardware, including the
receiver and satellite clocks, as well as
the Earths atmosphere. Double-dif-
ference observations can be formed by
subtracting two inter-station single-
diference observations.
Te inter-station single-diference is
derived by subtracting measurements
to the same satellite observed simulta-
neously at two stations. For example,
the single-diference and double-dif-
ference involving reference station m,
rover station k and satellites p and q are
pictured in Figure 1.
Te ambiguity parameters related
to GPS (CDMA) L1 or L2 double-dif-
ference observations can be written as:
Te symbol denotes the wave-
length of the carrier signal, which is
inversely proportional to its frequency.
Because all GPS satellites transmit on
the same frequencies, all signals in a
given L-band will have the same wave-
length.
Te double-diference integer ambi-
guity ( ) on the lef hand side of (1)
is constructed from the inter-station
single-diference ambiguities associ-
ated with satellites p ( ) and q ( ),
where p is a reference satellite common
to all other double-diference observa-
tions. Te double-diference ambigui-
ties are typically resolved or fxed using
integer least squares techniques such as
LAMBDA.
Double diferencing can also be
GLONASS
inter-
frequency
biases and
ambiguity
resolution
GNSS Solutions is a
regular column featuring
questions and answers
about technical aspects of
GNSS. Readers are invited
to send their questions to
the columnist, Dr. Mark
Petovello, Department of
Geomatics Engineering,
University of Calgary, who
will nd experts to answer
them. His e-mail address
can be found with
his biography at the
conclusion of the column.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
used when processing GLONASS observations. Te receiver
and satellite clock errors will also cancel provided the obser-
vations are in units of meters. However, the wavelengths of
GLONASS signals are not common for all satellites within
a given frequency band. Equation (2) can be generalized for
FDMAsignals by introducing wavelength identifers for sat-
ellites p and q such that:
In addition to the double-diference ambiguity,
GLONASS double-diference observations also consist of the
single-diference ambiguity related to the reference satellite p
scaled by the wavelength diference of the two signals
pq
.
We cannot simply estimate the single-diference refer-
ence ambiguity along with the double-diference ambiguity
using only carrier phase measurements. Tis approach would
lead to a singular solution with more unknowns than obser-
vations. Instead, these two ambiguities could be lumped
together in a modifed ambiguity term. However, the modi-
fed ambiguity would no longer be an integer and, hence,
could not be fxed.
In practice, the single-diference reference ambiguity is
ofen estimated with the aid of pseudorange observations.
For example, the reference ambiguity can be estimated using
Equation (3):
Te symbols P and represent the single-diference
pseudorange and phase observations, respectively. Once the
single-diference reference ambiguity has been determined,
Inter-station
single-difference
Double-difference
p
p
q
m k m k
FIGURE 1 The double-difference is the difference between two single-
difference observations. Double-differencing is used extensively in
data processing to effectively mitigate many of the common errors
affecting GNSS observations.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_______________________
_________________________________
___________________
26
we can resolve the double-diference
ambiguities using integer least squares
in much the same way as for GPS.
However, other factors must be
taken into consideration. Equation (3)
is infuenced by pseudorange random
measurement error, multipath, iono-
spheric errors and receiver dependent
biases. Tese residual errors make it
difcult to resolve the reference ambi-
guity to its correct integer value.
Equation (2) clearly indicates that
any error in the reference ambiguity
will be absorbed by the double-difer-
ence ambiguity as a function of the
delta wavelength term. For example,
approximately 4.6 millimeters of error
will be introduced for every meter
of error in the reference ambiguity
if the maximum diference between
GLONASS wavelengths is involved.
Reliable ambiguity resolution could
be compromised if this induced bias
becomes signifcant. However, we can
usually estimate the single-diference
reference ambiguity with sufcient
accuracy for reliable ambiguity resolu-
tion if the same brand of receiver is
used in the double-diference. How-
ever, residual receiver-dependent biases
may cause a signifcant bias in the
single-diference reference ambiguity
estimate when diferent receiver types
are involved.
In general, the hardware and signal
processing architecture of a receiver
will introduce frequency-dependent
biases in the pseudorange and carrier
phase measurements. Tese variations
in the measurements, which may be
diferent for code and phase, are com-
monly known as inter-frequency biases.
Te pseudorange and carrier phase
receiver biases for all GPS signals will
be the same for a given L-band since
all satellites transmit on the same
L1 and L2 frequencies. Tese biases
efectively cancel in double-diference
observations, even if diferent types of
reference and rover receivers are used.
Conversely, all GLONASS satellites
transmit on diferent frequencies; so,
these biases may even be diferent for
signals in the same L-band.
Te tracking channels in state-of-
the-art GLONASS receivers may be
calibrated to minimize the magnitude
of these inter-frequency biases in the
L1 and L2 observations. Furthermore,
receiver-dependent biases are generally
consistent for instruments developed
by the same manufacturer. As a result,
receiver biases are efectively mini-
mized in single and double-diference
observations and will not afect ambi-
guity resolution if the same brand ref-
erence and rover receivers are involved.
However, GLONASS double-difer-
ence phase observations for heteroge-
neous receiver pairs can be infuenced
by signifcant inter-frequency biases.
Figure 2 shows the GLONASS L2
double-diference carrier phase residu-
als for a mixed pair of state-of-the-art
geodetic-quality receivers separated by
approximately two meters. Estimates
of the receiver-satellite geometry as
well as the single and double-diference
ambiguity terms were removed from
the raw double-diference observations
to form the residuals.
Te expected value of the double-
diference residuals for such a short
baseline is approximately zero. Howev-
er, the residuals in Figure 2 are afected
by signifcant inter-frequency biases
approaching 0.5 cycles (~12 cm).
Tese biases may be attributed to
real variations in GLONASS phase
measurements introduced by the
FDMA receiver architecture and
apparent inter-frequency biases
induced by an incorrect estimate of the
single-diference reference ambiguity.
Remember that any error in the refer-
ence ambiguity will be common to all
double-diferences; however, the error
will manifest as inter-frequency biases
because of the frequency-dependent
delta wavelength term in Equation (2).
Te data presented in Figure 2
represents a typical example of the
inter-frequency biases observed
when the double-diference baseline
consists of diferent receiver types.
Tese biases need to be estimated and
removed from the observations in
order to resolve the double-diference
GLONASS ambiguities reliably.
However, the task of estimat-
ing inter-frequency biases for mixed
receiver types is not a trivial task,
particularly in real-time. In practice,
separating inter-frequency biases from
other error sources and the ambiguity
terms is difcult. For this reason, many
sofware solutions do not attempt to
fx GLONASS ambiguities if diferent
receiver brands are involved. Instead,
the inter-frequency biases are absorbed
by the real-valued or foat estimates of
the GLONASS ambiguities.
Tis approach generally improves
the GPS solution, especially ambiguity
resolution performance. However, the
full potential of GLONASS will only be
realized if the GLONASS ambiguities
are fxed to integers.
GNSS SOLUTIONS
FIGURE 2 The GLONASS double-difference carrier phase observations associated with this pair of
mixed receiver types are affected by signicant inter-frequency biases that may lead to unreli-
able ambiguity resolution if not handled appropriately.
GLONASS L2 Double Difference Carrier Phase Residuals
Epoch
R15-R14
R15-R24
R15-R08
0.15
0.10
0.05
0.00
-0.05
-0.10
-0.15
R
e
s
i
d
u
a
l
(
m
)
472500 472700 472900 473100 473300 473500
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
__________________________ ______________________
28
Rather than estimating inter-fre-
quency biases together with other
unknowns such as ambiguities, the
inter-frequency biases for various
receiver combinations can be pre-cali-
brated on a so-called zero baseline or
very short baseline in order to control
various error sources including atmo-
spheric errors and multipath. Tis
approach to estimate inter-frequency
biases is possible because the receiver
biases are generally stable over time.
Unlike antenna phase center varia-
tion (PCV) calibrations, no accepted
standard calibration tables for receiver
dependent biases currently exist in the
public domain. Terefore, manufactur-
ers normally calibrate diferent receiver
brands against their own reference
using proprietary techniques. Figure 3
shows the results of a proprietary solu-
tion implemented in real-time sofware
developed by Leica Geosystems to mit-
igate the inter-frequency biases evident
in Figure 2.
Te inter-frequency biases visible in
Figure 2 that might prevent ambiguity
resolution are no longer present in the
data. Tis result represents a signifcant
step towards interoperability among
diferent brands of receivers.
However, the calibration approach
does have some limitations. First, the
processing sofware needs to know
which types of receivers are involved in
order to apply the appropriate calibra-
tion values. Tis is not an issue if open
data exchange formats such as RTCM
SC-104 v3 (real-time format) and
RINEX v2 and v3 (post-processing for-
mat) are used, because they contain the
relevant receiver type information.
Secondly, manufacturers need to
ensure that the unit-to-unit receiver
biases for a given type of instrument
are consistent. Tis issue is currently
being addressed within the RTCM
SC-104 committee, which draws repre-
sentatives from government, academia,
and industry, including most receiver
manufacturers.
Finally, the calibrated double-dif-
ference observations may still be afect-
ed by residual inter-frequency biases
caused by unit-to-unit receiver varia-
tions, temperature diferences, and
aging hardware components. However,
the magnitudes of these residual biases
are typically only a small fraction of
the signal wavelength. Terefore, they
do not present a signifcant issue for
ambiguity resolution.
In summary, GLONASS (FDMA)
double-diference carrier phase obser-
vations consist of a single-diference
reference ambiguity term in addi-
tion to the usual double-diference
ambiguity. Furthermore, GLONASS
observations may be afected by inter-
frequency biases. Tese two issues
are not normally associated with GPS
(CDMA) observations and will hinder
or even prevent reliable ambiguity
resolution if not handled correctly.
Te Russian Federation plans to
add CDMA signals to the GLONASS
signal structure as part of its modern-
ization program, beginning with the
L3 frequency. However, Russian Space
Agency ofcials have indicated that
FDMA signals will be retained for the
foreseeable future in order to ensure
backward compatibility with FDMA-
only receivers.
Terefore, the issues that afect
GLONASS ambiguity resolution today
will also be relevant in the future.
Resolving the issues in practice is chal-
lenging, especially in real-time and
when diferent receiver brands are
involved.
Nevertheless, these issues are
manageable and once they have been
addressed, the addition of GLONASS
observations in a high-precision GNSS
solution can certainly improve posi-
tioning performance compared to GPS
alone.
FRANK TAKAC
Frank Takac received a
bachelor and master of
applied science from
RMIT University in
Melbourne, Australia.
He joined Leica
Geosystems in 2002
and has been involved in the development of
Leicas high-precision GNSS product range.
Frank currently leads the GNSS positioning
algorithms group responsible for real-time and
infrastructure software.
Mark Petovello is an Assistant
Professor in the Department
of Geomatics Engineering at
the University of Calgary. He
has been actively involved in
many aspects of positioning and
navigation since 1997 including
GNSS algorithm development,
inertial navigation, sensor
integration, and software
development.
Email: mark.petovello@
ucalgary.ca
GNSS SOLUTIONS
Calibrated GLONASS L2 Double Difference Carrier Phase Residuals
Epoch
R15-R14
R15-R24
R15-R08
0.15
0.10
0.05
0.00
-0.05
-0.10
-0.15
R
e
s
i
d
u
a
l
(
m
)
472500 472700 472900 473100 473300 473500
FIGURE 3 The inter-frequency biases affecting the GLONASS double-difference phase observations
in Figure 2 have been effectively mitigated. Consequently, the double-difference ambiguities can
then be xed reliably using integer least squares techniques.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
______
_________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
________________
30
On the Trails
of the Inuit
I
n Inuktitut, the Inuit language, Igliniit refers to rou-
tinely traveled trails. Countless trails are known and
used by Inuit in Canadas far northern territory of Nuna-
vut, a vast area the size of Western Europe flled with sea
ice, fords, and mountains.
Nunavut also has abundant wildlife and is home to
approximately 30,000 people, mostly Inuit, who live in 26
communities distributed around Nunavuts nearly two mil-
lion square kilometers (746,000 square miles) of land. All of
the communities, except one (Baker Lake), are coastal.
Since 2006 environmental researchers, geomatics engi-
neering students, and Inuit from Clyde River (Kangiq-
tugaapik) a community (population ~820) on the far
northeastern coast of Bafn Island (Figure 1) have worked
together in the Igliniit Project. Among their activities is the
In the Arctic, learning about the
location, conditions, and uses of trails
over time and space reveals a great deal
about the environment and relationship
of the people to their land and ice. In
this way, the Igliniit (Trails) Project
is making change more visible with a
new integrated GPS
system for expert
Inuit hunters, who log
thousands of kilometers
each year on dogsleds
and snowmobiles.
KYLE OKEEFE
THE UNIVERSITY OF CALGARY SCHULICH SCHOOL OF ENGINEERING,
CALGARY, ALBERTA, CANADA
SHARI GEARHEARD
NATIONAL SNOW AND ICE DATA CENTER, UNIVERSITY OF
COLORADO AT BOULDER, CLYDE RIVER, NUNAVUT, CANADA
GARY AIPELLEE, APIUSIE APAK, JAYKO ENUARAQ, DAVID
IQAQRIALU, LAIMIKIE PALLUQ, JACOPIE PANIPAK, AMOSIE SIVUGAT
CLYDE RIVER, NUNAVUT, CANADA
DESMOND CHIU, BRANDON CULLING, SHELDON LAM, JOSIAH
LAU, ANDREWLEVSON, TINA MOSSTAJIRI, JEREMY PARK, TREVOR
PHILLIPS, MICHAEL BRAND, RYAN ENNS, AND EDWARD WINGATE
THE UNIVERSITY OF CALGARY SCHULICH SCHOOL OF ENGINEERING
On the Trails
of the Inuit
Adapting GPS for
Nunavut Hunters and
Environmental Research
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
the technical problems and breakthroughs, the collaborative
model and not least what happened when two markedly
diferent communities discovered each other.
Building a Team
Te idea for the Igliniit Project developed in Clyde River,
where Dr. Shari Gearheard, a geographer and research scien-
tist with the National Snow and Ice Data Center at the Uni-
versity of Colorado at Boulder, has worked with Inuit hunters
and elders documenting environmental change since 1999.
Gearheard, who moved with her husband to Clyde River
in 2004, and some of her Inuit research partners came up
with the idea for creating a system that allowed hunters to
log their observations of the environment as they traveled,
so that the information could be mapped easily and immedi-
ately. Teir fndings could then be discussed and shared afer
the hunters returned from long (or short) journeys.
Te Igliniit Project came to life when it joined several
other northern projects to create ISIUOP, the Inuit Sea Ice
Use and Occupancy Project. Led by Dr. Claudio Aporta at
Carleton University in Ottawa, ISIUOP was successful in
obtaining a grant from the Government of Canada Interna-
tional Polar Year program. With a focus on sea ice, the Igli-
niit Project was now ready to test its concept.
Gearheard contacted Dr. Kyle OKeefe at the University
of Calgary Geomatics Engineering Department for support
with the technical side of the project. OKeefe took on the
challenge and used the opportunity to create a unique case
study for the departments undergraduate senior-year design
course.
Tree diferent student groups over three diferent years
of the course selected the equipment and worked on system
integration and programming. Te students built the system
based on design requirements, ideas, and feedback from the
hunters in Clyde River, a back and forth exchange that was
facilitated by Gary Aipellee (lgliniit Project manager and
translator) and Gearheard.
Te hunters sketched ideas and concepts, asked ques-
tions, and provided critiques, while the students in turn
asked their own questions and modifed the system to
accommodate the advice and requests from the new tools
intended users.
Afer a year and a half of development, the new system
was introduced in January 2008 during the International
Polar Year and tested by four hunters in Clyde River during
the rest of the 2008 sea ice season (until early July). In the
summer and fall of 2008 the students made further modifca-
tions to the units based on feedback from the hunters. Ten
the modifed units were brought back to the community in
January of 2009 for another year of testing, with two more
hunters joining the team.
The Igliniit Tool
Te Igliniit system is afordable and simple to mount. It is
light and easy to carry and easy to operate.
design, production, and testing of a new tool for Inuit hunters
one that combines the virtues of a personal digital assis-
tant (PDA), weather station, camera, and a handheld GPS
navigator.
Designed to be easily and afordably mounted on snow-
mobiles or dog sleds the regular modes of travel used by
Inuit hunters the integrated device does not attempt to
replace traditional navigation or land skills. Inuit knowledge
and skills related to navigation are highly complex and have
been passed down over generations.
Te Igliniit tracking system is merely a tool, one that
can complement and be combined with Inuit knowledge, as
other technologies such as snowmobiles and portable navi-
gation devices (PNDs) already have been.
Tis article describes the frst three years of the project,
FIGURE 1 Clyde River, Nunavut, and Calgary, Alberta, Canada
(courtesy of Google Earth)
P
h
o
t
o
:

E
d
w
a
r
d

W
i
n
g
a
t
e
Inuit hunter David Iqaqrialu of Clyde River
logs observations on his Igliniit PDA/GPS as he
travels the sea ice. Mounted on the back of his
snowmobile is a weather station, part of the
Igliniit system.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
32
Te tool consists of a GPS-equipped PDA and a weather
station that records temperature, humidity, and pressure
and automatically logs the location of the vehicle every 30
seconds. Te resulting georeferenced waypoints can later be
mapped to produce the travelers routes along with recorded
weather information.
In addition, the systems customized PDA screen has a
user-friendly touch interface in English and Inuktitut. Te
interface allows a hunter to press one or multiple icons that
represent observations made at a certain point. Tese obser-
vations/icons were determined by the hunters themselves,
and programmed by the students.
For example, the unit displays screens with icons that
represent animals, such as polar bear, fox, caribou, ptarmi-
gan, and so forth, as well as sea ice features including cracks,
rough ice, icebergs, and more (see Figure 2). Other features
allow a hunter to distinguish if an observation is a hazard
(for example, thin ice or certain types of cracks).
A keyboard can be displayed to enable a hunter to note
something that might not be part of the icon list. When the
hunter makes an observation, it is logged in the system to be
mapped later. A digital camera carried by the hunter pro-
vides georeferenced images that add to the documentation of
the journey.
In the projects current phase, the group is working on
generating maps of the data and toward automating the proj-
ect. Tis aspect of the work is led by Carleton University.
As maps from the Igliniit system are overlaid and accu-
mulated over time and space, the team hopes to achieve a
valuable quantitative and qualitative picture of the land, sea
ice, and how Inuit use both.
Quite apart from the technical aspects discussed in this article, the Igliniit
project has provided a rare opportunity for undergraduate students from
Calgary to explore and learn about a very distinct region of their own coun-
try.
The communities of Clyde River, Nunavut and Calgary, Alberta could not
be more different. Calgary is a typical North American city and Canadas fth
largest metropolitan area with a population of more than one million. It is
a booming commercial center where a number of technology, oil and gas,
manufacturing, banking and insurance companies are headquartered.
Many thousands of miles away, Clyde River is a small arctic hamlet
FIGURE 2 Igliniit user interface
showing the rst of ve pages of
icons.
FIGURE 3 Igliniit interface showing
the sixth page as displayed in Inuk-
titut. Pressing the English toggle
button switches the interface back
into English language mode
From Calgary to Clyde River
ON THE INUIT TRAILS
Above: Laimikie Palluq and
Calgary student Ryan Enns take
the Igliniit system for a test spin
on Palluqs dog team in January
2009. Note the weather station
mounted on an old hockey stick
at the back of the sled. Left:
Inuit elder and hunter Jacopie
Panipak gets to know the Igliniit
system in 2008. Mounting
hardware is in the foreground.
P
h
o
t
o
s
:

S
h
a
r
i

G
e
a
r
h
e
a
r
d
situated across Bafn Bay from Greenland (which has a higher population
density than the 10-year old Canadian territory of Nunavut.) Nearly all of
Clyde Rivers 820 citizens are Inuit, and subsistence hunting is a major part
of their livelihood.
The community has only a handful of cars and both visiting student
teams have been fascinated by the use of dogsleds and the dogs diet of
seal meat. The human residents depend on the land and sea as well. Clyde
River is supplied only by air and an annual sea lift, and meat from seals,
caribou, polar bear, and sh provides a major source of healthy food and
makes up a signicant portion of the local diet.
The University of Calgary undergraduates involved in this project have
been intrigued by the differences, but also surprised by the similarities. For
example, Clyde River youth are just as skilled as any kid in Calgary when it
comes to Internet, iPods, and video games. In fact most of the children of
the Inuit hunters help their fathers with technical troubleshooting, and were
the rst to load music into the systems.
The Igliniit system PDAs have been very popular with the hunters, not
just for their data collection capabilities but for other features, includ-
ing games and music. For example, some Inuit hunters helped build their
touch-screen skills by playing solitaire on the Igliniit units.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Combined with other methods of research, such as in-
depth interviews with elders and other local experts, Igliniit
data help add to the understanding of human-environment
relationships. Te information has application not only for
sea ice studies, but also for potential use in wildlife studies,
environmental monitoring, land use planning, and search
and rescue activities.
Off the Shelf
In September 2006, before the project was even funded, the
frst group of four students began their consultations with
Clyde River hunters to identify user requirements.
Te Inuit wanted to log positions and weather condi-
tions continuously including temperature, pressure, and
humidity. Tey also wanted to make and log their own obser-
vations of the environment. And the system had to work in
very cold weather -40C/F and below.
Te unit also had to be simple enough for almost every-
one in the community to use; so, the system interface had
to accommodate the special syllabic script of the Inuktitut
language (Figure 3), in addition to the roman orthography
used in English.
Te students selected and tested hardware for the data
collection system. Most fourth-year geomatics engineering
undergrads do not have electrical engineering or hardware
design experience; so, the project supervisors decided to limit
the project to commercial of-the-shelf products, which they
hoped would also keep unit costs down.
With these requirements in mind, several initial designs
involving handheld GPS receivers, personal digital assis-
tants, and portable weather stations were considered. But
most of these couldnt make it in the bitter weather of the sea
ice season.
Te custom interface and pressure, humidity, and tem-
perature requirements eliminated several commercially
available handheld GPS receivers with integrated weather
sensors.
Eventually, the students converged on a ruggedized PDA
as the most likely candidate for a customizable data-logger.
Te unit chosen has a minimum operating temperature spec-
ifcation of -30C (-22F) not perfect, but the closest they
could get to the operating temperature requirement given
the project constraints. Tey also selected the 12-channel, L1
C/A-code CompactFlash cardbased GPS receiver that is the
PDAs standard companion. Te unit provides postprocessed
accuracy of two to fve meters.
A suitable weather instrument was not identifed until
early in 2007. Again, due to budget constraints, the COTS
requirement, and the need for full operation at -40 degrees,
the student team identifed and eliminated several possible
weather stations before fnding a satisfactory instrument.
Te pocket weather meter that they selected was chosen
for its cost, environmental specifcations, and the availability
of a serial interface. Although the interface did not support
real-time data streaming as a standard feature, the manufac-
turer provided the team with documentation that enabled the
students to implement real-time data logging using several
unsupported features of the interface.
Inuit-Guided Software
In 2007, a second undergraduate project team developed
most of the Igliniit sofware in collaboration with the Clyde
River team. Te hunters, Aipellee, and Gearheard met regu-
larly to defne the user interface and sent their suggestions to
the students via Gearheard through email until they agreed
on a fnal interface design.
Internally, the sofware consists of two serial port readers
that communicate with the external weather sensor and the
GPS receiver card (which appears to the operating system as
a serial port). Te students wrote the sofware in C++ lan-
guage using the open-source wxWidgets library to develop
the graphical user interface (GUI).
Te user sees a single window with several buttons that
are always available (located at the bottom of the screen, Fig-
ures 2 and 3) plus a horizontally scrollable area that contains
fve pages of toggle buttons and one page of display boxes.
Each button represents either an action or a type of obser-
vation, as shown in Figure 2. Te user selects any combina-
Gary Aipellee leads the Clyde River group through discussions that
provide input to the students for interface development
P
h
o
t
o
:

S
h
a
r
i

G
e
a
r
h
e
a
r
d
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
34
ON THE INUIT TRAILS
tion of observation buttons and then presses save to record
the time and position associated with each observation. Some
important observations including catch, photo, danger
and overnight are always visible, and the rest are available
on one of the scrollable screens.
So, for example, a hunter would press fsh if he observed
a fsh. Te hunter would press fsh then catch if he was
successful fshing.
Te sixth screen (Figure 3) contains display boxes for cur-
rent position, speed, bearing, time, and weather information.
An additional button toggles the language of the applica-
tion between English and Inuktitut. Te sofware logs user
observations whenever the save button is pushed and keeps
an automatic record of position and weather information in
order to map the route traveled.
And thats not all hunters can also play games and
music included on the Igliniit system and they do. Te
extra features prove handy for something to do while waiting
out the weather at camp or keeping the kids busy at the fam-
ily cabin. (See the sidebar, From Calgary to Clyde River.)
Field Testing (with Plastic Food Wrap?!?)
For the Igliniit systems initial deployment in January 2008,
two students traveled to Clyde River with four complete data
collection systems. Preparing for this trip was a project in
itself the isolated community cant be reached by road; so,
food and other supplies are prohibitively expensive to buy
locally, and everything that could possibly be required for the
deployment had to be brought by the students. (Te village is
supplied only by air and a once-a-year sea-lif.)
In addition to the PDAs and weather instruments, the
students carried mounting hardware for the handheld com-
puters and weather station serial interface cradles.
Fortunately, the weather station interfaces included a
threaded mount for use with a tripod making it possible to
use commercially available camera mounts.
Unfortunately, the otherwise rugged interface cradle
couldnt handle the wet spring sea-ice conditions. Te solu-
tion? Students wrapped the interfaces with a self-adhesive
plastic food wrap.
Before the systems were installed on the snowmobiles, the
hunters and students spent the frst day getting to know each
other, practicing the application, and becoming familiar with
the PDA operating system.
Ten they were ready to go. Te hunters, most of whom
are also master mechanics (out of necessity), installed the
systems on four snowmobiles and took them outside for a test
drive in favorable weather.
Ironically, temperatures during the frst week were
unusually warm, around -15C (5F) warmer than Calgary
during that time.
Figure 4 shows the route of the frst test using Google
Earth. An accompanying photo shows where the snow
machines turned around at the easternmost point of the tra-
jectory. Two of the four systems worked well during this test,
one failed due to a loose battery, and the other due to a loose
connection in the GPS receiver CF-slot.
Working Out the Bugs
Afer the two students returned to Calgary, the hunters con-
tinued using the equipment for the remainder of the season.
A minor problem occurred almost immediately. During the
installation, one of the PDA power supplies was damaged,
and it took several weeks for the replacement part to arrive
by mail.
By the following week, the weather had returned to
normal and the PDAs began to freeze. Tough the PDA is
specifed to operate at -30C, the screens froze at around -
27C. Te team tried to solve the problem afer discovering
an Internet article posted by someone in Iqaluit, Nunavuts
capital, who was troubleshooting a frozen GPS unit. Tey
suggested placing an electric snowmobile handlebar heater
between the GPS receiver and its mounting cradle.
Once again, the team experienced the difculties of iso-
lation: parts had to be ordered from Calgary and mailed to
Clyde River. By the time the heaters arrived, the weather had
warmed again, and several of the hunters were uncertain
about using the heaters because of concerns that doing so
might drain the snowmobile battery or damage the electrical
systems.
FIGURE 4 Trajectory recorded in rst eld test plotted in Google Earth.
The dark dot at the eastern most point is an island called Umiujaq (see
photo), a local landmark.
Taking a break on the sea ice near Umiujaq, at the turnaround point
at the easternmost point in the trajectory shown in Figure 4.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Initial data sets contained some useful information, but
user observations were missing sporadically from the log. An
unknown bug in the sofware seemed to cause the discrete
data (entered by the user through the interface) to not be
logged all the time. Unfortunately, this sofware bug couldnt
be reproduced in Calgary and, as a result, went unfxed.
Afer about one month of deployment, the team encoun-
tered another problem. Some of the power and data cables
were not weathering very well and cracking in the cold tem-
peratures. In one case, a hunters dog chewed a cable. Again,
replacements had to be sent by mail.
By May 2008, afer a season of sea ice travel with constant
bumps in rough terrain, the mounting hardware started to
break. Large bumps caused some wire connections to fail and
GPS cards to dislodge. During consultations the hunters had
warned the team about how rough their travel can be, and,
indeed, the pounding took a toll on equipment.
Unfortunately the warmer, wetter conditions proved too
much for the plastic food wrap. Te improvised weather sta-
tion interface waterproofng no longer worked.
For the last few weeks of the 2008 season, the hunters
continued to take photos and use their Igliniit systems. Some
also used regular GPS units to log travel, and the team met
weekly to discuss their experiences.
During that time they put together a comprehensive
report to the Calgary team, outlining their likes, dislikes, and
wish list for the next year. Te hardware issues were serious,
P
h
o
t
o
:

S
h
a
r
i

G
e
a
r
h
e
a
r
d
Q&A with Clyde River hunters David Iqaqrialu (left), Laimikie
Palluq, and Jacopie Panipak during rst year of testing
FIGURE 5 A sample English user manual page
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
36
ON THE INUIT TRAILS
but they also thought the setup procedure for the whole sys-
tem was too complicated.
For example, many steps had to be repeated if the PDA
lost power since several components were being stored in vol-
atile memory, including the font used to represent Inuktitut.
As a temporary measure, a simplifed user manual was
developed in both English and Inuktitut. A sample page of
the user manual is shown in Figure 5.
The Wish List
Te list of improvements and changes the hunters requested
included:
1. Make a simpler operating system interface for the Igliniit
application.
2. Make the system deployable on a dog team
3. Solve the frozen PDA problem
4. Make sure that the user-entered data was recorded to fle.
5. Make the Inuktitut font stable (the font periodically dis-
appeared in two of the units)
6. Add a map display
7. Develop or obtain more rugged serial cables and mount-
ing hardware
8. Develop a more rugged 12V battery adaptor
Te hunters also asked for several additional icons to be
added to the observation list.
A third group of three students began working on these
refnements in September 2008, and the system changes
(described next) are currently being feld tested through
June/July 2009 by the hunters in Clyde River.
Rening the Hardware
Te students frst tried two approaches to solve the problem
of freezing. Tey found a way to heat the PDA internally with
a modifed snowmobile handlebar heater that ft an AA bat-
tery form factor. Te heater was installed in an optional PDA
AA battery boot that was modifed to allow the PDA to be
powered externally along with the heater.
Tey placed the heated PDA in
a temperature chamber cooled to
-40C. It was tested on low and high
settings. Unfortunately, at the low
setting, the heater was unable to
heat the PDA sufciently to keep the
whole screen from freezing, and, at
the high setting, the heater melted
the sof plastic part of the battery
boot.
Te three students then tried a
second approach. Tey designed an
insulted bag for the entire PDA. Te
three-layer prototype consisted of a
ballistic nylon shell, a layer of micro-
fber insulation and an internal layer
of nylon that included a fold-back
fap for accessing the screen.
Tey attached a fat handlebar heater to a thin aluminum
plate and inserted it into a pocket in the screen fap. Tis pro-
vided direct heat to the PDA screen when the fap was closed.
Accompanying photos show the bag open and closed, with
the heating element installed and the screen fap open.
Further controlled tests were conducted to fnd out
whether the bag could keep the PDA warm with and without
the use of the heater. A live GPS signal was re-radiated inside
the temperature chamber to confrm reception of GPS sig-
nals through the bag.
Te controlled-condition test results were very promising.
Te bag alone maintained the PDA screen at a temperature
above -10C for several hours without active heating. With
heating, a constant screen temperature of -10C was main-
tained for more than eight hours in an external temperature
of -40C.
Te bag was modifed to attach permanently to the
mounting hardware, with the mounting cradle inside the bag
(so that the PDA unit could be easily mounted and unmount-
ed). Additional hardware developments included mounting
brackets with shorter lever-arms that should work better over
rough terrain.
Te team also replaced several PDA batteries and built
custom serial and power cables using materials more suitable
for cold weather.
Improving the Software
Te students removed unwanted sofware from the PDA and
simplifed the application launch procedure.
Te aforementioned sofware bug that failed at times to
write hunters observations to fle could still not be repro-
duced in Calgary. However, the team made several sofware
modifcations that insured immediate fle writing.
Because the group had spent most of the Fall 2008 semes-
ter working on hardware improvements, they decided to
defer the addition of a map display until afer deployment in
January 2009.
Custom insulated bag (left) with main access ap
open. (at right) Insulated bag with main ap closed
and heating element installed in the open screen
cover. Photos by Edward Wingate
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
T
his annual conference has
become a traditional meeting
place for the exchange of ideas
between scientists and engineers
involved in navigation, motion control
and guidance all over the world. The
conference fee per person is $350,
if paid (in cash or by bank transfer)
before 1 April 2009, and $400 if
payment is made after 1 April 2009.
For students and post-graduates the
conference fee is reduced to 50%,
i.e., $175 and $200, respectively.
The fee includes: participation in
the Conference, one copy of the Con-
ference Proceedings, a sight-seeing
guided bus tour of the city, lunches,
cofee breaks during the 3 days, trans-
portation from/to the airport, trans-
portation to the Conference from the
hotel, and the banquet on 26 May.
The conference is held during the
White Nights season and there will
be opportunities for a sight-seeing
tour of St. Petersburg and a visit to
the Hermitage. An extensive three-
day spouses program will also be
ofered during the Conference at
extra cost and include city tours and
excursions to the palaces and muse-
ums of St. Petersburg and suburbs.
Conference languages are Russian
and English. Simultaneous interpreta-
tion will be provided from Russian into
English and from English into Russian.
Full information on the Conference
including Preliminary program, Con-
ference Schedule, Conference Regis-
tration form, etc. is available on the
16th Conference website:http://www.
elektropribor.spb.ru/cnf/icins09/enfrset.
html, where you will be able to click
the link to the website of Monomax
Ltd. ompany, the service-agent of the
16th Conference, and fnd informa-
tion regarding visa execution, hotel
reservation and cultural program.
Conference Topics:
Navigation, Control and Guid-
ance Systems, and Components
Integrated Navigation Sys-
tems for Marine, Land and
Aerospace Applications
Inertial Systems and Sensors
GLONASS, GPS, Galileo and
Augmentation Systems
MEMS
Algorithms and Software
Testing and Metrology
Conference Program Committee
Chairman Academician of the
Russian Academy of Sciences,
Prof. V. Peshekhonov (Russia)
Members Mr. L. Camberlein
(France), Dr. L. Crovella (Italy), Prof.
V. Gusinsky (Russia), Prof. D. Louki-
anov (Russia), Prof. L. Nesenyuk
(Russia), Dr. B. Rivkin (Russia), Dr. G.
Schmidt (USA), Prof. H. Sorg (Ger-
many), Dr. O. Stepanov (Russia)
Among the cosponsors of the
conference in 2009 are:
the Scientifc Council of the Russian
Academy of Sciences on the Problems
of Motion Control and Navigation,
Russian Foundation for Basic Research
(RFBR), the International Public
Association Academy of Naviga-
tion and Motion Control (ANMC), the
American Institute of Aeronautics and
Astronautics (AIAA), the Institute of
Electrical and Electronics Engineers
(IEEE; USA), lAssociation Aronau-
tique et Astronautique de France
(Aaaf), lInstitut Franais de Naviga-
tion (Ifn), and Deutsche Gesellschaft
fr Ortung und Navigation (DGON,
German Institute of Navigation).
Contact information:
FSUE SRC RF CSRI Elektropribor
30, Malaya Posadskaya St.,
St. Petersburg, 197046, Russia.
Tel.: 7 (812) 499 81 18, 499 81 57
Fax: 7 (812) 232 33 76
E-mail: ICINS@eprib.ru
The 16th Saint Petersburg International
Conference on Integrated Navigation Systems
The annual INS will be held by the State Research Center of the
Russian Federation Central Scientifc and Research Institute Elektropribor
on 25-27 May 2009 in St. Petersburg, Russia. You are invited to take part.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
__________________________
____
__________
38
The Second Season
In January of 2009 came the third teams turn to prepare for
and undertake the rigorous one-week trip to Bafn Island.
Te three students worked with the hunters to install
complete systems on six snowmobiles and two dogsleds and
began testing on January 23.
With lessons learned from 2008, they brought spare PDAs
along. One extra unit stayed in Clyde River and another
returned to Calgary with the students for continued trouble-
shooting and debugging.
Tey also brought a whole lot of spare parts.
Te students and hunters hope that improved hardware,
the PDA insulator bag, and minor sofware modifcations
will lead to improved data collection this sea ice season.
A Journey in Progress
Te focus this season is on developing the technology and
getting it to work reliably. Figure 6 shows a sample trajectory
from the 2008 ice season displayed on a map of the Clyde
River area. Once more data from the 2009 season is available,
the hunters will discuss the best ways to represent the data
and how the maps and data should be used.
Te Igliniit Project is funded to run through 2011. If
developed and tested successfully, the technology may be
useful in many Arctic communities and beyond for a range
of activities such as environmental monitoring, land use
planning, and search and rescue.
Acknowledgments
Te Igliniit Project is part of the larger
Inuit Sea Ice Use and Occupancy Project
(ISIUOP). ISIUOP involves several other
Inuit communities and researchers and is
coordinated by the Geomatics and Carto-
graphic Research Centre (GCRC) at Car-
leton University in Ottawa, Canada.
Te project is one of Canadas contribu-
tions to the International Polar Year (IPY)
and to the Sea Ice Knowledge and Use
(SIKU) Project, which is a larger IPY col-
laboration among several countries.
Eleven undergraduate students contrib-
uted to this work: 2006-2007: Desmond
Chiu, Sheldon Lam, Andrew Levson, and
Jeremy Park, 2007-2008: Brandon Culling,
Josiah Lau, Tina Mosstajiri, and Trevor
Phillips, 2008-2009: Michael Brand, Ryan
Enns, and Edward Wingate.
Six hunters work on the Igliniit Proj-
ect, four since the beginning: Apiusie
Apak, David Iqaqrialu, Laimikie Palluq,
and Jacopie Panipak. Jayko Enuaraq and
Amosie Sivugat joined the team starting
2008/2009. Gary Aipellee is a project coor-
dinator, researcher, and interpreter from Clyde River who
has played a crucial role in facilitating the entire project since
the earliest stages. (Tank you, Gary!!)
Te project would not be possible without the support of
the University of Calgary, Carleton University, the Govern-
ment of Canada IPY program, the Hamlet of Clyde River,
the Clyde River Hunters and Trappers Association, the Ittaq
Heritage and Research Centre, and the Nunavut Research
Institute.
Tis article is based in part on a paper presented at the
Institute of Navigations 2009 International Technical Meet-
ing in Anaheim, California, USA.
Manufacturers
Te Igliniit Trails system incorporates a TDS (Trimble)
Recon PDA and the XC Pathfnder GPS receiver from Trim-
ble, Sunnyvale, California, USA, and a Kestrel 4000 pocket
weather meter from Nielsen-Kellerman, Boothwyn, Penn-
sylvania, USA. Te PDA application sofware was written in
Microsof Embedded Visual C++, Microsof Corporation,
Redmond, Washington, USA, and the user sees a Windows
Mobile application. Te microfber insulation used in the
cold-proofed improvised PDA bag was Tinsulate, from 3M,
Minneapolis, Minnesota, USA. Te food wrap used to seal
the PDA interface connections was Glad PressnSeal from
the Glad Products Company, Oakland, California, USA.
FIGURE 6 A draft map reects one approach to representing Igliniit data. (The team is still experi-
menting with how to best map the data for various uses.) The map shows continuously logged
trajectories and user observations. Observations shown include seals, polar bears, polar bear
tracks, and thin ice.
ON THE INUIT TRAILS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Authors
Kyle OKeefe is an assistant professor of geomatics
engineering at the University of Calgary, Calgary,
Alberta, Canada, where he completed B.Sc. and Ph.
D. degrees in the same department. He has worked
in positioning and navigation research since 1996
and in satellite navigation since 1998. His major
research interests are GNSS systemsimulation and
assessment, space applications of GNSS, carrier phase positioning, and local
and indoor positioning with ground based ranging systems.
Shari Gearheard is a geographer and research scien-
tist with the National Snowand Ice Data Center at
the University of Colorado at Boulder. Since 1995,
she has worked with Inuit communities on projects
that link Inuit and scientic knowledge of the envi-
ronment and environmental change. Gearheard
lives and works full time in Clyde River, Nunavut.
Gary Aipellee is a professional interpreter/translator and a researcher
and coordinator for the Igliniit project.
Apiusie Apak, Jayko Enuaraq, David Iqaqrialu, Laimikie Palluq, Jacopie
Panipak, and Amosie Sivugat are expert hunters fromClyde River.
Desmond Chiu, Sheldon Lam, Andrew Levson, and Jeremy Park were
undergraduate students at the University of Calgary in 2006-2007. Together
they researched and selected the best possible hardware options available
to the Igliniit Project.
Brandon Culling, Josiah Lau, Tina Mosstajiri, and Trevor Phillips were
undergraduate students at the University of Calgary in 2007-2008. Together
they developed the main software application used in this work and man-
aged the deployment of the project in the rst eld season.
Michael Brand, Ryan Enns, and Edward Wingate are University of Calgary
undergraduate students currently working on the project. They were recent-
ly in Clyde River to deploy their improvements to the systemfor a second
season of eld testing.
References
Lauriault, T and G. Laidler (2009) Inuit Sea Ice Use and Occupancy Project
(ISIUOP) webpage. Last accessed 19 Jan 2009 https://gcrc.carleton.ca/
conuence/display/ISIUOP/Inuit+Sea+Ice+Use+and+Occupancy+Project
+%28ISIUOP%29
Ollivier, Kevin (2009) wxWidgets Cross-PlatformGUI Library.
http://www.wxwidgets.org, Last accessed 19 Jan 2009
P
h
o
t
o
:

C
y
n
t
h
i
a

P
a
n
i
p
a
k
The Igliniit team in Clyde RIver, January 2009. Back row from left:
Ryan Enns, Jayko Enuaraq, David Iqaqrialu, Laimikie Palluq, Edward
Wingate, Amosie Sivugat, Jacopie Panipak. Front row from left:
Michael Brand, Gary Aipellee, Apiusie Apak, Shari Gearheard
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
______________
_______________________________________
_________
_____________________________
40
T
he issue of intentional or inadver-
tent interference to GNSS signals
is a matter of growing concern
throughout the world.
In a study released the day before
the terrorist attacks on the Pentagon
and the New York World Trade Center
in September 2001, the U.S. Department
of Transportation assessed the national
transportation infrastructures vulner-
ability to civil GPS disruption.
Te agencys investigation and subse-
quent recommendations, known as the
Volpe report, warned that as GPS fur-
ther penetrates into civil infrastructure,
it becomes a tempting target that could
be exploited by individuals, groups or
countries hostile to the U.S.
A few years later, in a 2004 National
Security Presidential Directive on space-
based positioning, navigation, and tim-
ing (PNT), former U.S. President Bush
gave the Department of Homeland
Security (DHS) responsibility for lead-
ing development of a plan to address
concerns about interference to GPS.
A
Multi-Antenna
Defense
Although GNSS spoong
transmitting spurious signals
to fool a receiver has not yet
emerged as a major problem
for civil users, it represents
a growing risk. Certainly the
capability exists and, with ever
more security-related applications
coming online, the motivation
for spoong is increasing, too. In
this article, researchers discuss
a variety of countermeasures
and demonstrate one successful
method to detect GPS spoong
with a multiple antenna arrray.

i
S
t
o
c
k
p
h
o
t
o
.
c
o
m
/
A
n
d
r
e
y

P
r
o
k
h
o
r
o
v
PAUL Y. MONTGOMERY
NOVARIANT INC
TODD E. HUMPHREYS
UNIVERSITY OF TEXAS AT AUSTIN
BRENT M. LEDVINA
VIRGINIA TECH
Receiver-Autonomous
GPSSpoong Detection
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_
_
_
_
_
_
_
DHS issued a preliminary interference
detection and mitigation (IDM) plan
last year.
To date, most actual incidents involv-
ing GPS interference whether inten-
tional or unintentional have involved
in-band or out-of-band harmonic RF
transmissions that masked the weak
GPS spread spectrum signals. A good
deal of anxiety has been expressed
in recent years about inexpensive
GPS jammers that, at power lev-
els as low as one watt, could cause
wide areas of disruption to GPS
service.
Among other types of inten-
tional interference, the Volpe
report and the IDM plan men-
tion civil GPS spoofing, a technique
by which a GPS receiver is fooled into
tracking counterfeit GPS signals. Spoof-
ing is more sinister than intentional
jamming because it is surreptitious:
the targeted receiver cannot detect the
attack and, consequently, can be fooled
into generating erroneous data that may
even be hazardously misleading.
Previous work into civil spoof-
ing countermeasures begins with an
important internal memorandum from
the MITRE Corporation in which the
author, Edwin L. Key, appears to have
examined spoofng and spoofng coun-
termeasures in detail. (For details, see
the Additional Resources section near
the end of this article.) Te memoran-
dum recommends the following tech-
niques for spoofng detection:
1. amplitude discrimination
2. time-of-arrival discrimination
3. consistency of navigation and iner-
tial measurement unit (IMU) cross
check
4. polarization discrimination
5. angle-of-arrival discrimination
6. cryptographic authentication
Of the proposed techniques, angle-
of-arrival discrimination coupled
with physical security of the antennas
provides significant protection and is
relatively easy to implement with inex-
pensive single-frequency receiver tech-
nology.
In this article we demonstrate the use
of a dual-antenna receiver that employs
a receiver-autonomous angle-of-arrival
spoofng countermeasure essentially
an implementation of Keys ffh tech-
nique. It is based on observation of L1
carrier differences between multiple
antennas referenced to a common oscil-
lator. We believe that this defense could
be effective against all but the most
sophisticated spoofng attempts.
Spoong Scenarios
Spoofing scenarios can be broadly
divided into static (fxed target receiver)
and dynamic (moving target receiver)
cases.
Static Scenario. Te target receiver of
a static spoofng scenario could be, for
example, a timing receiver deployed to
synchronize the electrical power grid,
global trading, or a communications
network.
In all such timing applications, the
GPS antenna is situated with a clear view
of the sky, typically on top of a building or
a communications tower. A receiver-gen-
erated pulse per second (PPS) is used as
the time reference for synchronization.
One can envisage a scenario where
the spoofer knows the approximate
location of the targeted receiver anten-
na. Spoofer hardware and a directional
antenna could be used to mount an
attack at a distance of hundred meters
or more. As discussed in the paper by T.
E. Humphreys et alia cited in Additional
Resources, the general approach would
be as follows:
1. grow a replica signal in the shad-
ow of the correlation peak for each
satellite, replicating the received GPS
navigation data
2. increase the power of the spoofng
signals to overcome the GPS signals
3. slew the generated signals to be con-
sistent with a desired GPS position/
time.
Clearly, this technique could be used
to fool a timing receiver into generating
a PPS that is incorrect.
Dynamic Scenario. Since January
2005, in fshing waters controlled by the
European Union (EU), Commission
Regulation No. 2244/2003 has required
that operators of fishing vessels more
than 15 meters in length carry a satel-
lite-based vessel monitoring system .
Te VMS (typically employing GPS
today) records the voyage of the ves-
sel and automatically provides the
data to the fsheries monitoring cen-
ter of the EU member state where
the vessel is registered as well as the
member state in whose waters the
vessel is fshing.
Naturally, the data can be used
to detect passage into waters for which
the vessel is not licensed. This is an
example of a dynamic scenario where
the operator of the vessel is motivated
by fnancial gain to spoof the onboard
receiver. (Indeed, Article 6, Section 2 of
regulation 2244/2003 specifes that the
VMS data are not to be altered in any
way and prohibits anyone from attempt-
ing to destroy, damage, render inopera-
tive or otherwise interfere with the satel-
lite tracking device.)
In this case of complicit spoofng
the intent of the operator would be to log
a fctional voyage that does not disclose
illegal fshing activities. In such a situ-
ation, the complicit user could discon-
nect the GPS antenna and attach instead
a local GPS signal generator.
This is exactly the scenario envi-
sioned in a November 25, 2008, presen-
tation at a conference on MENTORE
(iMplemENtation of GNSS tracking &
tracing Technologies fOR Eu regulated
domains), an R&D project funded by
the European Commissions 6th Frame-
work Program and coordinated by the
Institute for the Protection and Security
of the Citizen (IPSC) of the ECs Joint
Research Centre (JRC).
In the presentation, JRC researcher
Gianmarco Baldini referred to the exis-
tence of a simple GPS fraud kit avail-
able for 2,000 (about $2,535) that could
feed spoofed signals into the VMSs RS-
232 port after disabling the antenna
and opening the VMS box. Baldini also
In one example of complicit
spoong, the intent of the operator
would be to log a ctional voyage
that does not disclose illegal shing
activities.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
42
MULTI-ANTENNA DEFENSE
mentioned sophisticated GPS signal simulators available for
about 100,000 ($126,700) that could be connected to or radi-
ated towards the VMS unit.
Other dynamic scenarios (largely of cinematic imagination)
might involve a malicious spoofer who seeks to guide an air-
craf into the ground or a mountainside. In our opinion, this
would be very difcult to achieve technically and appears to
be unlikely in practice. Nonetheless, the mitigation technique
presented here could be efective in detecting the presence of a
spoofer in such Hollywood Scenarios.
Spoofer Categories
Te JRC MENTORE paper cited earlier mentioned a couple
of categories of GPS spoofers. In this article we identify three
main types of spoofers.
GPS signal generator. Spoofers in this category are GPS sig-
nal generators readily available from several vendors. For use
as a spoofer, the signal generators RF output is amplifed and
transmitted, possibly using a directional antenna.
In this case, the transmitted signals are not phase- and fre-
quency-matched to the GPS signals being received from satel-
lites in the locality, and the navigation data do not replicate the
currently active navigation data. Although a receiver could be
fooled by this approach, particularly if the target receiver is frst
jammed and forced to reacquire, the spoofng signal generated
in this fashion typically looks like noise rather than a usable
signal to a receiver tracking it.
GPS Receiver Spoofer. Spoofers in this category are coupled
to a GPS receiver. Te GPS receiver tracks satellite signals at a
location and decodes the navigation data.
Te spoofer then generates a signal that mimics the incident
satellite signals in all respects. Conceivably, a spoofer could
add a calculated ofset to each satellite signal to compensate
for a specifed geometric ofset to the target GPS antenna. Te
spoofer is also able to vary the signal strength of the constitu-
ent signals so that they appear at the target antenna to have the
same relative strengths of the authentic signals.
Tis spoofng confguration has one transmit antenna and
is moderately sophisticated. This kind of spoofer has been
created, as discussed in the Humphreys paper cited earlier.
Although the technical knowledge to create such a spoofer is
not widespread, the required parts are freely available and may
be purchased for a few hundred dollars.
An accompanying photograph shows the Cornell Univer-
sity GRID dual-frequency sofware-defned GPS receiver. As
an example of a spoofng platform, the Cornell GRID receiver
can simultaneously track 12 C/A channels and generate 8 C/A
spoofng channels.
Coupled with the simple RF hardware shown in the second
photo, this platform has been used to investigate the challenges
involved in mounting a spoofng attack. Te hardware has been
successful, under controlled laboratory conditions, in spoofng
several diferent single-frequency GPS receivers, as illustrated
by the screen shots in the accompanying photos.
Sophisticated GPS ReceiverBased Spoofer. This kind of
design is similar to the equipment described in the previous
category but employs multiple transmit antennas. Further-
more, the spoofer is able to vary the carrier phase outputs that
are transmitted by each antenna to control the relative carrier
phases among these transmit antennas. Creating such a spoofer
is possible but technically difcult.
Cornell University GRID software-dened GPS receiver (left) RF transmitter prototype hardware (right)
Screen shots of spoofed commercial handheld GPS receivers
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Setting Up the Experiment
To help investigate our spoofng detection technique, we used
a dual-antenna array mounted on a roofop as shown in the
accompanying photograph. Tis assembly includes a pair of L1
GPS antennas separated by 1.46 meters. Between the antennas
is the GPS receiver itself.
Te receiver/antenna assembly was designed for vehicle
navigation and automatic machine control. We used it in this
project because the internal frmware is easily modifed and the
assembly produces L1 carrier phase measurements from both
antennas referenced to a common internal oscillator.
Te internal GPS receiver is based on a 12-channel, L1 C/A-
code chipset and uses proprietary sofware. Tere is nothing
special about the hardware platform itself; the platform choice
was merely convenient for the authors.
In the experiment, we installed the assembly on the roofop
in a known and fxed orientation. We chose a level confgura-
tion in which the baseline between the antennas is oriented
along the (true) north-south axis. Although we used a single
receiver and common oscillator, the diferential phase tech-
nique employed is equally applicable to two separate receivers,
each with a single antenna and oscillator, provided the baseline
between the antennas is known.
Methodologyfor Detectinga SpoongAttack
Exploiting the equipment confguration that we have described,
we developed a technique for detecting spoofng signals based
on their deviation from the characteristics of signals received
from actual GPS satellite transmissions.
Figure 1 shows the geometry for a single satellite in which
s is a unit line of sight (LOS) vector to a GPS satellite
b is the baseline vector between the two antenna in units
of L1 cycles.
Lines of constant phase emanating from the distant satellite
are represented by parallel lines orthogonal to s and separated
by the wavelength of the L1 carrier frequency.
For the ith satellite, a scalar equation for the L1 carrier phase
diference d
i
between the two antennas is given (in units of L1
cycles) by Equation 1
where:
b is the baseline vector between the antennas (in the body
frame) in units of L1 cycles
A is a direction cosine matrix to rotate vectors in the east-
north-up (ENU) frame to the body frame
is the unit line of sight (LOS) vector to satellite
i
in the
ENU frame
N
i
is an arbitrary integer ambiguity for satellite
i
B is a constant line bias or time varying delta-clock term
(depending on implementation)
i
is the summation of all carrier phase error terms for sat-
ellite
i
Equation 1 ignores terms due to the ionosphere and tropo-
sphere as these are common mode to the submillimeter level
with the assumed meter-level baseline between the antennas.
Te expression b
T
As
i
should be recognized as the inner product
between vectors s
i
and b.
For a dynamic case, the LOS vector is known in the ENU
frame, but the moving baseline b is known in the body frame.
Te direction cosine matrix A denotes the attitude of the anten-
na array and is needed to compute the inner product. (In other
contexts, Equation 1 can be used to determine the direction
cosine matrix A when it is desired to calculate the attitude of
the antenna array.)
In the following discussion, it is assumed that the inner
product between s
i
and b is known. For a fxed (roofop) installa-
tion the attitude of the antenna array is assumed to be a known
by design or pre-survey. For the case of a dynamic (vehicle)
installation, the attitude of the antenna array is assumed to
be determined by an orientation sensor (such as an inertial
attitude sensor) that is not susceptible to GPS spoofng
For a dual-antenna receiver with a common oscillator, B
is a constant associated with the diference in the electrical
length of the pathways from the antennas to the receiver. For a
receiver with separate oscillators (one for each antenna), B is a
non-constant bias dominated by the clock ofsets between the
two oscillators. In either case, B is common for all satellites.
Figure 2 shows a plot of delta phases (d) for four satellites
tracked for approximately one hour. Note that the various satel-
lites delta phases vary distinctly from one another. Tese data
were generated by the antenna array described earlier with the
antennas aligned in a north-south orientation. In Figure 2, the
integer (N
i
) for each satellite has been set to an arbitrary value
for convenient plotting.
Note that:
GPS antenna array
FIGURE 1 Antenna diversity geometry for a single satellite
A B
Receiver
b
S
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
44
Te observed change in dover time
is due to satellite motion
Te rate of change in d is propor-
tional to the baseline length
Noise on the measurements is due to
carrier multipath and carrier phase
noise
Based on t he known at t it ude
expressed in A and the known line
bias B, the expected values of the
delta phases shown in Figure 2 are
plotted 0.1 cycles below the mea-
surement data (for visual clarity). In
this case, the attitude was known to
approximately 0.1 degrees in pitch
and 0.3 degrees in azimuth.
Identifying a
Spoofed Signal
Te plot in Figure 2 illustrates the basic
idea for spoofng detection using multi-
ple antennas. If the d
i
measurements do
not agree with the expected phase pro-
fles within bounds set by the expected
noise and attitude uncertainty, then a
spoofng signal is identifed.
As showninFigure 3, a spoofer trans-
mitting froma single antenna has a very
diferent geometry fromthat of a receiver
tracking GPS satellites distributed across
the sky. Consistent with the geometry,
the d
i
profiles for a point transmit-
ter would all overlay each other except
for contributions due to multipath and
phase noise.
Te spoofng detection algorithmwe
implemented is, therefore, straightfor-
ward:
1. Te expected delta phases (d
i
) are
calculated based on known attitude
and line bias B.
2. Te measured delta phases are com-
pared (modulo 1 cycle) with the
expected delta phases.
3. For each satellite, the error between
the expected and the measured data
is calculated every 500 milliseconds
(or other update rate as desired).
4. Based on an error threshold that
is a function of satellite elevation,
expected worst-case, multipath and
attitude uncertainty, a limited up-
down counter is incremented or
decremented.
5. If the up-down counter reaches a
specifed maximum(based on sam-
ple rate and time to alarm), it triggers
a spoofng-detection alarm.
The algorit hm described here
requires that the attitude of the anten-
na array be known. Tis is not a great
problemfor a static array; however, the
dynamic case requires integration with
an independent attitude reference.
An alternative attitude-independent
implementation of the multiple-antenna
spoofng detection technique could be
designed to raise an alarm when the
delta phases are suspiciously close to one
another; that is, whenall the delta phases
overlay each other within a bound that
is just wide enough to accommodate
worst-case multipath and carrier noise.
Such an approach would have to
deal with rare cases where, for a brief
moment, the true satellite geometry hap-
pens to cause all delta phases to over-
lay each other (modulo one cycle). Tis
situation will occur on occasion and, if
not handled correctly, will lead to a false
alarm.
For a spoofer to defeat the algorithm
as implemented, the spoofing system
must emulate the expected carrier phase
deltas between the pair of antennas for
all satellites in track. As illustrated in
Figure 3, even a sophisticated spoofer
cannot emulate this geometry for several
satellites if the spoofer is limited to one
transmitting antenna.
A sophisticated spoofer with two
separate points of transmission might be
able to defeat the algorithm. However,
this would also require the spoofer to
know the geometry of the GPS anten-
na array, locate a matched transmitter
antenna very close to each GPS antenna,
and deal with other difcult problems
associated with multipath, signal leak-
age, and self-interference.
Such an attack would not be possible
without physical access to the antenna
installation. Of course, we could extend
the spoofng defense described here in
a straightforward way by using three
or more antennas making a multi-
transmitter spoofng attack even more
difficult. For this reason, we believe
the use of antenna diversity and physi-
cal security leads to a robust defense
against even sophisticated spoofing
attacks.
The implementation that we have
just described uses a single receiver
FIGURE 3 Antenna diversity geometry for a single satellite and point
transmitter
FIGURE 2 Carrier phase deltas di for four GPS satellites over one hour
1.5
1
0.5
0
-0.5
-1
-1.5
-2
L
1

c
y
c
l
e
s
0 500 1000 1500 2000 2500 3000 3500
Time (s)
Delta phase
A R
Receiver
b
S
Spoofer
electronics
Transmit
antenna
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
with a common oscillator. In this case,
the line-bias (B) is a predetermined
constant.
If B is unknown or non-constant,
one can simply remove it by subtract-
ing one satellite measurement from all
the others. (Typically, one subtracts the
measurement from a high elevation sat-
ellite.)
With this modifcation and a com-
mensurate increase in the error thresh-
old to accommodate increased phase
noise and multipath, the algorithm
is easily implemented using separate
receivers and antennas. The multi-
receiver approach only requires that the
vector between the two antennas is con-
stant and known.
Indoor Experiment
After developing the receiver autono-
mous spoofng detection (RASD) sof-
ware, we mounted the antenna array
depicted earlier on the roof and enabled
the sofware. Te algorithm was tested
for several days in an unspoofed set-
ting to validate that spurious (false)
detections were not fagged.
We used a detection threshold of
0.1 cycles for zenith satellites, with a
linear increase to 0.25 cycles for sat-
ellites at the horizon. Tese thresholds
were found empirically.
Initially, we intended that our experi-
mental design would employ the Cornell
University GRID receiver and the proto-
type RF transmitter hardware pictured
earlier to implement a spoofng attack
against the GPS receiver and antenna
array with a clear view of the sky, as fol-
lows:
1. Create a spoofng attack against the
test receiver with the RASD frmware
disabled and validate (using an oscil-
loscope) that the spoofer was able to
drive the PPS generated by the target
receiver away from truth.
2. Enable the RASD firmware and
execute the same attack, this time
validating that the receiver was able
to detect the attack and raise an
alarm.
Our plan was complicated by the
necessity to perform the experiment
outside and with a clear view of the sky.
Te issue: transmitting outdoors in the
L1 frequency band is illegal!
In view of this, we elected to sim-
plify the experiment and move indoors.
The revised experiment used a real
GPS signal from a rooftop mounted
GPS antenna. Te L1 band was ampli-
fed and re-transmitted indoors from a
point source as illustrated in Figure 4.
(Tis setup is identical to that occasion-
ally used at trade shows to allow GPS
receiver vendors to demonstrate active
GPS receivers indoors without cable
runs to the roof.)
We also moved the antenna array
indoors and mounted it near to the re-
transmit antenna. In this environment
the receiver was able to track the re-
transmitted signal without cycle slips
when the range to the transmit antenna
was less than approximately six meters.
The indoor receiver was also able to
decode all navigation data and calculate
a position fx (the position of the roofop
antenna).
Figure 5 plots the delta phase profles
of seven satellite signals tracked during
the indoor experiment. Consistent with
the geometry, one may observe that all
delta phases from the point source lie on
top of each other.
Having calculated the position
fix and the LOS vectors, the RASD-
enabled receiver immediately detected
and flagged the retransmitted signals
as coming from a spoofer not from
the satellites themselves. Although this
experiment did not detect an actual
spoofer, the situation is sufciently rep-
resentative that the approach and sof-
ware implementation were considered
successful.
Conclusions
Antenna diversity employing either
multiple separate receivers or a multi-
antenna single-oscillator receiver can
be used to defend against intentional
GPS spoofng by greatly increasing the
FIGURE 4 Indoor Experimental Setup FIGURE 5 Experimental delta phase results with point (re)transmitter
Rooftop Antenna
Indoor
Retransmit
Antenna
Indoor Antenna Array
The multi-receiver
approach only
requires that the
vector between the
two antennas is
constant and known.
1
0.8
0.6
0.4
0.2
0
-0.2
-0.4
-0.6
-0.8
-1
L
1

c
y
c
l
e
s
0 200 400 600 800 1000 1200
Time (s)
Delta phase
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
46
technical difculty required to mount a
successful attack.
In general, an additional spoofer
transmitter is required for each addi-
tional GPS antenna. Furthermore, a
spoofer would have to locate each trans-
mit antenna in close physical proximity
to the appropriate GPS antenna in the
array.
If the GPS antennas of static or
dynamic installations are further pro-
tected by physical security, it is possible
to create a robust defense against even
a sophisticated spoofng attack. In the
case of a complicit user, the presence of
multiple antennas makes it difcult to
intentionally defeat the system by direct
injection of an artifcial GPS signal.
In the spoofng defense implemented
here, a one-time survey of a fxed anten-
na array was sufcient to enable receiver
autonomous spoofng detection. A prac-
tical but slightly less robust defense that
does not depend on knowledge of the
attitude of the multi-antenna array can
also be implemented.
The technology to enable multi-
antenna spoofing detection is readily
available using any of the numerous GPS
receivers that produce L1 carrier phase
observables.
Acknowledgments
This article is based substantially on
material in a paper frst presented Janu-
ary 26, 2009, in Anaheim, California, at
the International Technical Conference
of the Institute of Navigation.
The authors would like to thank
Novariant for the use of the AutoFarm
roof array used for the experiment. Spe-
cial thanks to Dennis Connor of Novari-
ant for supporting this work. Additional
thanks to William J. Bencze for RF hard-
ware development support.
Manufacturers
The dual-antenna array used in the
experiments described in this article was
the AutoFarm antenna from Novariant,
Inc., Fremont, California, USA. The
arrays internal GPS receiver is based on
the GP2015/GP2021 chipset, Zarlink
Semiconductor Inc., Ottawa, Ontario,
Canada, and uses Novariant propri-
etary software. The spoofed handheld
receivers were the eTrex from Garmin
International, Olathe, Kansas, USA, and
the iPhone from Apple Inc., Cupertino,
California, USA.
Additional Resources
[1] Baldini, G., and J. Hofherr, IPSC Projects based
on Satellite Navigation Systems, 1st MENTORE
Event, Institute for the Protection and Security of
the Citizen, European Commission Joint Research
Center, Ispra, Italy, November 25, 2008
[2] Commission Regulation (EC) No 2244/2003,
Laying down detailed provisions regarding sat-
ellite-based Vessel Monitoring Systems, Ofcial
Journal of the European Union, L 333/17, Decem-
ber 12, 2003, Brussels, Belgium
[3] Humphreys, T. E., and B. M. Ledvina, M. L.
Psiaki, B. W. O Hanlon, and P. M. Kintner, Jr.,
Assessing the Spoong Threat: Development of
a Portable GPS Civilian Spoofer, Proceedings of
ION GNSS 2008, Institute of Navigation, Savanna,
Georgia, USA, 2008
[4] Key, E. L., Techniques to Counter GPS Spoof-
ing, internal memorandum, The MITRE Corpo-
ration, Bedord, Massachusetts, USA, February
1995
[5] United States Positioning, Navigation, and
Timing Interference Detection and Mitigation Plan
Summary, U.S. Department of Homeland Security,
Washington, D.C., USA, April 2008
[6]Vulnerability Assessment Of the Transporta-
tion Infrastructure Relying on the Global Position-
ing System, Technical Report, U.S. Department of
Transportation, John A. Volpe National Transporta-
tion Systems Center, Cambridge, Massachusetts,
USA, 2001
Authors
Paul Montgomery is the
principal engineer at
Novari ant, I nc. He
received a Ph.D in aero-
nautics and astronautics
from Stanford University
and was a founding
member of Novariant, where his accomplishments
have included adapting integrity beacon landing
system (IBLS) technology to the Outrider TUAV
(tactical unmanned aerial vehicle) and X-31 auto-
matic landing systems. In 2006, he was inducted
into the Space Technology Hall of Fame for his
contributions to the Novariant AutoFarm RTK
Autosteer technology.
Todd E. Humphreys is a
research assistant pro-
fessor in the Department
of Aerospace Engineer-
ing and Engineering
Mechanics at the Univer-
sity of Texas at Austin.
He will join the faculty of the University of Texas
at Austin as an assistant professor in the fall of
2009. He received a B.S. and M.S. in electrical and
computer engineering from Utah State University
and a Ph.D in aerospace engineering from Cornell
University. His research interests are in estimation
and ltering, GNSS technology, GNSS security, and
GNSS-based study of the ionosphere and neutral
atmosphere.
Brent M. Ledvina is an
assistant professor in the
Bradley Department of
Electrical and Computer
Engineering at Virginia
Tech. He received a B.S.
in electrical and com-
puter engineering from the University of Wisconsin
at Madison and a Ph.D. in electrical and computer
engineering from Cornell University. His research
interests are in the areas of software receivers,
GNSS applications, estimation and ltering, iono-
spheric physics, and space weather.
The technology to
enable multi-antenna
spoong detection is
readily available using
any of the numerous
GPS receivers that
produce L1 carrier
phase observables.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_______________________
48
I
n 2003 I published a paper called Te GPS EASY Suite
Matlab Code for the GPS Newcomer. Te paper consisted
of 10 parts, which are cited in the accompanying sidebar,
EASY Suite I Topics. Each installment included a printed
text and a fle of related Matlab code or scripts (an M-fle) that
could be downloaded from a designated web site.
Each part contained an answer to ofen-asked questions
in my classes. I did not, however, only ofer an answer to the
questions, but also a Matlab code that solves the problem. Te
article became a tremendous success and remained the most
downloaded paper from the GPS Solutions site for more than
a year.
Now, students love this sort of support. But I also received
positive reactions from professionals who used the code in
research papers, which saved them a lot of coding eforts.
Te original Matlab code also turned out to be the most
downloaded fle from the Aalborg website. It resulted in numer-
ous e-mails from interested readers asking for more fles. Tese
requests now answered by the creation of eight additional M-
fles. Some involve more complex problems and coding.
Since 2003, I have received a steady fow of additional ques-
tions from readers of Gilbert Strangs and my book, Linear
Algebra, Geodesy, and GPS. Last summer I decided to create
another set of common GPS problems/solutions and accompa-
nying Matlab code EASY Suite II. Tese will be presented in
serial fashion, as one or more exercises in this and forthcoming
issues of Inside GNSS.
Introducing the New Suite
Easy Suite II augments the set of basic computational tasks with
the following topics:
EASY11, stereographic sky plot of satellite orbits and plot of
time when satellites are above a given local horizon
EASY12, details of the LAMBDA method, explained
through a small numerical example
EASY13, receiver autonomous integrity monitoring
(RAIM), horizontal protection level (HPL), and vertical pro-
tection level (VPL)
EASY14, sample of space-based augmentation system
(SBAS) corrected positions and their presentation in Stanford
plots
EASY15, accuracy comparison between pseudorange based
stand-alone positions, baselines computed using pseudoranges
alone, and pseudoranges and carrier phase observations
EASY16, error analysis of a selected one-way observation
EASY17, satellite orbits in inertial and Earth-centered,
Earth-fxed (ECEF) systems, and curve defned by sub-satel-
lite points
GPS EASY Suite II
AMatlab Companion
Many practitioners in the GNSS eld are familiar with Matlab, a high-
level technical computing language and interactive environment for
algorithm development, data visualization, data analysis, and numeric
computation. In this new series, a prominent Danish GNSS researcher
uses Matlab to illustrate and explain a variety of common GPS issues.
KAI BORRE
AALBORG UNIVERSITY
I
m
a
g
e

o
f

G
P
S

c
o
n
s
t
e
l
l
a
t
i
o
n

b
a
s
e
d

o
n

p
u
b
l
i
c

d
o
m
a
i
n

l
e

f
r
o
m

W
i
k
i
m
e
d
i
a

C
o
m
m
o
n
s
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
EASY18, computation of diferential corrections at a base
station.
Te original suite was thought of as comprising the base for
an elementary course in GPS while, as refected in the subjects,
the new suite is of a more optional, topical character.
In 2003 and later, many users encountered difficulty in
fnding the necessary fles for running individual EASY-fles.
Terefore, I chose to copy all fles needed into single directo-
ries accessible online, so that each directory becomes self-con-
tained. Te price we pay is multiple copies of basic M-fles.
Te complete set of Easy Suite II Matlab codes can be found
in compressed (zipped) fles at <http://gps.aau.dk/~borre/
easy2>. Readers can fnd much of the theoretical background
for the EASY scripts in Chapters 14 and 15 of Linear Algebra,
Geodesy, and GPS (see Additional Resources for details). How-
ever, we have added some text that emphasizes certain issues
that are central in the codes, but may be difcult to fnd in a
textbook.
EASY11: GPS Sky Plots
Te EASY-Suite starts with a polar plot of satellite orbits (see
Figure 1) as viewed from a given location, a graph showing the
number of visible satellites, and the period of time when they
are visible during 24 consecutive hours.
EASY11 itself is based on an almanac downloaded most eas-
ily from the National Geodetic Survey (NGS) website <http://
www.ngs.noaa.gov/CORS/Data.html>. Te actual fle name is
brdc1550.08n.
Te RINEX fle has been reformatted into Matlabs binary
format for satellite ephemerides using the M-fle rinexe. Te user
enters (,) for the position on the ground where the plot is to be
used, and a value for the elevation mask to be set. Ten the azi-
muth and elevation angles for all visible positions of the included
satellites are computed and plotted as polar coordinates.
Next follows bookkeeping on how many and which satel-
lites are visible during the day and the time periods when they
can be seen. Figures 2 and 3 show the resulting plots.
Te Matlab code is simple, the result is impressive, and use-
ful. In early GPS days when the constellation was incomplete,
such plots were especially valuable for planning purposes to be
sure that enough satellites would be available for positioning.
Except in the most severe terrain and urban canyons, receivers
can fnd plenty of GPS (and GLONASS) satellites around the
clock, and this situation will become even more pronounced
when Galileo satellites are launched.
EASY12: LAMDA Method
Most students fnd the theory behind the Least-squares AMBi-
guity Decorrelation Adjustment (LAMBDA) for ambiguity
resolution difcult to understand. We shall try to smooth the
path.
First we describe in a subsection the linear algebra involved.
Next we add an M-script that elucidates how the method works
on a simple case with three ambiguities.
Linear Algebra for LAMBDA
Let the vector b contain the three components of the baseline
and the vector a contain ambiguities for the L
1
frequency and
possibly for the L
2
frequency. Te double diferenced observa-
tions are collected in the vector y:
Te normal equations are
FIGURE 1 Sky plot including all GPS satellites for a period of 24 hours at a
given position. The elevation mask is 10 degrees.
EASY Suite I Topics
The original Easy Suite, which can be found online at <http://kom.aau.
dk/~borre/easy/>, dealt with the following issues:
EASY1, time conversion
EASY2, computation of a satellites position froman ephemeris
EASY3 computation of a receivers position frompseudoranges
EASY4 computation of a baseline frompseudoranges alone
EASY5 computation of a baseline frompseudoranges and
phase observations using a least-squares solution
EASY6 the same, but nowusing a Kalman
lter for the baseline estimation
EASY6e the same, but introducing down-
weighting of older observations
EASY7 estimation of receiver clock offset
EASY8 check of cycle slips and receiver clock reset
EASY9 various coordinate representations of a given baseline
EASY10 estimation of ionospheric delay for the individual satellites
0
30
60
90
120
150
180
210
240
270
300
330
0
Skyplot for the position () = (57, 10)
Elevation mask 10
All PRNs except 1
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
____
________________________
____
_________
50
and the formal solution is
Te components of the solution vector are reals; however,
we want a solution of integers! Hence, we fnd an integer vec-
tor a such that
Afer the integer solution is found we substitute it for . Con-
sequently, the solution for changes to . In order to determine
, we multiply the lower block row in (3) by and subtract
from the upper block row:
Te upper block row gives:
Te right side is known and constant. If we change to , then
changes to and we have:
or
Te right side is known and is quickly found.
Minimizing a Quadratic Expression over
Integer
Te LAMBDA method starts as other least-squares computa-
tions by solving the normal equations (3). Te result is partly
the vector of real-valued ambiguities and partly the pertinent
covariance matrix .
Te core problem can be stated in a very clear form
In order to emphasize the integer nature of the problem we have
swapped the notation from a to I.
Tis is the problem we study, and in one case this is espe-
cially simple. If is diagonal, the best vector comes from
rounding each component of to the nearest integer.
Te components are uncoupled when is diagonal. Te
quadratic is purely a sum of squares . The
minimum comes by making each term as small as possible. So
the best is the integer nearest to .
In practice the individual components are highly cor-
related. Recalling that we ofen deal with 20 ambiguities, we
realize that a search procedure to fnd the minimum is unre-
alistic. Terefore, an idea about decorrelating the ambiguities
as much as possible, before starting the search, should lead to
an efective procedure.
This is just what the LAMBDA method does. It was
described by Peter Teunissen in 1993 and is, from a theoretical
point of view, still considered to be the best method.
Te frst step consists in transforming such that its of-
diagonal entries become numerically smaller. Tese entries
measure the correlation.
We start by computing the LDL
T
decomposition of the given
covariance matrix
Next we construct a matrix of integers Z from L by a
sequence of integer Gauss transformations and permutations
such that
is as diagonal as possible. Now the search defned by (4) is
substituted by a search over integers I:
EASY SUITE II
14
12
10
8
6
4
2
0
Elevation mask 10
#

o
f

V
i
s
i
b
l
e

S
a
t
e
l
l
i
t
e
s
0 6 12 18 24
GPS Time (hours)
8
16
24
32
Solid Lines Indicate Visible Satellites
P
R
N
s
0 6 12 18 24
GPS Time (hours)
FIGURE 2 Number of visible satellites
FIGURE 3 Time period when the visible satellites are 10 degrees or higher
above the horizon
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Te choice of Z is based on integer elimination start-
ing with the frst row of L. Tis will certainly involve
row exchanges and, therefore, Z will not be triangular.
Te essential idea for this lattice basis reduction was
given by A. K. Lenstra et alia in 1982 (See Additional
Resources), and the algorithm is sometimes called L
3
. For
further details see Strang and Borre, pages 495499.
Te result of the search is the integer vector . Finally, is
transformed back to the ambiguity space according to
Note that both Z and Z
-1
must have integer entries. A con-
sequence of this is that det(Z) = 1 as seen from Cramers rule.
Te condition on the determinant of Z means that the Z trans-
formation preserves the search volume. At the end of this pro-
cess, we have transformed a highly correlated space (elongated
ellipses) into a sphere-like search space, which diminishes the
search time tremendously.
EASY12 illustrates the computational steps including a few
numerical details. We choose a numerical example from P. J. de
Jonge and C. C. J. M. Tiberius (1996). Further computational
details are well described in that report.
A Numerical Example
Given the foat ambiguities
with covariance matrix
We introduce integer shifs of in order to secure that -1 <
1:
Tis leaves the remainders
Now we factorize into LDL
T
with the lower triangular
matrix
and the diagonal matrix
We then compute the initial size of the search ellipsoid as the
squared distances of partially rounded foat vectors to the foat
vector in the metric of the covariance matrix. In the present
example the search volume is determined by
2
= 1.245.
Te call computes the
integer transformation matrix Z, a decorrelated covari-
ance matrix , and the transformed version of the LDL
T
decomposition. All quantities are mentioned below:
Te transformed and shifed ambiguities are
Te transformed, decorrelated covariance matrix is
Note the diminished of-diagonal terms! Te lower triangular
L
t
used in the search is
and the diagonal matrix D
t
used in the search is
Determining the size of the search volume for the trans-
formed L
t
and D
t
yields
2
= 1.218. Te search domain is defned
as , for I integer. Te fnal and fxed
ambiguities are
Comparing this vector with in (9), we notice that the fnal
integer result could not be derived from the foat values; this is
due to the non-zero correlations between the individual ambi-
guities.
With this we end the simple numerical example demon-
strating the mechanism of the LAMBDA method.
In the next issue of this periodical, we will illustrate the con-
cept of receiver autonomous integrity monitoring (RAIM), and
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
52
EASY SUITE II
horizontal and vertical protection levels
as applied especially in aviation.
Manufacturers
Matlab is a product of Te MathWorks,
Inc., Natick, Massachusetts, USA.
Additional Resources
[1] de Jonge, P. J., and C. C. J. M. Tiberius, The
LAMBDA Method for Integer Ambiguity Estima-
tion: Implementation Aspects, Delft Geodetic
Computing Centre LGR series, No. 12, Delft, Neth-
erlands, 1996
[2] Lenstra, A. K., and H. W. Lenstra, Jr., and L.
Lovsz, Factoring polynomials with rational
coefcients, Mathematische Annalen 261 (4):
515534, (1982)
[3] Strang, G., and K. Borre, Linear Algebra,
Geodesy, and GPS, Wellesley-Cambridge Press,
Wellesley, Massachusetts, USA, 1997
[4] Teunissen, P. J. G., The least-squares ambi-
guity decorrelation adjustment: a method for
fast GPS integer ambiguity estimation, Journal
of Geodesy, 70: 6582, 1995
Author
Originally graduated as a
chartered surveyor, Kai
Borre obtained his Ph.D.
in geodesy from Copen-
hagen University, and a
Doctor of Technology
degree from Graz Uni-
versity of Technology. He has been a full professor
in geodesy at Aalborg University since 1976.
For more than 30 years Borre has conducted
research and performed education in the area of
satellite based positioning. In 1996 he established
the Danish GPS Center and since 2000 has been
head of a two-year international M.Sc. program
in GPS technology.
Borre is a coauthor of the widely used text-
books, Linear Algebra, Geodesy, and GPS and
A Software-Defined GPS and Galileo Receiver;
Single-Frequency Approach.
Since the early 1990s Borre has published
Matlab code for processing of GPS observations,
and since 2003 he published Matlab code together
with explanatory text, a very successful new peda-
gogical concept.
GPS produced a one-meter URE in
2008, Buckman said. Te GPS III satel-
lites, which will carry the new civil L1C
signal, are designed to have a URE that
is four times better.
Galileo. Turning at last to Europes
Galileo, the laborious process of con-
tracting out the fully operational
capability (FOC) system development
continues. In Munich, Fotis Karamitsos,
European Commission director-gen-
eral for transport and energy, and Paul
Verhoef, head of the Galileo unit, indi-
cated that agreements with companies
winning the lead contracts for six work
packages should be signed between Sep-
tember and the end of this year.
Discussions at the Summit revealed
tensions around negotiations with
China about a frequency overlay of
Compass signals on the security-ori-
ented Public Regulated Service as well
as the question of whether the costs
to build Galileo can be kept within
the 3.4-billion limit agreed by the
European Council and the European
Parliament.
In answer to a question at the
March 3 opening plenary, Karamitsos
insisted that we have no reason to
believe that FOC wont be delivered on
time and on budget.
Responding to a comment that sev-
eral member states and private compa-
nies have already suggested creating a
light version of Galileo fewer ser-
vices, signals, and/or satellites, Karamit-
sos said he that the European Union
(EU) member states have a legal obliga-
tion to deliver the full system. Galileo
satellites will be acquired in blocks of
10, 8, and 8.
Karamitsos complained of people
negotiating through the press, adding,
In this time of economic constraints it
doesnt make sense for our industry to
try to make money over the amount
allocated for the program.
According to one European source,
the reference was to Surrey Satellite
Technology Ltd. (SSTL), a UK frm
whose acquisition by EADS Astrium
closed in January as well as EU mem-
bers uninterested in using the PRS.
SSTL, which specializes in smaller,
economical satellite designs, built Gali-
leos GIOVE-A satellite now in orbit.
SSTL, along with its bidding part-
ner OHB System AG (OHB), has been
short-listed as a candidate for the Gali-
leo FOC space segment (with EADS as
the other contender) and are preparing
for the submission of a refned pro-
posal to the European Space Agency.
Versus Compass. Meanwhile, the
issue of the Compass/Galileo signal
overlay which recalls an earlier
attempt to overlay the PRS on the GPS
M-code continues unresolved afer
two meetings between Chinese and EC
representatives. Some years ago, China
attempted unsuccessfully to gain
access to the encrypted PRS, which
requires unanimous agreement of EU
member states before a non-EU nation
can do that.
PRS needs spectral separation,
insisted Paul Verhoef, head of the ECs
unit for Galileo and intelligent trans-
port, who acknowledged that negotia-
tions with China are going slower
than we hoped.
Chinas ambitious launch sched-
ule, which requires fnal decisions on
Compasss frequency plan, increases
the urgency of the dialog. We hope to
get agreement [with Galileo] before we
launch, but we cannot wait to do the
validation and development of the sys-
tem, Jing said in response to a ques-
tion from the Munich audience.
Te situation refects the ill will that
has arisen since the two sides signed
agreements in 2003 and 2004 to coop-
erate on Galileo, including a 200-mil-
lion Chinese contribution to program
development.
In the session on competition among
GNSS systems, Yin said that Chinas
industry had found it hard to compete
for contracts in the Galileo FOC pro-
curement, even though the nation had
allocated 70 million for the in-orbit
validation (IOV) phase. Several IOV
cooperation projects could not be imple-
mented smoothly, due to obstacles and
barriers, he added.
GNSS World continued from page 17
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Broadcom,
SiRF Top GPS
IC Ranking
A sophisticated vendor matrix that mea-
sures qualitative as well as quantitative
factors ranks Broadcom as the leading
supplier of GPS chips worldwide.
SiRF Technology and Texas Instru-
ments came in second and third in
the GPS Integrated Circuit Manufac-
turer Vendor Matrix, developed by ABI
Research based in Oyster Bay, New York,
with results announced March 12.
Rather than measure one or a few
market metrics, such as unit or dollar
volume and market share, the Vendor
Matrix is designed to provide a fuller
understanding of vendors positions in
specifc markets. Te matrix assesses
vendors endors on the key parameters
of innovation and implementation
across several criteria.
Broadcom ranked frst in implemen-
tation and SiRF in innovation.
Under innovation, ABI Research
examined design innovation, the use of
evolving fabrication technology, packag-
ing innovation, features (sofware/frm-
ware, power, size), fexibility (chip pack-
aging, OEM integration options), the
vendors recent product introductions,
and whether the ICs ofer multiple radio
integration.
Under implementation, the research-
ers evaluated companies in light of the
following criteria: the ability of the
design to meet performance and feature
objectives, the vendors sourcing strate-
gies and the strength of their supply
chains, their marketing/sales strategy,
structure, and recent successes, the
frms organizational alignments and
structures to support their GPS mar-
ket strategy, their market presence, the
efectiveness of their product strategies
and strength of their product lines, their
partnerships, their relative market share,
and their overall organizational health
Te ranking of the other top 10 spots
on the GPS IC vendor matrix is as fol-
lows:
4) Qualcomm Inc
5) STMicroelectronics
6) u-blox AG
7) Atheros Communications Inc
8) Infneon Technologies AG
9) Atmel Corporation
10) MediaTek Inc
People
NovAtel Names NewCEO
NovAtel, Inc., has named Michael Rit-
ter as the companys new president and
CEO.
Ritter succeeds
Jon Ladd, who
will take on a new
strategic advisory
role with NovAtels
parent company,
Hexagon AB, of
Sweden. A provider
of core GNSS technology, including GPS,
GLONASS, and Galileo receivers mostly
in high-precision OEM form factors,
NovAtel is based in Calgary, Alberta,
Canada.
A trained engineer, Ritter has more
than 20 years of experience in the high-
precision GNSS industry, most recently
as a senior executive with Trimble Navi-
gation Ltd. At Trimble, Ritter served for
several years as general manager of the
companys construction services divi-
sion. Prior to Trimble, Ritter held senior
positions with Schlumberger and various
other companies around the world.
NewBuilds
Racelogic Launches GPS
Data Logger/Replay System
Racelogic Ltd., of Buckingham, United
Kingdom, has introduced LabSat, a one-
box system designed to record and replay
real-world GPS data for use in designing
and validating GPS products.
Based on an in-house system devel-
oped for testing and calibrating Racelog-
ics VBOX products, LabSat connects to
a PC via USB 2.0 and records raw signals
from a GPS antenna directly to hard
disk, without limitations on the number
of satellites being tracked or the length
of the recording. VBOX is a GPS-based
system for measuring the speed and posi-
tion of a moving vehicle in performance
and brake testing.
According to the company, when the
logged data is replayed, a GPS receiver
will recreate the movement and satellite
reception encountered during the test,
including all original multipath, satellite
obstructions, and atmospheric efects.
Options include LabSat Signal Archi-
tect sofware that can simulate dynamic
scenarios without the artifacts of real-
world data and optional IMU modules
that enable users to record data from
inertial sensors and replay them syn-
chronized with GPS data.
Ventures
CSR Acquires SiRF
CSR and SiRF announced plans to merge
the latter company a pioneer in devel-
oping chipsets for GPS consumer prod-
ucts into the former in a pure stock
deal. SiRF shareholders will receive 0.741
of an ordinary share in CSR for each
SiRF share, which values SiRF at approx-
imately $136 million (91 million).
Tats a premium of approximately
91 percent on the closing SiRF share
price of $1.08 on February 9. Following
completion, SiRF shareholders will own
approximately 27 percent of the enlarged
CSR group.
Formally, SiRF will merge with a
newly organized U.S. subsidiary of CSR.
Completion of the merger is subject
to the approval of both SiRF and CSR
shareholders, clearance by regulatory
agencies, and certain other conditions.
CSR, with a focus on handset manu-
facturers, says it ranks frst in every Blue-
tooth market segment with a unit market
share in excess of 50 percent.
SiRFs interim CEO, Dado Banatao, is
expected to join the CSR board as a non-
executive member, and Kanwar Chadha,
a SiRF founder and vice-president of
marketing, as an executive director.
For an in-depth analysis of the CSR/
SiRF merger, visit our website: <http://
www.insidegnss.com/node/1356>.
INDUSTRY VIEW
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
____
54
June
JUNE 1-4
2009 JNC
Orlando, Florida USA
Military Navigation Technology:
The Foundation for Military
Ops is the theme of the 2009
Joint Navigation Conference,
cosponsored by ION and the Joint
Service Data Exchange.
<www.jointnavigation.org>
JUNE 2-5
NIST TIME & FREQUENCY
METROLOGY SEMINAR
Boulder, Colorado USA
The 2005 Nobel Prize winner in
physics will speak at this National
Institute of Standards and
Technology annual event.
<http://nistboulder.net/Seminar34.
htm>
May
MAY 3-6
ENC-GNSS 2009
Naples, Italy
The European Navigation
Conference-GNSS will be hosted by
the Italian Institute of Navigation
(IIN) this year. The theme is
Applications and Services.
<www.enc-gnss09.it/index.html>
MAY 6-7
2009 GPS PARTNERSHIP
COUNCIL
Los Angeles AFB, California
USA
The theme of this years meeting for
industry partners, sponsored by the
GPS Wing, is Leading Space-Based
PNT for the Warghter, the Nation,
the World.
<www.afcea-la.org>
MAY 12-15
INTERNATIONAL SATELLITE
NAVIGATION FORUM 2009
Moscow, Russia
This international forum and the
associated NaviTech Expo is aimed
at Russian end users of navigation
products and services in all sectors
and companies interested in the
Russian market. In Russian with
simultaneous English translation.
<www.glonass-forum.ru/eng>
MAY 25-27
ICINS 2009
St. Petersburg, Russia
Conference for scientists and
engineers involved in navigation,
motion control and guidance
for land, aerospace and marine
civil applications. Organized
by Elektropribor, the Russian
Federations state research center.
<www.elektropribor.spb.ru/cnf/
icins09/enfrset.html >
GNSS TIMELINE
April
APRIL 14-17
AAGG 2009
Mendoza, Argentina
Argentine Association of Geophysics
and Geodesy Scientic Reunion
and rst Workshop of Continuous
GNSS Stations from America and the
Caribbean. <www.aagg2009.org/
english/home>
APRIL 19-24
EGU GENERAL ASSEMBLY 2009
Vienna, Austria
The European Geosciences Union
meeting features a geodesy program
with sessions of particular interest
to Inside GNSS readers.<http://
meetings.copernicus.org/egu2009>
APRIL 20-24
EFTF/IFC 2009
Besanon, France
The European Frequency and Time
Forum and IEEE International
Frequency and Control Conference
includes a section on Timekeeping,
Time and Frequency Transfer and
GNSS Applications. <www.eftf-ifcs-
2009.com/
APRIL 28-29
GPS-WIRELESS 2009
San Mateo, California USA
Mobile information companies will
discuss converging navigation,
location-based services, eet and
asset management for enterprise
markets, online portal, personal
location, and homeland security
markets. <www.gps-wireless.com>
APRIL 28-MAY 1
SPACE WEATHER WORKSHOP
Boulder, Colorado USA
This annual program by the Space
Weather Prediction Center (SWPC)
highlights space weather impacts
on communications, navigation,
spacecraft operations, aviation, and
electric power. <www.swpc.noaa.
gov/sww>
ADVERTISERS INDEX
Company Page Number
Forsberg Services Ltd. 7
European Navigation Conference 2009 14
GPS Networking 25
Hemisphere GPS 27
ICINS 2009 St. Petersburg 37
IfEN 9
Joint Navigation Conference 2009 22
L3/Interstate Electronics Inside Front Cover
NavCom Technology Inc. Inside Back Cover
NavTechGPS 25
NovAtel, Inc. 23 , Back Cover
OmniSTAR 21
Pacic Crest 13
PCTEL, Inc. 10
Racelogic Ltd. 29
Septentrio 5
Spirent Communications 3
Telematics Detroit 2009 47
u-blox 15
Need more information?
Its online!
<WWW.INSIDEGNSS.COM/EVENTS>
JUNE 2-3
TELEMATICS DETROIT 2009
Novi, Michigan USA
Conference and exhibition covering
the digital in-car and mobile
industries. It features senior-
level speakers from Ford, Toyota,
TomTom, Yahoo!, Garmin, Panasonic,
U.S. Department of Transportation,
Visteon, Google, and others.
Inside GNSS readers receive
discounted admission register
online at the url below with code
PARTNER1525
<http://www.telematicsdetroit.com/
register>
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_______
____
___
____
___________
_________________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
__________________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_____________________________________________________

GNSS 20090301 Mar 2009

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

GNSS 20090301 Mar 2009

Uploaded by

Copyright:

Available Formats

Previous Page | Contents | Zoom in | Zoom out | Front Cover | Search Issue | Next Page q

You might also like