GNSS 20090101 Jan 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
4 QUESTION SURVEY
WE INVITE YOUR FEEDBACK!
Dear Reader,
In addition to our print and web versions of Inside GNSS, we will soon offer a full-fledged digital version
of our magazine. We are offering some of our current readers an advance view of the digital version and
welcome your comments in a brief on-line survey.
As you peruse this issue, please consider some of the benefits that a digital edition has to offer:
l Timely Delivery (an announcement of the availability of the digital version of each issue can be
delivered to your email inbox automatically, often before the printed issue arrives in the mail)
l Searchable (search across all available issues)
l Active links (click on underlined words to send an email or visit a Web site)
The following are a few tips for navigating the digital edition once you've opened it:
l To move from page to page, use the forward and back buttons in the toolbar at the top of the screen.
l To enlarge text for easier reading, click once anywhere on the page. Click again to bring the text
back to its original size.
l The Search function allows you to conduct a keyword search within the entire issue, or across all
available issues.
l For offline reading, download a fully functional copy of the digital edition to your hard drive by
selecting the Save function from the toolbar at the top of the screen.
We encourage you to forward this email to friends and colleagues with whom you would like to share
this digital edition.
If you are viewing the Web browser version, just click the Share function on the toolbar.
Please take a moment to complete our four-question survey, which you can access by clicking the link
below. Your participation is greatly appreciated!
Sincerely,
Glen Gibbons
Editor and Publisher
Inside GNSS
January-February 2009
www.insidegnss.com
GNSS FRIEND OR FOE?
First Look at QZSSs Indoor
Messaging System
AUTONOMOUS INTEGRITY:
Computing Protection
Levels for Multiple Errors
GPS & REGIME CHANGE
Part 2:
What Lies Ahead
Get in Line
Flying Satellites in Formation
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
Previous Page | Contents | Zoom in | Zoom out | Front Cover | Search Issue | Next Page
_________
_________________
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 J A NUA RY/ F E BRUA RY 200 9 www.insidegnss.com
TECHNICAL ARTICLES
28 Autonomous
Integrity
An Error IsotropyBased Approach for Multiple
Fault Conditions
Miguel Azaola-Senz and
Joaqun Cosmen-Schortmann
A new technique for computing protection levels
in liability critical applications those in
which large undetected GNSS position errors can
have severe legal or economic consequences.
37 QZSSsIndoor
MessagingSystem
GNSS Friend or Foe?
Andrew Dempster
The IMES positioning and communications sys-
tem developed as part of Japans Quasi-Zenith
Satellite System may be the perfect answer for
seamless ubiquitous positioning or perhaps
a jamming threat to the integrity of GPS.
TABLE OF CONTENTS
JANUARY/FEBRUARY
2009
VOLUME 4
NUMBER 1
TOC BY THE
NUMBERS
8
Thinking Aloud
10
360 Degrees
11
GNSS Hotspots
ARTICLES
12
GNSS World
ICG Round 3
14
GNSS Solutions
GPS Jamming, linear
carrier phase combos
20
GPS & Regime Change
Part 2, What Lies Ahead
28
Autonomous Integrity
37
IMES: Friend or Foe?
43
Working Papers:
Formation Flying, Part 2
DEPARTMENTS
52
Industry View
53
GNSS Inside
54
GNSS Timeline
54
Advertisers Index
FEATURE ARTICLE
20GPS&Regime
Change, Part2
What Lies Ahead
Glen Gibbons
The massive turnover that occcurs when the
executive branch moves from the hands one
political party to another will see the departure
of many ofcials with GPS expertise. Who will
ll the gap and what will that mean for the
nations premier space-based positioning,
navigation, and timing program?
International Committee on GNSS: Round 3
Four steps forward, one step back? The path
to GNSS interoperability is strewn with good
intentions . . . and sausage. See the report
on the latest meeting of the worlds GNSS
providers, beginning on page 12.
w
w
w
.
b
a
r
a
c
k
o
b
a
m
a
.
c
o
m
Cover photo of PRISMA satellite courtesy of
Swedish Space Corporation (SSC) and Intespace
_
_
_
_
_
_
_
_
_
_
_
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
January/February 2009 Volume 4/Number 1
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, L. J. Sellers
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 J A NUA RY/ F E BRUA RY 200 9 InsideGNSS 7
14 GNSSSolutions
What about GPS jamming and maritime safety, and
linear carrier phase combinations?
Mark Petovello with Alan Grant and
Paul Williams
43 WorkingPapers
GNSS in Space, Part 2: Formation Flying Radio
Frequency Techniques and Technology
Thomas Grelier, Alberto Garcia, Eric Pragin,
Laurent Lestarquit, Jon Harr, Dominique Seg-
uela, Jean-Luc Issler, Jean-Baptiste Thevenet,
Nicolas Perriault, Christian Mehlen, Christophe
Ensenat, Nicolas Wilhelm, Ana-Maria Badiola
Martinez, and Pablo Colmenarejo
8 Thinking Aloud
Critical Infrastructure
Glen Gibbons
10 360 Degrees
News from the world of GNSS
Obama Chooses Raytheon VP for DoD
Russia Sends Up GLONASS-M Trio
ESA Names Hein to Galileo Post
12 GNSS World
52Industry View
53GNSS Inside New Products
54Advertisers, Calendar
DEPARTMENTS COLUMNS
_______________________ ____________________________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
8
GLEN GIBBONS, JR.
Editor
S
o, President Obama wants to
spend some money on infra-
structure, eh? Well, heres an
idea: send some of it GPSs way.
Infrastructure isnt just concrete
and rebar. We can also build highways
to the stars and pardon the clichs
bridges to the future rather than
bridges to nowhere.
And talk about bang for the buck.
Te billion dollars or so that the
United States spends on GPS each
year produces many tens of billions of
dollars in products and services.
Of course, a big chunk of that
GPS market is outside of this country.
But afer our recent lamentable
contribution to global fnancial
troubles, perhaps its time to remind the
world about the unprecedented U.S.
generosity in creating an entirely new
public utility and making it available
everywhere.
Not only that, but U.S. policy forced
other GNSS providers to be generous,
too. As the would-be Galileo public-
private partnership discovered, you
cant compete with free.
Anyway, back to Obama and
infrastructure.
Te Global Positioning System
has many unusual, novel, perhaps
even unique features. But the one that
relates to the current topic is that GPS
is both a critical infrastructure in itself
notably its ground control and space
segments and the pervasive, strategic
installation of high-performance
receivers and a contributor to
other critical infrastructures, such
as communications networks or
transportation.
Tat should earn GPS double the
attention, if not twice the budget.
But theres more. GPS not only
allows us to do things that we couldnt
do before; it allows us to do them more
efciently greater productivity at
less cost, whether surveying forest
boundaries or guiding a thousand
airplanes at once.
And though those efciencies
may reduce the job opportunities at
individual enterprises, they stimulate
a far greater amount of job creation
overall design and engineering,
manufacturing, professional feldwork
most of it high-skilled and higher-
paying than the positions that were lost.
Te United States really hasnt had
an industrial policy since just before
and during World War II, when the
Roosevelt administration converted
much of the nations jobless into
public employees (Works Progress
Administration, Civilian Conservation
Corps), its manufacturing sector
into an armament assembly line, and
gasoline and foodstufs into ration
coupons.
Afer that, we saw occasional,
isolated initiatives the interstate
highway system, the lunar missions
of the 1960s, SEMATECH large-
scale infrastructure and technology
programs that could have served as
potential components of an industrial
policy, if one had existed.
GPS can help thread the new
infrastructure eforts together, and
expand the role that it already plays.
Many commercial GPS manufacturers
are looking forward to the
opportunities that building or restoring
highways, bridges, and (imagine!)
maybe even railroads will bring.
But the United States is still running
the GPS program as though we had all
the time in the world. Well, no ofense
to those atomic clocks on board the
GPS satellites (another frst of its kind),
but the world is quickly catching up
with us in matters of GNSS. And, if
we take a close look at the worlds four
GNSS program schedules, over the next
few years just about every other GNSS
system is going to pass GPS by in terms
of signal availability, modernity, and
diversity.
Te United States risks seeing its
GPS brand decline amid the growing
choices in the GNSS marketplace.
Its time that the GPS leadership,
civil and military, revisited its
prevailing philosophy and began
launching for scheduled capability,
rather than as needed to sustain an
aging constellation.
And, while theyre at it, they should
take another look at the size of the
constellation. Every other GNSS system
is committed to a true 30 satellite/30
slot confguration. If the advent of the
biggest infrastructure investment in
American history isnt the right time to
do the same with GPS, when is?
As American poet Edwin Markham
asked on behalf of the man with the
hoe gazing at the ground, Give back
the upward looking and the light/
Rebuild in it the music and the dream
The United States is still running the GPS program
as though we had all the time in the world.
THINKING ALOUD
Critical Infrastructure
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
InsideGNSS J A NUA RY/ F E BRUA RY 2009 www.insidegnss.com
360 DEGREES
IMAGE CREDITS: 2. GRIN-NASA; 3. NASA; 4. WIKIMEDIA CREATIVE COMMONS;
6. LARS PIND, WIKIMEDIA CREATIVE COMMONS.
El Segundo,
California
ON HOLD After
Obama Chooses
Raytheon VP
for DoD DepSec
P
resident Barack Obama has
named William J. Lynn III, a
senior vice-president at Raythe-
on Corporation, as his nominee
for U.S. deputy secretary of defense, a
key post that among other responsibili-
ties includes serving as the co-chair of
the Space-Based Positioning, Naviga-
tion, and Timing (PNT) Executive Com-
mittee (ExCom).
As the number two official in the
Department of Defense (DoD), Lynn will
report directly to Robert Gates, the cur-
rent secretary of defense who will contin-
ue in that position in the new administra-
tion. He succeeds Gordon England, who
has paid a good deal of attention to GPS
during his term in ofce and enhanced
the role of the PNT ExCom as an arbi-
ter and advocate for the GPS program
throughout the federal government.
Gates has been intimately involved
in the process of identifying and inter-
viewing appropriate candidates for vari-
ous vacancies throughout the depart-
ment, according to Pentagon Press
Secretary Geof Morrell.
Lynn returns to the capital as no
stranger to either DoD or Congress.
From 1997 to 2001, Lynn served as one
of fve under secretaries of defense, act-
ing as the departments comptroller and
principal advisor to the Ofce of the Sec-
retary of Defense (OSD) for all budget-
ary and fscal matters.
Another significant DoD position
held by Lynn was that of OSD director of
program analysis and evaluation (PA&E)
from 1993 to 1997, where he oversaw the
DoDs strategic planning process.
Before entering the DoD in 1993,
Lynn served for six years on the staf of
Senator Edward Kennedy as liaison to
the Senate Armed Services Committee.
Although Lynn apparently hasnt
had a close association with the GPS
program, his company does. Raytheon
is the prime contractor for the Federal
Aviation Administrations GPS Wide
Area Augmentation System (WAAS)
and leads one of two competing teams
seeking the contract to build the next-
generation GPS operational control seg-
ment (OCX).
As Raytheons senior vice-president
for government operations and strat-
egy, Lynn is the chief liaison with fed-
eral executive and legislative branches
for a company that does. According to
an Associated Press report, until early
last year Lynn lobbied Congress and fed-
eral agencies on a variety of programs
including missiles, sensors and radar,
advanced technology programs, and
intelligence funding.
Raytheon recently announced that
its OCX team completed the segment
design review and modernized capabil-
ity engineering model demonstration
on Dececmber 13, 2008. Te company is
working under a $160 million Phase A
system design and risk reduction contract
awarded by the GPS Wing in November
2007. A team led by Northrop Grumman
is the other contender for the OCX con-
tract. A fnal decision on the OCX prime
contract is expected later this year.
Russia Sends Up
GLONASS-M Trio
T
he Russian Federal Space Agen-
cy (Roscosmos) successful ly
launched a Proton-M rocket and
three GLONASS-M satellites at
10:53 a.m. (GMT) on December 25 from
the Baikonur spaceport in Kazakhstan.
The spacecraft were placed into
orbital plane 1 on the GLONASS con-
stellation, where they will occupy slots
2, 3, and 8, according to the Roscosmos
Information-Analytical Center. Built
by Reshetnev Informational Satellite
Systems, the modernized GLONASS
space vehicles will join 17 operational
satellites now in orbit. Newly launched
GLONASS satellites are usually brought
online within a month or so.
Changes in the GLONASS program
this year mean that the full constellation
in the future will have 30 satellites rather
than 24, with each of three orbital planes
hosting eight nominal and two backup
satellites.
Te current GLONASS constellation
provides 95 percent coverage of Russian
territory and 83 percent global coverage,
according to Roscosmos and Reshtenev,
with each launch improving these val-
ues.
In 2007, Reshetnev doubled its pro-
duction of GLONASS satellites, accom-
plished in part through introduction of
computer-aided acquisition and logistics
support technologies and 3-D computer
design systems, representatives of the
public company said.
By 2010, Russia will launch the frst
of its GLONASS-K series satellites,
with four of these being checked out
on orbit over the following two years.
A new unpressurized design ensuring
operation of the spacecraf systems in
Oregon, USA
Gas Wars The governor
of Oregon, a socially
innovative state, seeks a
controversial mileage tax
to replace the gas tax on
vehicles. GPS units would
track distance, not traveler
information. But privacy
advocates dont trust that
plan. And conscientious
Greens would get no credit
for driving gas sippers.
Pasadena, California
Sibling Rivalry
The December
2008 International
Committee on GNSS
meeting (ICG-3)
in Pasadena hit
rough waters as
providers negotiated
interoperability.
GNSS leader U.S.
wants to stay there
but cooperatively,
recommitted Russia
wants to expand its
appeal, technically
innovative EU wants
some elbow room, and
China wants to climb far
and fast. (See story on
page 12.)
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
www.insidegnss.com J A NUA RY/ F E BRUA RY 2009 InsideGNSS 11
360 DEGREES
GNSS Hotspots
NewYork, USA
Gift of the Magi A
Manhattan surveillance
company loaned GPS micro-
devices and hidden cameras
to Christian and Jewish
congregations for their
holiday displays in December.
When moved, the bugs set
off an alarmto a cell phone or
email address and the stolen
items could be tracked.
Bottomline? Dont mess with
the Three Wise Guys.
Washington, DC
What Next? The newObama
administration is keeping
Defense Secretary Robert
Gates, but many other GPS
champions are moving on.
Will DoDs newNumber 2
WilliamLynn win priority
listing for the worlds rst
and only fully operational
GNSS as it ghts to stay on
top? (See story on page 20)
Brazil
Grand Theft Auto Brazils
unusually high auto theft
and violent robbery and
hijacking rate has inspired
a newlawto mandate GPS
anti-theft and tracking
devices on all newly-
registered vehicles, as of
August 1, 2009. The required
systems alert a call center
with the vehicle location and
can turn off the car remotely.
South Africa

Its a bird! Its a plane!
Stuck in a remote clinic
during an epidemic with no
way to deliver test samples
to a diagnostic laboratory?
The South African National
Health Laboratory Service has
adapted war technology for
healing. MedicAir Couriers
are miniature UAVs that
followa pre-programmed
route to the lab using GPS and
microelectronic gyroscopes.
GLONASS
Compass
Military
Breaking
Policy
Signal Launch
Other Systems Galileo
GPS
Bright Idea
Technology
Commercial
Consumer
Satellite
History
Glitch
Conference
an outer space environment is expected
to improve satellite survivability while
reducing the size and weight of the SVs.
Te smaller, lighter design will enable
Roscosmos to launch two GLONASS-K
satellites at a time on Soyuz-2 rockets
rather than Protons, reducing the launch
costs by half, according to Roscosmos.
ESANames Hein
toGalileoPost
T
he European Space Agency (ESA)
has appointed Prof. Dr. Gnter
Hein to serve as Head of Galileo
Operations and Evolution with
duties in Noordwijk, Netherlands, and
Paris, France.
With an initial four-year term begin-
ning December 1, Hein will report
directly to Ren Oosterlinck, ESAs
director of the Galileo program and
navigation-related activities (D/GAL).
Nominated for the post by the Ger-
man government, Hein wil l share
responsibility for the evolution of the
Galileo system, the relation and interop-
erability of Galileo and EGNOS to other
GNSSs, preparation of the future opera-
tional phase of the Galileo infrastructure,
implementation the related technology
actions, and associated activities.
Hein took up his new responsibilities
quickly, joining European Commission
representatives on a mission to Beijing
to negotiate with the Chinese Technical
Working Group on Compass regarding
the problem of signals and frequencies of
Galileo and the Chinese GNSS system.
Hein has been a full professor and
director of the Institute of Geodesy and
Navigation (IGN) at Germanys Uni-
versity FAF Munich since 1983. He will
take a leave of absence from the univer-
sity, retaining all the rights as professor.
Bernd Eissfeller, formerly IGNs vice-
director and since 2000 a full professor of
navigation at the University FAF Munich,
is now heading now the institute.
Hein has been a member of the
European Commissions Galileo Sig-
nal Task Force and founded the annual
Munich Satellite Navigation Summit. He
has also been a contributing editor for
Inside GNSS since 2006, coordinating
the Working Papers column.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
12
R
epresentatives from the worlds GNSS and augmen-
tation providers spent a sun-laden week in Pasadena,
California, advancing aspirations for their separate
enterprises while refning the basis for compatibility
and interoperability among them.
Te occasion was the third meeting of the International
Committee on GNSS (ICG-3), a voluntary United Nations
backed association that brings together GNSS and augmenta-
tion providers including the United States, Russia, Euro-
pean Union, China, India, and Japan associate members
representing key user communities.
Afer its formal launch in 2006, the ICG sketched out
a bold vision of cooperation and shared principles in joint
statements of the committee and a Providers Forum during
the groups second meeting in Bangalore, India, in September
2007.
Te third meeting in Pasadena, however, refected a more
cautious approach by most members as they began collective-
ly translating general principles into work plans that could
lay the foundation for technically and operationally aligned
systems without undercutting separate and frequently
divergent goals for individual GNSSs.
As one American delegate observed in an aside compar-
ing the India and U.S. meetings: Four steps forward and one
step backward.
Hosted by the United States, the meeting was chaired by
Ken Hodgkins, director of the U.S. State Departments Ofce
of Space & Advanced Technology, with local arrangements
taken care of by the Jet Propulsion Laboratory (JPL), Cali-
fornia Institute of Technology, and the ofces of the Interna-
tional GNSS Service (IGS) Central Bureau located at JPL.
Te UN Ofce for Outer Space Afairs (OOSA) represent-
ed by Sharafat Gadimova, an OOSA program ofcer, serves
as executive secretariat for the ICG.
Is Interchangeability Possible?
At ICG-3, Brad Parkinson, a Stanford University professor
emeritus and the original manager of the GPS Joint Program
Ofce in the 1970s, set the tone for the meeting in
the opening plenary with an appeal for collaborative
eforts that would move beyond interoperability to
interchangeability. Tis concept, which allow use
of any four [GNSS satellites] anytime, requires a
closer alignment of GNSS system time and coor-
dinate reference systems than currently proposed,
Parkinson said.
Parkinson, who is vice-chair of the advisory
committee to the U.S Space-Based Positioning, Nav-
igation, and Timing (PNT) Executive Committee,
also called for high-level support of earlier delivery
of GPS III satellites, currently scheduled for 2014,
and a more robust constellation of 30 space vehicles
(SVs) in 30 orbital slots, plus spares.
His comments echoed a growing concern at the
possibility of GPS brownouts if a substantial num-
ber of GPS spacecraf failed at the same time, creat-
ing holes in the coverage until new generations of satellites
can be brought online.
Global Market Pressures vs. National
Priorities
Te pressures to advance individual programs create centrif-
ugal forces that threaten to overcome the centripetal impulse
represented by the ICG.
GLEN GIBBONS
Clashing trajectories pressure GNSS providers . . . but growing
commercial markets inspire a search for compatibility at the third
plenary meeting of the International Committee on GNSS.
GNSS WORLD
ICG-3: Friendly Persuasion
The organizing committee of ICG-3: International Committee on
GNSS, third meeting. International GNSS Service Central Bureau
Director Ruth Neilan (rst row, far left) organized the Pasadena
event. Next to Neilan is meeting chair Ken Hodgkins, director of
the State Departments Ofce of Space and Advanced Technology.
Mike Shaw (rst row, fourth from right) directs the National
Coordination Ofce for Space-Based Positioning, Navigation, and
Timing (PNT). OOSA program ofcer Sharafat Gadimova (third from
right) serves as executive secretariat for the ICG.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Compatible & Interoperable
Te most intense discussions at ICG-3 revolved around the
Providers Forum terms of reference, work plan, and work-
ing principles of compatibility and interoperability and their
further defnition. Taking place in open sessions of the ICGs
four working groups and closed Providers Forum meeting,
delegates ground through wording of the documents in a pro-
GNSS providers are acutely aware that GNSS represent
critical infrastructures that can reinforce a nations politi-
cal leadership at the global level. At the same time, however,
GNSS supports commercial industries and markets that
continue to grow steadily even in the current economically
challenged conditions.
Participating in those markets requires an essential com-
patibility with a common technical standard now repre-
sented in an ad hoc way by GPS, but likely to expand into a
more generic form as the other systems are
completed.
Te presence of user communities and
industry representatives at ICG-3, many
as invited speakers, underscored the pres-
sures that government ofcials are feeling
to ensure that individual GNSSs remain
part of the mix incorporated into manu-
facturers designs.
Divergent tendencies also refect difer-
ent priorities currently existing among the
GNSS programs, which might be summa-
rized as follows:
The UnitedStates: maintaining its lead as
the frst and only fully operational GNSS
while ensuring open civil standards and a
level playing feld in international trade
in GNSS products and services; providing
a multilateral forum where representa-
tives can discuss scientifc and technol-
ogy matters that national policy prevents
them from addressing bilaterally (i.e., with
China).
EuropeanUnion: fostering the growth of
European high-tech expertise while pre-
serving its options for managing the sys-
tem and still maintaining momentum to
implement Galileo as globally competitive
GNSS; convincing China to move of of
portions of radio spectrum where Galileos
security-oriented public regulated service
(PR) will operate (not necessarily in that
order).
Russia: Increasing the appeal of
GLONASS to consumer manufacturers
and markets and establishing ground
monitoring facilities outside the Russian
territory; developing its industrial base to
produce domestically designed equipment
for military and commercial applications.
China: Bringing its system online quick-
ly for national security and commercially
competitive reasons, capturing global
mass markets for Chinese-built goods, and
establishing Compass as the worlds pre-
eminent GNSS system.
ICG-3 continued on page 41
________________________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
14
InsideGNSS J A NUA RY/ F E BRUA RY 200 9 www.insidegnss
GNSS
Solutions:
What is the effect
of GPSjammingon
maritime safety?
A
lthough GPS jamming
incidents are relatively rare
they can occur; and when they
do, their impact can be severe.
Te General Lighthouse Authorities
of the United Kingdom and Ireland
(GLAs) comprise the Commissioners
of Irish Lights, the Commissioners
of Northern Lighthouses and Trinity
House, who between them provide
aids to navigation (AtoNs) for the
beneft of all mariners in British and
Irish waters. In order to investigate
the efects of GPS jamming, whether
by intentional or accidental means, the
GLAs conducted a trial in 2008 on the
efect of GPS denial on marine aids-to-
navigation, and ship-borne and shore-
based navigation and information
systems.
Todays mariners commonly
use GPS enabled devices to navigate
their vessels, however large, from
port to port and berth to berth. Te
International Maritime Organization
(IMO) mandates the carriage of
electronic position-fxing systems
by all vessels over 300 gross tons
and those carrying passengers on an
international voyage in accordance
with the Safety of Life at Sea (SOLAS)
convention. Te GPS position is
ofen fed into other vessel systems, for
example an electronic chart display
and information system (ECDIS), the
vessels automatic identifcation system
(AIS), or a plotter.
Te use of diferential GPS (DGPS)
is preferred; mariners improve their
positioning accuracy and ensure
integrity of their GPS derived position
by using the large number of DGPS
radiobeacons located around the
world.
Although GPS receivers for
navigation are commonplace and very
conspicuous on the bridge, the use of
GPS is ofen more inconspicuous in
other AtoN and positioning devices.
Examples include its use for providing
position input to the onboard AIS
transponder, as well as the digital
selective calling (DSC) system, which
has the capability to include the vessels
position as part of a distress signal.
In addition to vessel-based
systems, marine aids-to-navigation
use GPS. AIS timeslots may be
synchronized using GPS as a source
of accurate time. AIS also provides
AtoN position information based on
GPS input. Synchronized lights use
GPS as a common timing source,
and diferential GPS services provide
accuracy and integrity to the mariner.
Terefore, GPS denial, whether
intentional from malicious jamming or
unintentional due to malfunctioning
equipment such as television antennas,
may afect safety both on the bridge
and on-shore.
During the jamming trial, a GLA
vessel was ftted with two typical
marine-grade DGPS receivers, a
survey-grade GPS receiver, and an
eLoran receiver. Te UK Ministry
of Defence assisted in the trial by
providing and operating a GPS
jamming unit.
Te jamming unit transmitted a
known pseudorandom noise code on
the L1 frequency, with an efective
radiated power of 1.5 W. On full power,
using a directional antenna, this unit
was capable of jamming GPS over a
30-kilometer envelope (Figure 1). Te
trial vessel made several runs between
two waypoints.
Each waypoint was positioned
outside the jamming area (including
the areas afected by the main lobe
and side lobes as indicated by the red
and black hatching areas in Figure
1) to allow onboard GPS devices to
GPSjamming
andlinear
carrier phase
combinations
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
reacquire satellites before starting the next run. Each passage
took approximately one hour at a steady speed of 10 knots.
Te results were a mixture of the expected and the
surprising. We had expected that the GLAs trial eLoran
service would not be afected by the GPS jamming unit,
due to the diferent operating frequencies and dissimilar
failures modes, and this was indeed the case. eLoran gave
a consistent position accuracy of 8.0 meters (95%) during
periods of GPS jamming, demonstrating the advantages of
an independent electronic navigation systemwith dissimilar
failure modes to GPS.
Te typical marine-grade GPS receivers did not fare so
well. Both units were forced to operate in stand-alone GPS
mode as the local DGPS reference station was also jammed.
Both receivers reported erroneous positions, ofen indicating
implausibly high speeds and equally implausible position
errors.
Figure 2 details the recorded positions fromone of the
GPS receivers in which each recorded position has been
color-coded depending on the reported speed. Where the
receiver is operating correctly, the resulting positions are
reported as blue. Te efect of GPS jamming can be seen with
the yellow, orange and red positions, which are reporting
erroneous speeds and positions.
Tis fgure also shows a comparison of the worst-case
GPS position with that provided by the eLoran receiver at
the same measurement epoch. It can be clearly seen that
the two reported positions difer signifcantly, being over 22
kilometers apart.
Not only shipborne systems were afected. Te AIS
enables vessels to communicate with other vessels and shore-
based infrastructure to exchange information such as their
call sign, position, destination, estimated time of arrival, and
other pertinent information. Tis information is ofen used
by vessel trafc systems (VTS) on shore to monitor trafc
in and out of port and other waterways. During periods of
GPS jamming the trafc picture can be compromised due to
FIGURE 1 On the left is a screen capture showing the area affected by the
GPS jamming unit (shown on the right) along with the passage between
the waypoints used in the trial, plotted using Meridian SeaTrack soft-
ware.
FIGURE 2 Google Earth plot showing reported position from one of the
GPS receivers during a passage through the jamming zone. Highlighted
is a comparison of the worst-case erroneous GPS position (red circle)
with the corresponding eLoran position (green square). Colors indicate
reported speed: blue <15 knots, yellow <50 knots, orange <100knots,
and red >100knots.
____________________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
16
What are linear
carrier phase
combinations
andwhat are
the relevant
considerations?
L
inear carrier phase combina-
tions are formed by adding or
subtracting carrier phase mea-
surements on two or more fre-
quencies. Such combinations are used
to improve the resulting measurement
in some manner relative to the original
measurements.
In this context, improvement
usually implies removing/reduc-
ing certain errors so as to facilitate
the ambiguity resolution process or
increase the measurement (and, there-
fore, position) precision. We must note,
however, that improvement in both
areas is not possible and thus a design
trade-of is required.
In this solution, we will discuss
how linear carrier phase combinations
are formed and the key considerations
associated with this process. A discus-
sion of some of the common GPS com-
binations is also provided.
Linear Combinations
Denoting the carrier phase measure-
ments (in units of cycles) on frequen-
cies f
1
and f
2
as
1
and
2
respectively,
the linear carrier phase combination,
a,b
, is computed as
where a and b are selectable coef-
cients. Te wavelength of the new mea-
surement is given by
where c is the speed of light, and
1
and
2
are the wavelengths corresponding
with frequencies f
1
and f
2
respectively.
Although equation (1) only consid-
ers two measurements, we could also
add other measurements (e.g., c
3
,
d
4
, etc.). Tese additional measure-
ments are omitted here for simplicity
without detracting from the generality
of what follows.
Before proceeding, frst consider
the equation for a carrier phase mea-
surement made on a single frequency,
f
1
, which, for our current purposes, can
be written as
erroneous positions being reported by
the vessels GPS receivers (Figure 3).
It is also important to note that,
during the trial, shipborne systems
reliant on GPS input failed to maintain
GPS lock and alarmed audibly. Tis
resulted in a higher level of noise on
the bridge lasting several minutes and
also in the denial of some shipborne
systems, such as the vessels ECDIS,
gyro calibration, dynamic positioning
system, and input to the DSC. A lack
of familiarization of the vessels crew
in such situations could clearly afect
their ability to respond, particularly
if an outage occurs while the vessel is
performing a difcult maneuver.
In reply to the original question
posed by the column, the GLA trial
demonstrated that some typical marine
grade GPS receivers can be afected
substantially, reporting erroneous
positions, and implausible speeds.
In addition, other GPS dependent
systems can be adversely afected; the
vessels AIS unit for example. In such a
situation the trafc image for the area,
whether viewed from other vessels or
by shore-based infrastructure, would
be seriously confused.
Particularly important to the
GLAs was the fact that some AtoNs
were afected. DGPS services were
disrupted, AIS AtoNs were afected,
and synchronized lights were also
vulnerable. Tis was to be expected
and now the GLAs are in a position to
be able to identify when GPS denial
occurs and to be able to respond
appropriately.
IMO and the GLAs promote the
use of multiple, dissimilar navigation
systems, just for this kind of event.
As such, the GLAs recommend that
all mariners be familiar with diverse
navigation systems, and are promoting
the use of eLoran as a terrestrial back-
up and complementary system to
satellite navigation.
ALAN GRANT AND PAUL WILLIAMS
AlanGrant is a principal Engineer for the
Research and Radionavigation Directorate of
the General Lighthouse Authorities of the UK
and Ireland. He received the degrees of B.Sc.
and Ph.D. from Staffordshire University and
the University of Wales, respectively. He is
an Associate Fellow of the Royal Institute of
Navigation, a member of the US Institute of
Navigation, and is a chartered physicist
Paul Williams is a principal engineer with the
Research and Radionavigation Directorate of The
General Lighthouse Authorities of the UK and
Ireland. As the technical lead of the GLAs eLoran
Work Program, he is involved in planning the
GLAs maritime eLoran trials and works on a wide
range of projects from real-time differential-
Loran system development to the quality
assurance of Loran additional secondary factor
(ASF) data. He holds B.Sc. and Ph.D. degrees in
electronic engineering from the University of
Wales, and is a chartered engineer, an Associate
Fellow of the Royal Institute of Navigation,
and a board member of the International Loran
Association.
GNSS SOLUTIONS
FIGURE 3 Extract of a vessel trafc image dur-
ing the jamming trial. GPS jamming resulted
in erroneous positions being reported for the
trial vessel Pole Star, giving the impression
that she has sailed across the peninsula.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
where is the geometric range from
the receiver to the satellite, G the com-
bined efect of all geometric errors,
I
i
is the ionosphere error, Nis the inte-
ger carrier phase ambiguity, and n is
the combined efect of the stochastic
errors, namely, measurement noise and
multipath.
Geometric errors are characterized
as being the same magnitude on all fre-
quencies and, in the context of GNSS,
include the troposphere and satellite
orbit errors (receiver and satellite clock
errors can also be included here, but
are normally removed during double
diference processing). Te ionospheric
error is given by
where TEC is the total electron content
(TEC).
Both the geometric and ionospheric
errors are included within the brackets
implying that they are defned in units
of length. Te measurement noise,
however, is outside the brackets and is
quantifed in units of cycles. We will
explain the reason for this distinction
in a moment.
In the context of the foregoing
discussion, three main considerations
need to be borne in mind when deal-
ing with carrier phase combinations:
the integer nature of the ambiguities,
the magnitude of the errors in units
of cycles, and the magnitude of the
errors in units of meters. Each of these
is discussed in detail in the following
sections.
Integer Nature of the
Ambiguities
In order for the ambiguities of the
combined carrier phase measurement
to be integer, the a and b coefcients
must also be integer. Note that this
does not preclude using non-integer
coefcients. In fact, the well known
ionosphere-free combination falls into
this category (which we will discuss
later). Rather, the key is that if non-
integer coefcients are used, the linear
combination cannot be used to resolve
the ambiguities as integers.
Magnitude of Errors
in Units of Cycles
Errors in units of cycles (cycle-
errors) are important because they
have direct implications on the
ambiguity resolution process. Specif-
cally, when cycle-errors are small, the
ambiguity resolution process is more
reliable. Conversely, large cycle-errors
make ambiguity resolution less reliable.
With this in mind, the various
errors are investigated below by sub-
stituting equation (3) into equation (1)
and considering only the error source
_______________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
18
GNSS SOLUTIONS
of interest. With this in mind, the geometric cycle-errors are
given by
where the frequency independent nature of the geometric
errors in used in the frst step, and equation (2) was used in
the last step.
Te importance of this result is that the geometric errors
are scaled by the new wavelength. In other words, selecting
a and b to obtain a longer wavelength reduces the geometric
errors in units of cycles. Tis makes ambiguity resolution
more reliable. Similarly, shorter wavelengths will increase the
errors and make the ambiguity resolution process more dif-
fcult.
For the ionosphere, the error in the linear combination is
where the relation of c = f was used in the third line.
In this case, the efect of the ionosphere is not explicitly
related to wavelength and must be evaluated on the per-com-
bination basis. Instead, the error is expressed as the function
of the error on one of the two input measurements. From the
preceding discussion, if a is set to unity and b is set to -f
2
/f
1
,
then the efect of the ionosphere error is removed altogether.
Tis is the well-known ionosphere-free
linear combination.
Finally, to assess the stochastic errors,
we must employ error propagation. For
this, we assume that the measurement
errors are a function of the wavelength
this is true for both noise and multipa-
th efects and that the errors are white
and uncorrelated between frequencies.
Tis latter assumption is certainly not
valid for multipath efects. Nevertheless,
the assumption simplifes the derivation,
which can, as necessary, be expanded to
consider more realistic multipath efects.
Tat said, with the stated assump-
tions, the stochastic errors are quantifed by their standard
deviation and are given by
Table 1 summarizes the efect of some common GPS lin-
ear combinations in terms of errors in units of cycles. In par-
ticular, the table shows the ratio of the errors for the linear
combination to the corresponding error of the L1 measure-
ment. To put it diferently, the table shows the amplifcation
of the errors relative to L1 (absolute value).
Te beneft of the widelane combination ( 0.86.m)
becomes clear here. In particular, it reduces all of the geo-
metric and ionospheric errors relative to L1, thus simplifying
the ambiguity resolution process. In contrast, the narrowlane
combination ( 0.11 m) increases all errors, suggesting that
it should be avoided unless the errors are small. (Te beneft
of this combination will be investigated in the next section.)
As expected, the ionosphere-free combination removes the
ionosphere error and is therefore still important, for example,
over very long baselines or during solar maximum when the
ionosphere errors are expected to be large. However, we must
remember that the ionosphere-free combination does not
maintain integer ambiguities. Te last combination listed is
an integer-maintaining combination that closely approxi-
mates the ionosphere-free case in terms of mitigating the
ionosphere error, which may be useful in some applications.
It is also interesting to note that in all cases the stochastic
errors are increased. We should expect this because, regardless
of the values of a and b, the sum squared formulation in equa-
tion (6) guarantees an increase in the variance of the errors.
Magnitude of Errors in Units of Length
We can determine the magnitude of the various errors in
units of length by multiplying their cycle-errors by their
wavelength. Errors in units of length are important for posi-
tioning purposes.
Efectively, by scaling the carrier phase measurement to
units of length, one obtains a range that can be used to com-
pute ones position (similar to the pseudorange case). It fol-
a b Common Name Geometric Errors Ionosphere Error Stochastic Errors
1 -1 Widelane 0.22 0.28 1.41
1 1 Narrowlane 1.78 2.28 1.41
1 -f
2
/f
1
Ionosphere free 0.39 0 1.27
4 -3 N/A 1.66 0.15 5
TABLE 1. Amplication of errors in units of cycles (relative to L1) for some common GPS linear combinations
a b Common Name Geometric Errors Ionosphere Error Stochastic Errors
1 -1 Widelane 1 1.28 6.41
1 1 Narrowlane 1 1.28 0.79
1 -f
2
/f
1
Ionosphere free 1 0 3.23
4 -3 N/A 1 0.09 3.01
TABLE 2. Amplication of errors in units of length (relative to L1) for some common GPS linear combinations
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
lows, therefore, that smaller measure-
ment errors will give correspondingly
smaller position errors. Table 2 shows
the same comparison as Table 1, but for
the case when the errors are expressed
in units of length.
Tree things are worth noting.
First, the geometric errors are unaf-
fected. Tis is not surprising because
these errors, by defnition, were invari-
ant to frequency.
Second, the design trade-of dis-
cussed earlier is now evident. For
example, the widelane combination
is shown to increase the measure-
ment error due to the ionosphere by
about 28 percent. (Te noise is also
increased signifcantly, but this is less
of a concern because it can be averaged
out.) Somewhat ironically however,
the motivation for using the widelane
in the frst place is usually because
the errors most notably due to the
ionosphere are too large for reliable
ambiguity resolution of the L1 ambi-
guities directly.
Tird, the narrowlane combina-
tion is shown to have better stochastic
error performance relative to L1 when
expressed in units of length. Tis is
generally considered to be the advan-
tage of this combination. Specifcally,
for short baselines (e.g., for attitude
determination), where the stochas-
tic errors dominate, the narrowlane
combination is preferred. Of course,
this comes at the expense of having to
resolve the ambiguities for a shorter
wavelength a relatively less reliable
approach.
Summary and Outlook
Te foregoing analysis focused on
dual-frequency combinations. How-
ever, with the modernization of GPS
and the upcoming launches of Galileo
and Compass, multiple frequency com-
binations will be possible. Despite this,
the considerations discussed in this
article will still hold and can be used
as a stepping stone for more advanced
combinations and subsequent data
processing.
MARK PETOVELLO
Mark Petovello is an assistant professor in the
Department of Geomatics Engineering at the
University of Calgary and editor of the GNSS
Solutions column.
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 algorithmdevelopment,
inertial navigation, sensor
integration, and software
development.
Email: mark.petovello
@ucalgary.ca
_______
________
_________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
20
GPS & Regime
Change
Part 2: What Lies Ahead
C
hange ahead for the GPS program? Youd better
believe it.
At frst glance, the biggest change will be whos not
there anymore and whos not there yet as Bush legacy
turns into Obama presidency.
As January 20 arrived and the rite of resignation presaged
Inauguration Day, widespread vacancies opened up among
the federal posts most important for GPS policy and top-level
management (see sidebar, Roll Call). And with one notable
exception that of deputy secretary of defense few
replacements had been named.
In itself, that ensures that, in the near term, the GPS pro-
gram will roll on relatively unaltered. Te program wont
be exactly on autopilot indeed, an experienced cadre of
career civil service remain behind the GPS wheel so much
as cruise control.
Even afer the repopulation of federal ofces begins in
earnest, we might expect slow and careful changes, a refec-
tion more of personalities than wholesale changes in policy.
Tats because, as Part 1 of this series documented, the
past eight years have seen considerable consolidation in the
framework of the GPS program.
On the other side of that months-long transition from old
regime to new, however, a formidable set of issues involving
GPS and, more generally, space-based positioning, naviga-
tion, and timing (PNT) await Team Obama: sustainment
of the GPS constellation, the pace of modernization and
changes in acquisition processes, confrmation or revision
of the GPS policy making process, GPSs relationship with
other GNSS systems (and a whole set of associated issues of
interoperability), a PNT architecture plan in the making, and
somewhere out there perhaps a new presidential policy.
Filling the Gaps
Despite the widespread departures, theres no risk that
nobodys minding the GPS store. Career ofcials in the senior
executive service (SES) continue to staf the PNT NCO, key
ofces, and interagency groups responsible for implementing
GPS programs.
Te NCO, under executive director Mike Shaw, has fund-
ing assured through the 2009 fscal year (FY09).
An ex ofcio body, but one rich in GPS expertise, is
the PNT ExCom Advisory Committee led by its chair, Jim
Schlesinger, former head of the CIA and secretary of defense,
and vice-chair, Brad Parkinson, the frst program manager of
the GPS Joint Program Ofce (now GPS Wing). Te commit-
tee was reauthorized late last year and will provide continuity
of information and advice for the incoming deputy secretar-
ies and designees who comprise the ExCom membership.
Other experienced GPS resource people include Joel
Szabat, deputy assistant secretary for transportation policy,
who continues as DoTs highest PNT-related career ofcial
How will the new Obama
Administration, worldwide
economic crisis and conicts, and
the accelerating development
of three other major systems
shape the future of GPS?
GNSS WORLD
GLEN GIBBONS
C
r
e
a
t
i
v
e

C
o
m
m
o
n

-

J
o
e

C
r
i
m
m
i
n
g
s
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
and chaired NCO executive steering group meetings; John
Grimes deputy, Cheryl Roby, and Ray Swider, a program
analyst responsible for PNT policy and planning in ASD/
NII; Jason Kim, senior policy analyst in the Ofce of Space
Commercialization; Ken Hodgkins, director of State Depart-
ments Ofce of Space and Advanced Technologies, and his
deputy, David Turner (former director of the Interagency
GPS Executive Board that preceded the PNT ExCom).
Te GPS Wing, under its commander Col. David Mad-
den, remains the lead acquisition agency for GPS satellites,
the operational control segment, and military user equip-
ment.
Te process of regime change in the federal government
is, indeed, so broad and thorough-going that in 2000 Con-
gress passed a Presidential Transition Act to develop and
deliver orientation activities for key prospective Presidential
appointees and prepare a transition directory laying out the
roles and responsibilities of key government posts.
When the new ofcials do take up their duties, they will
receive what are called transition books, documents pre-
pared by career civil servants that outline the appointees
roles and responsibilities and orient them to their new posi-
tions. On the long list for the DoT and DoD deputy secre-
taries essentially, the chief operating ofcers for those
departments will be cochairing the PNT ExCom. Absent
explicit instructions from above, however, each appointee
from department secretary on down decides which endeav-
ors move to the top of the list.
A NewGPS Action Faction
As Inside GNSS went to press, Obamas appointments to date
have provided little cluse as to wher GPS falls on that priority
list. None of the top appointees appear to have a particularly
deep background on the subject.
More generally, however, the new president has clearly
shown that defense and security issues are as high on his
agenda as economic issues. He persuaded Bushs current
Secretary of Defense Robert Gates, a leader generally well-
respected by both Republicans and Democrats, to continue
in his post. (Te other Republican cabinet nominee for
secretary of transportation is Ray LaHood. More about
him in a little while.)
Whatever the reasons, Gates appointment will bring
much continuity throughout the department. Obamas
transition team authorized Gates to invite a number of DoD
political appointees to remain in their current positions.
Among those was John Young, the under secretary of defense
for acquisition, technology, and logistics.
Around 40 DoD positions require Senate confrmation,
including the deputy secretary, undersecretaries, and assis-
tant secretaries and some deputies. Te remaining appoint-
ments can be made by the defense secretary.
So, now that Gatess go-to guy deputy secretary Gor-
don England is gone, where will the GPS action faction
arise in DoD?
For deputy secretary of defense, Obama chose William
J. Lynn III, senior vice-president for government operations
and strategy at Raytheon Corpora-
tion, who had previously served in a
number of Pentagon posts. (See 360
Degrees, page 10.)
Te nomination drew criticism
from observers who pointed out
another Obama campaign promise
not to appoint high ofcials who
had served as lobbyists within the
previous year. However, the tradition
of regular migrations between DoD
and defense contractors (and back
again) is older than President Eisenhowers 1961 admonition
about the unwarranted infuence of a military-industrial
complex.
Although Lynn apparently hasnt had a close association
with the GPS program, his company does. Raytheon is the
prime contractor for the WAAS program and leads one of
two competing teams seeking the contract to build the next-
generation GPS operational control segment (OCX).
As Raytheons chief liaison with federal executive and
legislative branches, Lynn had advocated for a variety defense
projects. Although his most recent posting to the Pentagon
William J. Lynn III
_____________________________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
22
in civil aviation with programs such as the Federal Aviation
Administrations GPS Wide Area Augmentation System
(WAAS), but also exist in maritime, rail, and perhaps eventu-
ally, highway applications.)
Most observers have acknowledged the relevant back-
ground and high quality of Obamas selections with
notable exceptions. One is Leon Panetta, nominee for CIA
director, a man with deep government experience, but
little in matters of intelligence or security. Te other is Ray
LaHood, a Republican congressman from Illinois who ended
his House sojourn this month.
Obama picked LaHood over several potential Democratic
nominees with transportation experience. Having served
for six years during the 1990s on the House Transportation
Committee, LaHood doesnt draw a complete blank on the
subject. (In contrast to Maria Cino, a Bush deputy secretary
of transportation appointed in 2005 with only a background
in Republican Party fund-raising. Cino disarmed skeptics at
frst staf meeting by taking out her drivers license and say-
ing it was her credentials for the transportation job.)
Nonetheless, past DoT secretaries, such as Norman
Mineta and Federico Pea, were personally familiar with the
GPS role in intelligent transportation systems as well as civil
aviation. LaHood has a much steeper learning curve, should
he even decide to take up the subject of GPS.
Critical Infrastructure, Global Competition
Unlike its performance in many other areas, the Bush
administration did a pretty good job in institutionalizing
GPS.
as comptroller focused more on budgetary and fnancial
matters, he previously served as OSD director of program
analysis and evaluation (PA&E) from 1993 to 1997, where he
oversaw all aspects of the DoDs strategic planning process.
Te PA&E directorate can play a substantial role in shap-
ing the fortunes of defense programs. For instance, the Air
Force recently felded a PA&E inquiry about how design of
the control segment afects the space segment and the
fnancial implications for associated budgets.
So, given his background, Lynn seems as though he could
pick up where England lef of, as a leader champion, even
for GPS.
The Transportation X Factor
A big unknown is the kind of role Obamas secretary of
transportation will play. DoT has inherited a domestic GPS
leadership role among civil agencies largely because of its
regulatory and operational responsibilities for safety-of-life
modes of transportation. (Te latter are particularly apparent
GNSS WORLD
Roll Call
Like the refrain of Chilean singer Patricio Manns anthem to regime
change, Los Caidos (The Fallen), recent weeks have comprised a
litany of decision-makers who are gone:
Deputy Secretary of Defense Gordon England, who elevated the
visibility of GPS within the department of Defense (DoD) and
infused the Space-Based PNT Executive Committee (ExCom) with
initiative and leadership
John Grimes, assistant secretary of defense for networks and
information integration (ASD/NII) and the DoDs CIO, charged
under a 2008 revised DoD directive (4650.05) with overseeing
PNT policy and all aspects of GPS for the Ofce of the Secretary of
Defense (OSD)
Vice Admiral Thomas J. Barrett, USCG (Ret.), deputy secretary of
transportation, co-chair of the PNT ExCom
Department of Transportation (DoT) Acting Under Secretary Tyler
Duvall
Paul Brubaker, administrator of the U.S. Department of Transpor-
tations (DOT) Research and Innovative Technology Administration
(RITA), which is co-authoring the national PNT Architecture project
and is responsible for oversight of the Nationwide Differential GPS
System and other crucial research; also RITA deputy administrator,
Cheryl McQueary
Ed Morris, director of the U.S. Commerce Ofce of Space Com-
mercialization, which hosts the PNT National Coordination Ofce
(NCO) charged with carrying out the ExComs decisions. Morris
recently signed on with ITT Space Systems (see article in Industry
View, page 52)
John Negroponte, deputy secretary of state, whose agency leads
international consultations and negotiations with other nations on
space-based PNT issues a role of growing importance as GNSS
systems proliferate
Paul Schneider, deputy secretary of homeland security, and Robert
Jamison, Department of Homeland Security (DHS) under secretary
for National Protection and Programs Directorate (DHS is charged
with detecting and mitigating domestic interference to GPS and
providing a back-up plan in case interruptions in GPS service).
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
An updated per-
formance specifcation
for the GPS Standard
Positioning Service
(SPS) based on the L1
C/A-code signal that is
in all mass-market con-
sumer GPS products as
well as commercial and
professional equipment.
An active and well-func-
tioning PNT ExCom,
NCO, and Advisory
Committee. A selection
process under way for a
new generation of satel-
lites (GPS III) and opera-
tional control system
(OCX).
Agreements or regu-
lar exchanges with other
GNSS providers, includ-
ing the European Union,
Japan, Russia, and India.
Renewed fers to the
International Maritime Organization and International
Civil Aviation Organization that the GPS SPS would remain
available as a cornerstone for civil users for the foreseeable
future.
But Obamas administration will have to maintain the
momentum.
Challenges abound. As the biennial DoD report on GPS
sent to Congress late last year pointed out, system robustness
and competitiveness must be sustained in the face of aging
satellites and emerging GNSS competitors.
Te GPS constellation is healthy, populated by 30 or so
satellites compared to the specifed full operational capabil-
ity (FOC) of 24, and performing far better than predicted.
Nonetheless, as the DoD report pointed out, of the 31 satel-
lites on orbit at the time of the report, 20 are past their design
life, and 19 are with redundancy in either the navigation mis-
sion equipment or the satellite bus, or both.
Funding for modernization of the space and ground seg-
ments appears secure and, barring further delays in the Block
IIF generation of satellites, sufcient space vehicles should
be available to sustain the constellation. But the report ends
with a caveat: To remain the preeminent space-based PNT
system, it is imperative that modernization of the GPS space-
craf and ground control system continues on schedule.
Tat schedule, however, will bring modernized GPS capa-
bilities to fruition years afer the other GNSS providers have
built modern systems. Galileo and Chinas Compass (Beidou
2) programs both could be fully felded before the frst GPS
III satellite with the modernized L1 civil signal (L1C)l reaches
orbit.
U
.
S
.

A
i
r

F
o
r
c
e

p
h
o
t
o

i
l
l
u
s
t
r
a
t
i
o
n
/
S
t
a
f
f

S
g
t
.

D
o
n

B
r
a
n
u
m
__________
_______
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
24
GNSS WORLD
Markets: Infrastructure Rising
Consumer GPS product sales have taken a much-reported
hit over the past year when consumer spending as a whole
had already turned downward before the September collapse
of fnancial and housing markets. Stand-alone GPS prod-
ucts such as portable navigation devices (PNDs) are coming
under further pressure by the growing popularity of smart
phones as a platform for location-based applications.
On the commercial side of the market, however, the pic-
ture is a lot rosier. And at least one likely presidential initiative
rebuilding Americas infrastructure should have a strong
upside for vendors of higher-priced, high-accuracy GNSS.
Road grading, stake-out surveys, automated machine
control, and a range of other applications that have steadily
become civil engineering best practices should beneft hand-
ily from multi-billion-dollar job-stimulus plans targeting
transportation systems.
Tat initiative will come on top of steady growth in preci-
sion agriculture spurred by high-precision GPS position-
ing that has moved from yield monitoring, through efcient
applications of fertilizers and pesticides, to automated culti-
vation and seeding that eliminates the overlapping rows (and
tedium) of human-guided equipment.
Obamas pronouncements on behalf of green technol-
ogy could beneft the GNSS industry if the operating efcien-
cies made possible by the GNSS are recognized and incorpo-
rated into these economic stimulus programs.
Even oil and natural gas exploration is expected to fuel
steady sales of real-time diferential GNSS services and
equipment over the middle and long term, despite the recent
decline in commodity prices and concerns about reducing
the nations carbon footprint.
Policy: Looking Ahead, Not Back
Te two major eforts at GPS-related policy making are the
1996 Presidential Decision Directive (PDD)/National Science
and Technology Council-6 issued by the Clinton Adminis-
tration and the 2004 National Security Presidential Directive
(NSPD) that superseded it.
If the former was driven by the undermining of selective
availability (SA) by the WAAS program and other widely
available real-time diferential GPS services, then the rise of
Galileo drove much of the urgency for the latter.
Te 1996 PDD reafrmed the dual-use principle and
opened a path for DoD to develop new post-SA solutions to
its security concerns. Te 2004 policy, on the other hand,
sought to respond to the proliferation of GNSS capabilities in
the world.
Galileo, for example, threatened to overlay DoDs planned
M-code signal with an encrypted Public Regulated Service
(PRS). Earlier that year, DoD had gotten involved in the
negotiations under way between the State Department and
EU diplomats that led to a June 2004 agreement on GPS-
Galileo cooperation. Te Pentagons main concern was that a
technical agreement that moved PRS of of the M-code band
Future Waypoints
2009
1st Quarter. Publication of 2008 Federal Radionavigation Plan (FRP).
March 24. Tentative launch date for a modernized GPS Block IIR-M satellite IIR-
20(M) with an experimental L5 signal payload, from Cape Canaveral, Florida
Summer. Launch of IIR-21(M), last of the GPS Block IIR satellites built by Lock-
heed Martin Company
Summer. Publication of an initial National PNT Architecture Transition Plan
3rd Quarter. Next-Generation GPS Control Segment (OCX) Award Phase B contract
October. GPS IIIA satellite program, Key Decision Point KDP-C, Defense Space
Acquisition Board (DSAB), review results of GPS III Capabilities Insertion Program
and possible acceleration of capabilities for subsequent phases
Fall. Launch of rst GPS Block IIF satellite
Fall. Deployment of L2C civil navigation (CNAV) message type 0 on Block IIR
satellites
September 1418. Fourth meeting of International Committee on GNSS (ICG-4) in
St. Petersburg, Russia
Throughout the year: launch of three to four Compass satellites
2010
Critical design review (CDR) and authorization to build OCX
Military GPS User Equipment (MGUE) contract award: Ground-Based GPS
Receiver Application Module (GB-GRAM) [GPS Wing]
First Galileo In-Orbit Validation satellite launch
24 GLONASS-M satellites on orbit. Launch of experimental GLONASS-K spacecraft
Throughout the year: launch of seven to eight more Compass satellites
2011
GRAM Standard Electronics Module Type E (GRAM-S/M) contract award for air
and maritime platforms [GPS Wing]
Flight test of rst CDMA signals on GLONASS; full constellation with on-orbit
spares
2012
Civil Aviation. Design approval for GPS local area augmentation system (LAAS,
known generically as ground-based augmentation system) Cat-III landing
system
OCX Block 1 available
2013
L2C signal initial operational capability (IOC)
OCX Block 2 Available
Fully operational capability (FOC) Galileo constellation completed
2014
M-code (with ex power capability) on 24 satellites
Launch of rst GPS IIIA satellite
Full-rate production, GB-GRAM
2015
Full-rate production, handheld MGUE and GRAM-S/M [GPS Wing]
2016
L2C signals transmitting all data types on 24 satellites
2017
Fielding of DoD modernized military GPS user equipment
2018
L5 signal transmitting on 24 satellites
WAAS approach procedures published to all instrument runways in the National
Air Space
2020
L1C IOC
2021
L1C signal on 24 satellites (FOC)
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
be worked out simultaneously
(rather than subsequently) with
the overall pact.
Te present policy frame-
work, while meeting the imme-
diate needs of GPS programs,
is founded on presidential
directives that essentially were
responding to past events and
external developments. Should
the time come for an Obama
turn at GPS policy, it might ben-
eft from the forward-looking
spin that the new president ofen puts on his initiatives.
One focal point for such a venture could by the National
PNT Architecture that DoTs RITA and DoDs National
Security Space Ofce are shepherding through the bureau-
cratic maze under an interagency agreement. Architecture
plan advocates would like to see the United States achieve a
robust system-of-systems mix of PNT technologies over the
next 15-20 years that would give GPS (and GNSS) a leading,
but far from exclusive role. An efective plan would guide
near- and middle-term investment decisions by agencies with
PNT responsibilities.
Uncivil Agencies, Unequal Equities
And then there is the perennial question of civil equities and
the contribution of civil agencies to support the costs of post-
L5 civil modernizations, as mandated by the 2004 presiden-
tial policy.
Te profusion of GPS activities within civil agencies ofen
produces confusion in intra-agency responsibilities. Tis
fragmentation of activities and constituencies causes confu-
sion and competition. Especially when it comes to money,
civil agencies ofen compare unfavorably with the apparent
unity of DoD GPS interests.
For example, DoD (as well as DHS and the Department
of Veterans Afairs) is operating under a full FY09 budget
approved in legislation last September. As ofen occurs, the
rest of the federal government is operating under a continu-
ing resolution that provides funds at the same level as FY08
through early March 2009.
An immediate complication for civil GPS is that its sched-
uled contribution is $20.7 million in FY09 (from the FAA
budget), but its FY08 level was $7.2 million. Te FY09 legis-
lation provides that Funds available to the Department of
Defense for the Global Positioning System during the current
fscal year may be used to fund civil requirements associ-
ated with the satellite and ground control segments of such
systems modernization program.
________________
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
Although
civil GPS lead-
ers expect an
FY09 budget
or budgetary
relief to be
worked out
over the next
few months,
the situation
underscores
the recurring
dependency under which non-defense agencies labor. (In
contrast, the FY09 funding bill allocated $136 million to the
DoD for GPS procurement and $819 million for GPS-related
R&D.)
One possible alternative would be to assess small fees ($1
or less) on GPS chips or receivers a practice that the Gali-
leo program is seriously considering and use these pro-
ceeds to pay for civil improvements. A fnal topic that might
bear watching is possible restoration of a National Space
Council, as operated under the elder George Bush presidency,
and the implications that might have for the PNT ExCom
process itself.
Whatever the content, an Obama administration might
want to kick the status of the next presidential policy on
GPS up a notch: to an executive order that has more binding
authority on departmental leaders.
PNT = GPS or GPS+
Despite the fact that DoD is a single agency, diferences ofen
arise between the Pentagon as the PNT policy and funding
conduit, the U.S. Air Force as the executive agency for GPS,
and the military services with their difering requirements.
One emerging issue is the timeline for felding modern-
ized military GPS user equipment (MGUE). Some Pentagon
ofcials fear that the GPS Wing will not meet its own sched-
ule for full rate production of MGUE platforms for air, land,
and sea users. (See sidebar, Future Waypoints.)
A 2006 mandate signed by Grimes called for most MGUE
to be available by 2014, when the 24th M-code satellite is
declared operational. One of our greatest challenges is sup-
plying the warfghter with afordable advanced military user
equipment and protecting them from jamming, deception
jamming, and unintentional interference, notes the 2008
report on GPS that went to Congress over Grimess signature.
Another dividing line may be over the question of wheth-
er space-based PNT is efectively just GPS by a fancier name,
or whether it is GPS plus other PNT systems and technolo-
gies, as envisioned by the national Architecture initiative.
Us versus Them
If ambivalence about realizing the dual-use goals of GPS per-
vades the domestic realm, it amplifes itself many times over
in the international realm, where GPS is one among several
GNSSes. Are Galileo, GLONASS, and Compass friends or
foes, competitors or collaborators, helpers or hindrances to
meeting the nations PNT goals?
If the 1996 policy grew out of a world view of a unilateral
U.S. role in GNSS, the 2004 directive stemmed from a clear
understanding that it had become a multi-system world and,
with the rise of terrorism, a darned unsafe one.
Under the recent Bush presidency, the image of the Unit-
ed States in the eyes of many other nations and their willing-
ness to work with it deteriorated markedly. Obama with
his personal popularity overseas (so far) and an internation-
alist perspective wrought by his upbringing may have a
good chance to reverse, or at least ameliorate, that situation.
In the GPS realm, that cause is already well advanced. Te
creation of the International Committee on GNSS (ICG) with
strong U.S. support, including its GNSS Providers Forum
(a development encouraged by the DoD), has established a
forum for multilateral eforts in such areas as compatibility,
interoperability, and common signals. Bilateral initiatives,
ofen revolving around security and trade issues, are solidly
under way with all of the other system operators except
China, where only a few rounds of bilateral talks on signal
compatibility have taken place.
GNSS WORLD
REGIME CHANGE continued on page 51
C
r
e
a
t
i
v
e

C
o
m
m
o
n
s
,

P
h
o
t
o

b
y

P
e
t
e

S
o
u
z
a
______________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Autonomous
Integrity
Within the great effort being spent on improving GNSS integrity-assurance techniques in
order to achieve higher service levels for civil aviation, special attention is being paid to user-
based solutions, as reected in the conclusions of the Federal Aviation Administrations GNSS
Evolutionary Architecture Study (GEAS) panel. This article introduces an autonomous technique
for computing protection levels based on error isotropy that, unlike conventional methods,
adapts in real-time to ranging error size and also handles multiple-fault conditions.
MIGUEL AZAOLA-SENZ AND JOAQUN COSMEN-SCHORTMANN, GMV
An Error IsotropyBased Approach
for Multiple Fault Conditions
Copyright iStockphoto.com/Alexander von Sydow
28
I
n trying to ensure integrity of GNSS
navigation systems for civil aviation,
various approaches have produced a
range of diferent concepts, most of
which assume the computation of a pro-
tection level. Tis computation is usually
accomplished either autonomously (that
is, entirely based on information gath-
ered by the user receiver) or with some
degree of external assistance.
Such information may be provided
by integrity augmentation systems (for
example, space-based or ground-based
augmentation systems SBAS and
GBAS). It may also be provided directly
by the GNSS constellation, as it is fore-
seen with the future GPS III remark-
ably enough, the GPS SPS Performance
Standard already includes integrity per-
formance specifcations and Galileo.
Autonomous protection-level com-
putation techniques, however, have
never been seriously considered as reli-
able sole means for ensuring integrity
in safety-of-life (SoL) applications, not
only because of the poor performances
achieved, but also due to the somewhat
delicate assumptions all of them rely
upon. As a result, such techniques have
mostly been considered as complemen-
tary to external integrity systems. One
example: GPS+receiver autonomous
integrity monitoring (RAIM) is not
allowed as a primary navigation means
for precision approach operations.
Recently, in regard of the improve-
ments on accuracy and reliability
expected from the future constellations
GPS III and Galileo, new approaches
have been proposed for the apportion-
ment of integrity requirements. Tis is
ref lected, for instance, in the conclu-
sions presented in the Phase I report of
the USA GNSS Evolutionary Architecture
Study (GEAS). Te report suggests that
the allocation of the burden for provid-
ing integrity should be balanced towards
the user receiver, thus conferring user-
based integrity (that is, receiver autono-
mous integrity) a higher responsibility.
User-based integrity is also gaining
importance due to the emergence of a
new feld of GNSS applications, the so
called liability-critical applications (i.e.,
those where undetected GNSS large
position errors can generate signifi-
cant legal or economic negative conse-
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
___________
quences). Some leading examples of such
applications are road tolling/congestion
charging (both for highways and city
areas), law enforcement (e.g., speed fn-
ing or surveillance of parolees) or pay
as you drive insurance schemes.
Unlike air navigation, liability-criti-
cal applications ofen take place in harsh
operating environments dominated by
local efects such as multipath. Under
such conditions these applications can-
not always be monitored or aided by
external (global, regional, or even local)
augmentation systems.
Even in civil aviation, some landing
operations could also be subject to large
multipath that could put the navigation
integrity at risk. For those scenarios the
proposed technology would mitigate the
associated risk.
One key assumption of convention-
al RAIM schemes is that simultaneous
faulty measurements are extremely
unlikely. Tis single-fault assumption,
however, fails to hold in a typical liabil-
ity-critical application scenario, where
multipath is the primary source for large
measurement errors and will quite fre-
quently afect more than one measure-
ment at a time. Te single-fault assump-
tion also fails to hold in the future air
navigation scenario, where the large
number of satellites made available by
the joint use of several constellations
(GPS/Galileo/GLONASS) will signifi-
cantly increase the probability of mul-
tiple simultaneous faults.
Other assumptions common to all
existing RAIM schemes include one or
another statistical model of the indi-
vidual measurement errors, trying in
particular to bound the tails of their
distributions. Tis sort of assumption is
somewhat risky and difcult to verify,
especially when the target confidence
level is very high, as in the case of SoL
applications such as civil aviation.
Moreover, under heavy multipath
conditions most statistical assumptions
of this nature just do not hold as errors
caused by multipath are strongly depen-
dent on the geometric characteristics of
the local environment. (An especially
acute example of this is non-line-of-
sight (NLoS) multipath that is, when
a signal is tracked by GNSS equipment
as it refects from some surface despite
the fact that a direct view of the satellite
is occluded by some obstacle.) Hence, it
is almost impossible to come up with a
statistical characterization of such errors
that can be used for integrity monitor-
ing.
In this article we present a novel
technique for autonomous computa-
tion of protection levels, the isotropy-
based protection level concept, or IBPL
for short. This technique makes no
particular assumption on the statistics
of individual measurement errors and
provides coverage against multiple fault
conditions. It takes advantage of a pos-
sible future multi-constellation scheme
as its performance improves rapidly
with the amount of satellites used for
positioning.
Discussion in this article will show
that asymptotic performance of the IBPL
with respect to the number of satellites is
comparable to that obtained with SBAS
protection levels. This fact makes the
IBPL a very promising technique, not
only for liability-critical applications
(the framework where it was born)
but also, and very particularly, for SoL
applications. We believe that IBPL fts
remarkably well in the scheme proposed
by the GEAS panel mentioned earlier,
which recommends a shif of the integ-
rity responsibility towards the on-board
equipment.
IBPL simplies the interoperation of multiple GNSS
constellations for integrity purposes, avoiding the
problemof combining different integrity concepts
_____________
_________________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
30
AUTONOMOUS INTEGRITY
Furthermore, as a fully autonomous
method, the IBPL-based approach does
not require integrity information to be
transmitted on the GNSS or SBAS sig-
nal in space. Tis dramatically simpli-
fes the interoperation of multiple GNSS
constellations for integrity purposes,
avoiding the problem of combining
different integrity concepts from the
various constellations or augmentation
systems.
Autonomous Integrity:
Two Approaches
For liability-critical applications, par-
ticularly in urban areas, local effects
such as multipath especially NLoS
multipath are by far the main source
of errors and, consequently, the main
threat to accuracy and integrity. In this
framework, the conventional notion of
faulty measurement as a large measure-
ment error caused by a satellite malfunc-
tion is no longer useful.
Instead, a large measurement error
is much more likely to be caused by
some local efect (primarily NLoS mul-
tipath) than by a system failure of any
kind. Furthermore, large errors caused
by local efects can be so common that
any efficient detection and exclusion
algorithm would frequently reject too
many observations to allow navigation.
Tis is why we need to consider other
approaches to the integrity problem, at
least for harsh environments.
On one hand, as we have just
explained, the traditional concept of
achieving integrity operates on the prin-
ciple of rejecting faulty measurements.
This is what we will call the measure-
ment rejection approach (MRA), which
works well in open-sky environments.
Focusing on integrity and disregard-
ing its potentially reduced availability,
MRA can also work in other environ-
ments as long as the detection/exclusion
algorithms used do not assume that only
a single fault can occur at one time, and
removal of such assumption is itself a
serious challenge.
Another approach is to character-
ize measurement errors and be able to
compute a protection level that protects
against them, without the need for iden-
tifying and removing degraded measure-
ments, even if they are contaminated
with very large errors. Tis is the error
characterization approach (ECA). In an
ECA implementation, as soon as mea-
surement errors increase or decrease, so
does the computed protection level. No
matter how large the error of a measure-
ment is, that measurement will be used
for navigation, but the computed protec-
tion level has to account for it so as to
keep bounding user position error.
Both approaches can lead to the
same level of integrity, the trade-off
being a matter of protection level sizes
and their associated availability. Service
availability, understood as the probabil-
ity that the navigation system is usable
for a particular application at a particu-
lar time, requires that:
A navigation solution exists with an
associated protection level.
Te protection level is small enough
to suit the requirements of the par-
ticular application.
Te frst requirement conficts with
the MRA, whereas the second one con-
ficts with the ECA.
In open-sky environments, where
local efects are rare and small, both the
MRA and the ECA can yield quite the
same availability performance. However,
in harsh environments, the MRA alone
is clearly insufcient, and that is why the
authors developed the ECA concept and
the IBPL as the key for a successful ECA
implementation.
However, when the authors looked
back to the open-sky environments in
which both approaches are expected to
work almost as well, the IBPL proved to
be such a powerful technique as to sug-
gest its use for SoL applications (of which
civil aviation is probably the clearest
exponent).
IBPL: the Concept
Te IBPL algorithm does not implement
measurement rejection techniques but
rather computes a protection level based
on the all-in-view least squares solution.
Of course, other IBPL solutions are pos-
sible, for instance, when diferent subsets
of measurements are used and the one
with smallest IBPL is selected. However,
in its simplest form (as described in this
article), this algorithm is a strict ECA
concept implementation. On the other
hand, this does not exclude the possibil-
ity that some refnements can be made
for open-sky applications by including
some kind of fault detection/exclusion
mechanism.
Te idea for the basic IBPL algorithm
is to use the vector of least squares esti-
mation residuals (or the residual vector)
as a characterization of the position
error: the larger the residual vector, the
larger the state estimation error vector
(from a statistical perspective). Te rela-
tion between both is taken to be linear;
so, the protection level depends linearly
on the size of the residual vector.
Of course, the state estimation error
also depends on the dilution of preci-
sion (DOP); so, if we are interested in
a horizontal protection level (a vertical
protection level would be obtained anal-
ogously), we would compute it as:
where r is the least squares residual vec-
tor and k is the proportionality constant
that relates the residual size with the
state estimation error.
Tis constant k, which depends on
the target confdence level 1 of the
protection level as well as on the num-
ber of measurements used for the esti-
mation, is called the isotropic confdence
ratio (ICR). It is defned so as to ensure
that the state estimation error (or, more
precisely, its image in the measurement
space through the observation matrix
H) is bounded by the size of the resid-
ual vector up to the ICR and the target
confdence level 1 , according to the
formula:
The IBPL algorithm does not implement measurement
rejection techniques but rather computes a protection
level based on the all-in-view least squares solution.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
For computing k from the preced-
ing relation, we assume that the mea-
surement error vector has an isotropic
distribution in the measurement space
(and, hence, the name for the constant
k). Tat is, the error vector can point in
any direction of the measurement space
with the same probability. Note that
this does not imply any particular dis-
tribution of the individual measurement
errors (e.g., Gaussianity), nor that they
are unbiased or have known variance.
Individual errors can be arbitrarily large
or biased as long as they defne an error
vector that has the same a priori prob-
ability of pointing in any direction of the
measurement space.
Notice that, when conventional navi-
gation and RAIM methods assume cen-
tered Gaussianity and independence
of errors, isotropy holds as a trivial con-
sequence; so, isotropy is less a stringent
assumption than those taken in conven-
tional techniques.
Isotropy implies that the pointing
direction of the error vector defines a
uniform distribution in the unit N-1-
dimensional sphere of the N-dimen-
sional measurement space. On the other
hand, the condition is a
condition on the ratio between the sizes
of the vectors and r, which happen
to be orthogonal in the measurement
space. Therefore, the said condition
defnes a region of the sphere, the area
of which depends on k.
Hence, in order to compute the ICR
for a given confdence level 1 , it suf-
fces to impose that the area of the region
so defned is a fraction of the total area
of the sphere. Tus, if the whole sphere
represents all possible pointing direc-
tions for the error vector, and hence has
probability 1, the region so obtained rep-
resents a set of possible pointing direc-
tions with probability .
Note that the dimension of the mea-
surement space is equal to the number of
measurements; so, we are dealing with a
multi-dimensional sphere, and the result
depends on the dimension. Hence, the
value of the ICR depends on both and
the number of measurements, N:
In solving this problem, one comes
across multidimensional integral equa-
tions that can be numerically solved for
a pre-selected set of values of and N
and tabulated for faster real time perfor-
mance. Table 1 presents an example of
this. Similar tables can be computed for
diferent navigation modes; for instance,
in the case of relative or kinematic navi-
gation, where the clock parameter gets
removed from the problem by double-
diferencing measurements of both the
base and the rover receivers, there are
only three parameters to estimate, which
yields a slightly diferent table.
Note from Table 1 the high sensitiv-
ity of the ICR to the number of mea-
surements. Observe, for instance, that
for = 10
-7
the value of k drops from
1.510
7
to 4.410
3
just by passing from
5 to 6 measurements. Tis means that
when measurement redundancy is low,
residuals constitute much less reliable a
measure of position errors. (In particu-
lar, although not made explicit in Table
1, with four measurements the ICR
would grow to infnity, because no fnite
bound can be guaranteed in the absence
of measurement redundancy, regardless
of the target confdence level.)
As a consequence, availability perfor-
mance becomes poor in environments
with reduced visibility of the sky, such
as urban areas, as the isotropy-based
protection level tends to be very large.
However, this type of protection level
exhibits a great performance in open-
sky areas.
Remarkably, the IBPL concept pro-
vides protection against simultane-
ous multiple faults, because it does not
depend on the number of faulty mea-
surements occurring at the same time
nor on the sizes of their errors: simulta-
neous faults combine to produce a cer-
tain measurement error, but no a priori
privileged directions exist for the mea-
surement error vector to point to; hence,
the isotropy assumption is not violated
by multiple fault conditions.
Validating IBPL Integrity
Of course, we need to validate this new
protection level concept and its underly-
ing isotropy assumption in terms of the
achieved integrity, and that must be done
by experimentation with real data. We
have to show that the theoretical con-
fdence level of the isotropy-based pro-
tection level is satisfed in real life. Our
discussion here cannot be considered as
a full validation of the IBPL concept, but
it provides signifcant information about
the validity of the proposed algorithms.
For that purpose it is very convenient
to be able to compute protection levels
for various confidence levels espe-
cially for low ones, say, from 1-10
-4
down
to 1-10
-1
because much smaller mea-
surement campaigns are required (in
order to have a representative statistical
sample) than for a high confdence level
such as 1-10
-7
(required for civil aviation
applications).
N \ 10
-1
10
-2
10
-3
10
-4
10
-5
10
-6
10
-7
5 14.94 150.0 1500.0 1.510
4
1.510
5
1.510
6
1.510
7
6 4.30 14.09 44.70 141.42 447.21 1414.21 4472.13
7 2.67 6.19 13.52 29.22 62.98 135.72 292.40
8 2.03 4.00 7.31 13.11 23.37 41.60 74.00
9 1.68 3.02 4.99 8.03 12.80 20.33 32.25
10 1.46 2.47 3.82 5.74 8.51 12.55 18.46
11 1.30 2.12 3.13 4.49 6.32 8.85 12.35
12 1.18 1.87 2.68 3.71 5.04 6.79 9.10
13 1.09 1.69 2.36 3.18 4.20 5.50 7.16
14 1.02 1.56 2.12 2.80 3.62 4.64 5.90
15 0.96 1.44 1.94 2.52 3.20 4.02 5.02
TABLE 1. Example of ICR pre-computed values
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
32
Experimental validation has been
driven based on an open-sky data set
28,800 epochs long of real GPS measure-
ments from December 17, 2007, collected
at the International GNSS Service (IGS)
station Villafranca, Spain. (Further vali-
dation with much larger data sets is fore-
seen in the near future.) Integrity results
are summarized in Table 2, which shows
the horizontal misleading information
(MI) event rates obtained in both sce-
narios for diferent values of . (An MI
event occurs when the protection level is
exceeded by the position error.)
As can be seen in Table 2, the MI
rates are very close to their theoretical
values (given by the corresponding ).
The reader will notice, however, that
these results show some degree of con-
servativeness. Tis is due to the fact that
DOP is a scalar measure of the ratio of
a vector transformation, given by the
matrix H, between the measurement
space and the state space and, as such,
DOP is conservative.
The definition of the IBPL can be
refned to account for this efect in order
to obtain tighter MI rates. Apart from
that, these results represent a promis-
ing confrmation of the integrity of the
method and, therefore, of the validity of
its underlying isotropy assumption.
Although demonstrating perfor-
mance in an urban environment is not
the aim of this article, it is worthwhile
to mention that parallel tests with urban
data have been carried out with similar
integrity results, thus confirming the
robustness of IBPL against multiple fault
conditions.
IBPL Performance Results
From the same open-sky test run for
IBPL integrity validation we derive per-
formance fgures in the form of accumu-
lated histograms of protection level sizes.
Figure 1 represents an IBPL accumulated
histogram that, for each possible size
between 0 and 200 meters, represents
the relative frequency at which IBPL has
occurred below that size throughout the
test. Te histogram accounts for 28,800
samples (just as many as the number of
measurement epochs) and has diferent
curves, each representing a different
confdence level 1 from 1-10
-1
up to
1-10
-7
.
Observe that IBPL accumulated his-
tograms constitute a good way to fgure
out service availability (for whatever type
of application the IBPL could be used) as
they provide information both on IBPL
sizes and on associated frequencies.
Observe also from the figure that
IBPL availability is not far from that
achieved with conventional RAIM pro-
tection-level computation techniques,
although IBPL is not relying on the
single fault assumption nor on assumed
measurement error size statistics. RAIM
assumptions on error sizes tend to be
conservative to compensate for certain
model weaknesses, such as the fact that
real-life errors do not follow a Gaussian
distribution. Tis is partially compen-
sated by the single-fault assumption on
which RAIM relies; so, at the end they
produce quite a similar result.
We must, however, consider two key
diferences:
RAIM-based protection levels rely
on more stringent assumptions,
e.g. Gaussianity with knowledge of
the covariance matrix or absence of
multiple fault conditions.
Isotropy-based protection levels
adapt to each situation, changing
their size to account for measure-
ment quality and the number of
satellites, which makes them much
more suitable for the forthcoming
multi-constellation scenario.
Te signifcant availability improve-
ment made possible by IBPL appears
when the number of satellites used for
navigation is big, as with the combined
used of two or more constellations (e.g.,
GPS + GLONASS + Galileo). Starting
with the same real-data set used in Fig-
Horizontal IBPL Accumulated Histogram
1
0.8
0.6
0.4
0.2
0
A
c
c
u
m
u
l
a
t
e
d

F
r
e
q
u
e
n
c
y
0 50 100 150 200
Horizontal IBPL (meters)
1e-1
1e-2
1e-3
1e-4
1e-5
1e-6
1e-7
FIGURE 1 IBPL availability in open sky, GPS only FIGURE 2 IBPL open sky availability, GPS + GLONASS/Galileo (extrapola-
tion)
10
-1
10
-2
10
-3
10
-4
Horizontal
MI Rate
0.088 0.0070 0.000417 0
Vertical
MI Rate
0.055 0.0036 0.000069 0
TABLE 2. Horizontal Misleading Information Results
1
0.8
0.6
0.4
0.2
0
A
c
c
u
m
u
l
a
t
e
d

F
r
e
q
u
e
n
c
y
0 10 20 30 40 50
1e-1
1e-2
1e-3
1e-4
1e-5
1e-6
1e-7
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
ure 1, we have extrapolated the IBPL
behavior with a second constellation
(which could be considered to be either
GLONASS or Galileo). IBPL extrapo-
lation is made by assuming twice the
number of satellites as with only GPS.
Note that the dilution of precision is
kept the same as in the GPS-only case,
because the second constellation is not
simulated but only a doubled number of
satellites assumed for the computation of
the ICR. (More details about the extrap-
olation technique will be discussed in
the following section.) The results are
plotted in Figure 2, in which the scale
of the horizontal axis has been reduced
with respect to Figure 1 for clarity.
The improvement in performance
using two constellations is clear when
compared with GPS alone represented
by Figure 1. Performance is also compa-
rable to current SBAS protection levels.
(One should expect improvements of
SBAS protection levels associated with
a multi-constellation scenario to be
caused mainly by a decrease in the DOP,
which would also imply an improvement
of IBPL with respect to the ones shown
here.)
Recall that Figures 1 and 2 corre-
spond to a clear sky environment. IBPL
size in other, more aggressive scenarios,
such as an urban environment, would
increase due to:
larger measurement residuals caused
by larger measurement errors,
because of local efects such as mul-
tipath (especially NLoS multipath)
decreased satellite visibility caused
by obstacles (e.g., buildings) that
partially occlude the sky view.
Figure 3 shows an example of IBPL
multi-constellation performance in an
urban canyon, which is an extrapola-
tion based on a real GPS data set (with
46,881 samples) obtained with a car
driving through Salamanca quarter in
Madrid. Te extrapolation method is the
same one used for the open-sky example
(Figure 2).
Since the size of IBPL depends lin-
early on the size of the residuals, and
this in turn depends linearly on the
size of the errors, all results presented
herein would be substantially improved
with the accuracy increase that is
expected from the evolution of GPS and
GLONASS systems, as well as from the
future Galileo.
Asymptotic Convergence of
IBPL to SBAS PL
Another remarkable property of the
IBPL concept is its convergence to the
defnition of PL currently used in SBASs
(see Annex J of RTCA/DO-229D cited
in the Additional Resources section near
the end of this article). Roughly speak-
ing, the SBAS defnition of PL is:
where K is the percentile 1 of the
centered chi distribution. If the hypoth-
eses assumed in RTCA/DO-229D held
(namely, centered Gaussianity with
known standard deviation ), then the
size of the residual vector of the least
squares position solution would satisfy:
On the other hand, the isotropy-
based PL is:
It can be shown by means of Lam-
berts W-function that:
This proves that the IBPL concept
converges to the SBAS PL concept when
the number of measurements grows.
As an illustration of the preceding
theoretical discussion, the extrapolated
IBPL curve for 1-10
-7
confidence level
(from Figure 2) has been compared with
the corresponding SBAS protection level
curve. Te results are presented in Figure
4, which shows, along with the extrapo-
lated IBPL histogram, the corresponding
SBAS protection level histogram.
Both curves correspond to the same
data set as that used for integrity valida-
tion in the earlier section on validating
IBPL integrity (i.e., real GPS measure-
ments collected at IGS station Villafran-
ca). For the SBAS protection levels we
used that same data set together with
real EGNOS messages broadcast dur-
FIGURE 3 IBPL availability in deep urban canyon, GPS + GLONASS/Galileo
(extrapolation)
FIGURE 4 Extrapolated horizontal IBPL vs. horizontal SBAS PL
1
0.8
0.6
0.4
0.2
0
A
c
c
u
m
u
l
a
t
e
d

F
r
e
q
u
e
n
c
y
0 50 100 150 200
1e-1
1e-2
1e-3
1e-4
1e-5
1e-6
1e-7
Horizontal IBLP vs SBAS HPL
1
0.8
0.6
0.4
0.2
0
A
c
c
u
m
u
l
a
t
e
d

F
r
e
q
u
e
n
c
y
0 10 20 30 40 50
Horizontal Protection Level (meters)
Real SBAS HPL
Extrapolated Horizontal IBPL
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
34
ing the time interval to which the data
set corresponds; so, both curves are fully
comparable. Note, however, that this is
an example based on a particular data
set and intended only to illustrate the
preceding theoretical discussion; other
real-life examples could lead to slightly
diferent results.
Because the IBPL extrapolation
assumes twice the number of satellites as
with only GPS, it takes at each measure-
ment epoch the ICR that would corre-
spond to twice the number of measure-
ments used for position computation,
but leaves the DOP untouched. We have
to take into account the fact that using
more measurements tends to increase
the size of the residual vector. Assum-
ing that the size of the residual vector
statistically depends on the number N
of measurements according to the factor
, and that the noise levels of both
GPS and Galileo measurements are the
same, the residual vector sizeincrease
factor that results from using 2N mea-
surements (instead of N) is:
Of course, this has to be understood
as a mere approximation that relies
on several assumptions. However, the
extrapolated IBPL values that result
should not difer too much from the ones
that will be obtained with an additional
GNSS constellation. Even better results
would be expected with the combined
use of GPS, Galileo. and GLONASS.
About the Isotropy
Assumption
Once the isotropy assumption has been
accepted, the level of integrity achieved
with the IBPL concept can be proven
mathematically, and is therefore incon-
trovertible. Te only controvertible point
of the method is the isotropy assump-
tion itself, or, more precisely, the extent
to which this assumption represents the
real world.
Te frst thing to notice is that isot-
ropy is a condition on the behavior of
a vector (the vector comprised by the
individual errors of the diferent mea-
surements) that lies in the measure-
ment space. Measurement space is a
vector space whose dimension equals
the number of measurements used in
the computation of a navigation solu-
tion and must not be confused with the
geometric three-dimensional space that
contains the position we want to esti-
mate (nor with the four dimensional
space associated with the 3-D position
and the clock).
Hence, the isotropy assumption has
no relation with satellite line-of-sight
geometry (at least not a direct one,
though there are some subtleties about
this idea that we will discuss later on).
Te pointing direction of the error vec-
tor depends only on the values of the
diferent measurement errors, regardless
of the geometric confguration of their
lines of sight. So, we can imagine two
completely diferent geometric confgu-
rations of the same set of satellites, but
if each satellites measurement error was
the same in both confgurations, then
the corresponding measurement error
vectors would also coincide, both thus
pointing in the same direction within
the measurement space.
Isotropy means that the error vector
has no privileged directions in the mea-
surement space toward which to point.
One could argue that the coordinate axes
of the space are in fact privileged direc-
tions, as each one is used to represent
the measurement error of one individual
satellite. So, it may seem that a measure-
ment error in a particular satellite would
make the measurement error vector to
align somewhat with the corresponding
axis. However, measurement errors do
not come alone: each measurement has
its own, and the key point is how they
combine to form the measurement error
vector. Usually, errors combine random-
ly to produce an error vector that can
point randomly in any direction.
Let us consider for a moment the
classic assumption that measurement
errors distribute normally, with null
mean and some known variance, com-
mon to all satellites. Te multivariate
statistical distribution of all errors (that
is, the distribution of the measurement
error vector) is then isotropic. To visu-
alize this, just observe that the associ-
ated 1-sigma hyper-ellipsoid is actu-
ally a hyper-sphere (and this is a good
example to remark the independence
of isotropy with respect to line-of-sight
geometry).
Isotropy even holds when we allow
diferent variances for diferent satellites
(after the normalization process that
transforms the weighted least-squares
problem back into an ordinary least-
squares one). Terefore, isotropy is, at
the very least, a less restrictive assump-
tion than the usual Gaussian statistical
models assumed in most RAIM tech-
niques.
However, the authors believe that
isotropy is more than just a less strin-
gent assumption. It is instead actually
representative of the real world, because
even when the axes of the measurement
space could be privileged lines, they are
also free to point to any direction in the
measurement space themselves.
To understand this point, let us recall
that the important thing about the head-
ing direction of the error vector is that
it determines the relation between the
two orthogonal components, H and
r, of the error vector. So, as the image of
H (as a vector subspace of the measure-
ment space) changes its orientation with
respect to the axes of the measurement
space, the axes change their pointing
directions relative to the image of H.
Therefore, the relation between H
and r changes even if the error vector is
bound to a particular axis.
In the preceding discussion, how-
ever, we have made use of the change of
orientation of the image of H within the
measurement space, and that is directly
linked to variations of the satellite line-
of-sight geometry. In this sense, the
isotropy assumption is related to satel-
lite geometry.
The IBPL method does not require, in principle,
any ground monitoring [...] thus simplifying ground
segment design.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Another subtle reason suggests why
satellite geometry can indirectly afect
the isotropy hypothesis: although we
have remarked that the pointing direc-
tion of the measurement error vector is
independent of the satellite geometry
(but is determined only by the values of
the measurement errors), this indepen-
dence would be lost if some correlation
existed between satellite geometry and
measurement errors (e.g., larger errors
at lower elevations). However, this prob-
lem can be overcome by modeling such a
geometric correlation of errors by means
of weighting strategies like those used
in all other integrity provision schemes
(either autonomous or augmentation-
based).
What we claim is that the isotropy
assumption is a solid one when we con-
sider the overall statistics, involving all
possible geometries and letting time run
forever. Tat is, if we are given a suf-
ciently large set of samples from real life,
each of them taken randomly (randomly
also in what concerns satellite geometry),
the MI rate obtained from such a sample
with the IBPL would be consistent with
the theoretical confdence level used to
compute said IBPL.
It is a matter of fact that civil aviation
requirements do not allow us to aver-
age satellite geometries when evaluat-
ing integrity performance; so, in order
to satisfy these requirements we would
need isotropy to hold for each single
geometry. Tat is, we would have to show
that, for each possible satellite geometry,
the real-life statistical behavior of rang-
ing errors associated to that particular
geometry is isotropic. Tat is not only
difcult to prove, but possibly false.
Te authors are aware that this is a
potential issue for the application of the
IBPL to civil aviation, and further work
is being carried out in order to overcome
this problem. We think it worthwhile to
remark, however, that the issue of aver-
aging geometries is less critical for liabil-
ity critical applications, in which mul-
tipath is the predominant error source
and the motion relative to the obstacles
causes satellite geometry to change very
quickly and quite randomly, even for a
fxed route.
We have centered our discussion on
the isotropy assumption on the faulty
case, because in fault-free conditions it
does not seem reasonable to think that
the axes of the measurement space are
privileged directions (at least, it is clear
that faulty conditions are a much bigger
challenge for the isotropy assumption).
To illustrate this discussion, we will
include an example of IBPL response to
a real-life fault condition. The failure
took place December 25, 2005, when
the on-board clock of GPS satellite PRN
25 started drifing anomalously at 21:04
(GPS time). At 21:25 the satellite was frst
f lagged as unhealthy in its broadcast
navigation message, and by that time,
the satellites clock had drifted about
100 meters away from the value report-
ed in its broadcast navigation message,
causing large position errors for many
users.
We have used RINEX data from IGS
station at Villafranca del Castillo (Spain)
in order to test the reaction of the IBPL to
such an event. Te resulting horizontal
position error (in red) and IBPL (in blue)
are plotted over time (from 21:00:00 to
21:59:59) in Figure 5.
Observe that we have depicted IBPL
at 1-10
-2
confdence level, since it is closer
in size to the position error than at high-
er confdence levels and therefore makes
the picture clearer. In particular, this
treatment makes it easy to see that the
IBPL evolution mimics that of the posi-
tion error, keeping it properly bounded
at any time during the fault condition.
Because we are claiming the ability
of IBPL to handle multiple fault condi-
tions as well, we have tried to fnd a real-
world example. However, even though
we know that some of them exist, we did
not manage to fnd any RINEX fles that
refect such multiple (even double) fault
conditions. So, we decided to simulate
our own, taking advantage of the pre-
ceding single fault example.
For that purpose we manipulated
December 25, 2005, incidents RINEX
HPE vs Horizontal IBPL under a single failure condition
(real data from IGS station VILL, 12/25/2005, 21:00:00 - 21:59:59)
180
160
140
120
100
80
60
40
20
0
M
e
t
e
r
s
7.55 7.6 7.65 7.7 7.75 7.8 7.85 7.9 7.95
x 10
4
GPS Second of Week
HPE
HPL(alpha = 0.01)
HPE vs Horizontal IBPL under a double failure condition
(simulated data)
250
200
150
100
50
0
M
e
t
e
r
s
7.55 7.6 7.65 7.7 7.75 7.8 7.85 7.9 7.95
x 10
4
GPS Second of Week
HPE
HPL(alpha = 0.01)
FIGURE 5 Horizontal IBPL vs. horizontal position error under a single
failure
FIGURE 6 Horizontal IBPL vs. horizontal position error during a double
failure
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
36
fle to include a clock drif for PRN 24
similar to that of PRN 25 originally pres-
ent in the fle. Te results are depicted
in Figure 6. As in the single failure case,
IBPL is always bounding position error
during the period of satellite failures.
Conclusions
The isotropy-based protection level
concept arose as the result of investi-
gations concerning GNSS liability criti-
cal applications, in particular in urban
environments. The authors found,
however, that this notion also shows a
great availability performance in open-
sky environments and could therefore
become a major breakthrough in open-
sky SoL applications such as civil avia-
tion. Isotropy-based protection levels
are completely autonomous, are easily
computable in real time, and rely on a
single, quite verisimilar and verifable
hypothesis.
Unlike other approaches for integ-
rity being defned as part of the GEAS
initiative, the IBPL method does not
require, in principle, any ground moni-
toring, though detection and exclusion
of faulty satellites by the ground seg-
ment would help guarantee isotropy,
leaving the protection level computa-
tion to the user through the IBPL
and thus simplifying ground seg-
ment design.
IBPLs sensitivity to the number of
satellites becomes a clear advantage in
open sky. With currently no more than
10 satellites in view on average (GPS
only) and 20 or even more when con-
sidering either GLONASS or the future
European Galileo system, this PL con-
cept will predictably yield great perfor-
mances, with smaller protection levels
than those achieved nowadays by exist-
ing SBASs such as the U.S. Federal Avia-
tion Wide Area Augmentation System or
the European Geostationary Navigation
Overlay Service.
Manufacturers
Open sky real data have been down-
loaded through the Internet from the
IGS data server in the form of observa-
tion and navigation RINEX fles record-
ed at the IGS station at Villafranca del
Castillo (Spain), which operates an
Ashtech Z-XII3 receiver. Urban can-
yon data were collected by the authors
in Madrid (Spain) by means of a SiRF-
star II receiver, SiRF Technology, San
Jose, California, USA. Data processing
algorithms were developed by GMV
Aerospace and Defence S.A. using Mat-
lab from Te Mathworks, Inc., Natick,
Massachusetts, USA.
Additional Resources
[1] Cosmen-Schortmann, J., and M. Martnez-
Olage, M. Toledo-Lpez, and M. Azaola-Saenz,
Integrity in Urban and Road Environments and Its
Use in Liability Critical Applications, Proceedings
of the Position Location and Navigation Sympo-
sium (PLANS) 2008, Monterey, California, May
68, 2008
[2] GNSS Evolutionary Architecture Study, Phase I
- Panel Report, February, 2008, available on-line
at <http://www.faa.gov/about/ofce_org/head-
quarters_offices/ato/service_units/techops/
navservices/gnss/library/documents/media/
GEAS_PhaseI_report_FINAL_15Feb08.pdf>
[3] RTCA Inc., RTCA/DO-229D, Minimum Opera-
tional Performance Standards for Global Posi-
tioning System/Wide Area Augmentation System
Airborne Equipment, 2006
Authors
Miguel Azaola-Saenz has
an M.S. in mathematics
from the University
Complutense of Madrid
and a Ph.D. in geometry
and topology from the
University of Cantabria
(at Santander), Spain. He has worked at GMV since
2001 as an engineer in several GNSS projects
related to both the ground and the user segments.
Since 2005 he has been involved in user segment
R&D activities at engineering and management
levels.
Joaqun Cosmen-Schort-
mann has an M.S. in
aerospace engineering
from the Polytechnic
University of Madrid. He
has worked at GMV since
1986. In 1995 he became
responsible for GNSS business in GMV with a very
active role in the development of EGNOS and
Galileo systems. He is currently the CEO Advisor
for GNSS matters.
______________________
__________________________
__________________________
________________________
_______________________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
J
apans Indoor MEssaging System
(IMES) promises to provide seam-
less indoor and outdoor position-
ing with no extra hardware for a
GPS-enabled phone, receiver, or other
portable device. But what is it?
IMES is a lesser-known part of the
regional Quasi-Zenith Satellite System
(QZSS) being developed by the Japan
Aerospace Exploration Agency (JAXA).
Te QZSS program will place three sat-
ellites in high-altitude orbits transmit-
ting ranging signals to improve naviga-
tion performance in areas of Japan that
prove difcult with GPS alone. Te frst
launch of a QZSS satellite is scheduled
for 2010.
Five of six QZSS signals transmitted
from these satellites will use the same
signal structures, frequencies, spreading
code families, and data message formats
as GPS and GPS satellite-based augmen-
tation system (SBAS) signals.
An annex to the interface specifca-
tion for QZSS (IS-QZSS) sets forth the
IMES signal design. (See reference in
Additional Resources section near the
end of this article.)
Although QZSS satellite signals will
only be available in a western Pacific
Ocean region centered over Japan, IMES
is a separate terrestrial element based
on an open specifcation that could be
implemented anywhere.
According to the annex, the IMES
signal is designed by JAXA to contrib-
ute to the development of QZSS-ready
receivers as well as satellite positioning
applications by realizing the seamless
positioning environment. However,
IMES signal transmitters are not part of
QZSSs Indoor
Messaging System
A lesser-known aspect of Japans Quasi-Zenith Satellite System (QZSS) program
is development of IMES or Indoor Messaging System, which uses pseudorandom
noise codes (PRNs) and operates in the GPS L1 frequency. The Japan Aerospace
Exploration Agency (JAXA) designed the IMES signal to contribute to
development of QZSS-ready receivers as well as satellite positioning applications
by supporting development of seamless indoor/outdoor positioning. This
article explores whether IMES is the perfect answer forseamless ubiquitous
positioning or possibly a serious jamming threat to the integrity of GPS.
ANDREWDEMPSTER
UNIVERSITY OF NEWSOUTH WALES
C
o
p
y
r
i
g
h
t

i
S
t
o
c
k
p
h
o
t
o
.
c
o
m
/
H
e
n
g

K
o
n
g

C
h
e
n
GNSS Friend or Foe?
___________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
38
the QZSS component; rather, the IMES
specification is intended to aid third-
party vendors in further development
and installation of the system.
Primarily, IMES is designed to pro-
vide accurate positioning indoors where
reception of GPS and other GNSS sig-
nals is blocked or unreliable. By being
able to operate indoors using GPS-capa-
ble equipment (by upgrading a receiver
frmware with the IMES navigation bit
decoding algorithms), the system is
meant to pick up where GPS fails, in a
seamless way.
IMES is real ly competing with
assisted-GPS or GNSS (AGPS/AGNSS),
which has been used with some success
although substantial inaccuracy for
indoor positioning. In particular, its advo-
cates advance IMESs utility for enhanced
911 (E911) automatic location of emergen-
cy calls from mobile phone users.
At the recent International Sympo-
sium on GPS/GNSS in Tokyo, a number
of technical presentations and exhibitors
discussed the implementation of IMES
technology by JAXA, GNSS Technolo-
gies (GNSST), and other companies. Hit-
achi and GNSST had receivers on their
exhibit stands and graphics explaining
how IMES worked.
For someone with a GPS back-
ground, these graphics were somewhat
confusing. It looked as though IMES was
a system of GPS retransmitters mounted
to ceilings something that wouldnt
help much at all for indoor navigation.
Furt her i nvest igat ion, howev-
er, revealed that these transmitters,
although they operate in the GPS L1
band, are in fact transmitting a com-
pletely separate signal. Te IMES signal
simply gives the location of the nearest
transmitter, which an appropriately con-
fgured receiver can then take to be its
position.
At this point, of course, the alarm
bells go of. Transmitting in L1? To GPS
users, thats tantamount to jamming. So,
is IMES the perfect answer to the seam-
less ubiquitous positioning problem
or is it a dangerous jamming threat to
the integrity of GPS?
Perhaps the answer lies somewhere
in between. . . . With that thought in
mind, this article briefy describes the
IMES program and raises several key
issues about its practical use in combi-
nation with GNSSs.
The IMES Signal
In June 2008, JAXA released version
1.0 of the IS-QZSS document, which
includes the IMES specifcation. Te RF
characteristics of IMES are the same as
the L1 C/A code for GPS and QZSS.
Transmitted at the GPS L1 center
frequency (1575.42 MHz), IMES has
a bandwidth of 2.046 MHz or more
including the main lobe. Like the L1 C/
A code, the IMES signal is right-hand
circularly polarized and BPSK-modu-
lated with a pseudorandom noise (PRN)
code.
In the current interface specifcation
for GPS (IS-GPS-200D) the U.S. gov-
ernment has approved allocation of the
Gold (or PRN) codes 173 to 182 for use by
other GNSS applications such as IMES.
The received power level at an IMES-
capable receiver is specifed to fall in the
range between -158.5 dBW to 94 dBW.
Word structure, bit rate, and modulation
are the same as L1 C/A code.
IMES Messages
IS-QZSS defnes four diferent types of
messages, as shown in Table 1. Two mes-
sages give the location of the transmit-
ter. Message type 0 gives the latitude (23
bits) and longitude (24 bits) in WGS-84.
However, height or altitude appears as a
building foor number (8 bits). Message 1
provides latitude (24 bits), longitude (25
bits), altitude (12 bits), and foor number
(9 bits with units of 0.5 foor).
Messages 3 and 4 simply send an
identifer, which, according to GNSST,
can then be used to address a location
in a database corresponding to that ID.
Messages 3 and 4 also transmit a BD
bit, which is a border or boundary indi-
cator, set whenever the transmitter is the
one nearest the outdoors or a GPS-
accessible area. (As yet undefned, Mes-
sage type 2 is reserved for later develop-
ment.)
IMES can be used by a suitably
modifed stand-alone GPS receiver but is
primarily intended for use with a GPS-
enabled mobile device. Notably, messag-
es 3 and 4 rely on the device to be able to
access a database via a network.
The Additional Resources section
includes a number of articles, particu-
larly those by D. Manandhar et alia, that
describe IMES, its applications, and test
results to date in greater detail.
Whats Good about IMES?
What IMES delivers that other indoor
location systems does not is reliability
and accuracy. When receiving an IMES
signal, the receiver has a strong idea of
where it is to within tens of meters. Te
accuracy is better than AGPS, especially
in height, and receivers are not required
to have high sensitivity.
The big selling point for IMES in
Japan is its ability to add a lot of value to
simple positioning with additional data,
such as maps and route guidance. For
example, locations extracted from the
databases used with messages 3 and 4
can be accompanied by location-based
service (LBS) information.
In fact, the location may not even
be returned at all. Te sorts of applica-
tions used by Hitachi and GNSST to
promote or, in Australian slang, spruik
IMES include indoor navigation, fnd-
ing products in a store, geofencing of
children, location-specifc instructions
The big selling point for IMES in Japan is its ability to
add a lot of value to simple positioning with additional
data, such as maps and route guidance.
Message
ID
Frame
Length
(words) Contents
Minimum
repetition
cycle (sec)
0 3 Position 1 12
1 4 Position 2 12
3 1 Short ID 6
4 2 Medium ID 12
TABLE 1. IMES Message denitions
QZSSS INDOOR MESSAGING
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
in case of emergency, asset management,
and so forth.
Whats Not SoGoodabout It?
Te technology is in its infancy; so, the
list of problems that we will raise here
looks large, but each issue needs to be
dealt with if IMES is in fact going to suc-
ceed AGPS.
1. Infrastructure. For IMES to work,
the transmitters need to be very densely
located in all indoor spaces where loca-
tion is required, at separations of 2030
meters. Tis represents a massive invest-
ment in infrastructure. However, unlike
AGPS, this investment does not neces-
sarily need to be made by the telecom
companies alone or even at all. Also, if
the IMES infrastructure is incomplete,
AGPS may still be required to fill the
gaps.
2. Jamming. IMES should only afect
a small area around the transmitter.
However, within this area, IMES will
completely jam GPS.
Te QZSS specifcation suggests that
the IMES power levels are high enough
to jam GPS outdoors. A maximum
transmitted signal strength of -94 dBW
is about 65 dB stronger than an unob-
structed GPS signal received outdoors.
Te receiver will generally receive the
signal at a much lower level; the mini-
mum is specifed to be -158.5 dBW, but
jamming is still likely to occur nearby to
a transmitter. Indoors, where the direct
GPS signals are strongly attenuated, the
situation would be worse.
Te Gold codes are designed so that
a receiver should be able to detect a GPS
signal that is about 21dB weaker than the
strongest GPS signal. Tis is known as
the cross-correlation margin. Advanced
signal-processing techniques, such as
those developed by Eamonn Glennon
at the University of New South Wales,
can increase this margin substantially,
but even with indoor GPS, cross-corre-
lation margins of 65 dB have not even
been considered.
Ot her systems t hat have been
assigned codes in the IS-GPS-200D,
such as SBAS and QZSS itself, transmit
at levels that do not exceed the cross-
correlation margin. In practice, in Japan
the levels will not be as high as those in
the specification. GNSST suggest that
transmitters will be limited to -100 dBW
(or -70 dBm), still signifcantly stronger
than GPS.
Initial GNSST experiments suggest
that this transmitted signal strength only
has a noticeable efect on a GPS receiver
within one meter (with the stronger level
of the specifcation afecting receivers
out to three meters distance). These
afected regions seem small, and inde-
pendent tests should be used to confrm
the GNSST results. If they are valid,
IMES will only jam a relatively small
area.
Te system model of IMES is also not
backward compatible, meaning that if a
user has an AGPS device that has happily
worked for years, once IMES is installed,
that device can no longer use AGPS.
3. Seamlessness. Interestingly, in
the papers presented about IMES, the
word seamless appears prominently.
Te idea is that the same equipment can
use GPS outdoors and IMES indoors,
seamlessly. Some papers even claim
that the seamlessness has been proven
by experiment.
Tis isnt quite the case, however.
Modif ied receivers and mobi le
phones have been shown to work indoors
with IMES signals and outdoors with
GPS, but as yet none of the published
experiments appear to showing a receiv-
er happily using IMES and transitioning
without interruption to GPS outside, or
vice versa. Te transition from GPS to
IMES should be seamless that is,
the receiver holds on to GPS until it is
jammed or blocked indoors, by which
time it has picked up IMES (as long as it
is looking for it).
Going the other way is more prob-
lematic. IMES can be used until it is too
weak to be received, but then the receiver
should be able acquire four GPS satel-
lites. If the equipment is being used in
an environment broadcasting Messages
3 or 4, then it will have been warned by
the BD bit to start looking for GPS sig-
nals. In a Message 1 or 2 environment,
no such warning will have been given.
Because IMES requires the receiver
to receive one IMES signal at a time, a
handover between IMES transmitters
must take place. So, while receiving from
one transmitter, the receiver must be
searching for others, not knowing which
of the PRNs is best to search for.
The near-far phenomenon means
that the current IMES transmitter
will jam others until they are within
the cross-correlation margin. Con-
sequently, once a transmitters power
has been set, maximum and minimum
distances emerge for the placement of
other transmitters.
Tese transmitters must use diferent
PRNs or else they will spoof each other.
Tis means that the IMES transmitters
must be laid out based on a code-reuse
pattern similar to the frequency-reuse
patterns employed by cellular networks,
with the added complication of operat-
ing in three dimensions, i.e., diferent
foors of a building by defnition have
diferent IMES transmitters.
Returning to the question of GPS/
IMES transitions, although a huge
Copyright iStockphoto.com/Rene Mansi
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
___________
40
efect is not likely, this cross-acquisition
between GPS and IMES will use more
energy than for one system alone, if for
no other reason than it increases the
PRN search space by the 10 codes used
for IMES.
4. You are the ground segment. Te
IMES transmitters, like GPS satellites,
tell the receiver where they are. Howev-
er, in the case of GPS satellites, a secure,
centralized ground segment infrastruc-
ture is used to create the transmitted
messages.
In the case of IMES, the lat/long/
height/f loor that is transmitted must
come from somewhere else. Each of the
locations must be surveyed in. Te accu-
racy of this survey need not be great,
because the resolution of the transmit-
ted messages is 13 meters and, in any
case, that position applies for regions of
tens of meters.
However, the surveyed position of an
IMES transmitter still needs to be accu-
rate to, say, meters in WGS-84. Given
the enormous numbers of transmitters
needed for this system, a huge scope for
error arises in a) the measurement of the
position, b) the data entry of the posi-
tion, and c) the installation of the equip-
ment (a transmitter with a perfectly well
entered position being installed in the
wrong place).
5. They are the ground segment. Mes-
sages 3 and 4 dont tell you your position
directly. Tat information comes from
a database. Te business model put for-
ward by GNSST suggests that the data-
base may be owned by, for instance,
department store, underground mall
and etc.
What that means is that users may
not get the position theyre looking for,
just what the department store or under-
ground mall wants them to receive. As
with casinos in Las Vegas, navigating
ones way out of the shop as soon as pos-
sible may not be in the store owners best
interest.
6. Security. Tens or hundreds of
thousands of these devices are needed
for IMES to operate properly. Tey will
have to be installed in public areas such
as shopping centers and railway stations,
and private areas such as shops. Teir
installation needs to be cheap because
of the sheer numbers involved.
However, if an IMES transmitter
can be installed relatively simply, it is
also likely to be easy to remove. A stolen
transmitter, or one simply bought from
the manufacturer, could cause havoc.
Driving through a city with a high-pow-
ered IMES transmitter on the dashboard
could cause all GPS navigators nearby to
either cease working, or worse, if they are
IMES-enabled, to think they were in the
location of the stolen device (i.e., inside
a building). It makes GPS jammers (or
IMES spoofers) readily available to any
small-time thief.
7. Frequency allocation and regulation.
Ultimately, IMES will only ever operate
in countries where the IMES signal has
been sanctioned in the L1 band. This
may in some cases require a change in
the law before it can operate. In Aus-
tralia, for instance, any transmission in
the L1 band must not have an intention
to jam GPS, otherwise it is considered
a criminal act. IMES works by deliber-
ately jamming GPS (or at least AGPS) in
indoor environments.
Although IMES transmitters are not
pseudolites, it would make sense if they
were regulated in a similar way. Recently,
the European Conference of Postal and
Telecommunications Unions (CEPT)
has taken the lead in trying to come up
with a global standard for pseudolite
regulation. The relevant committee is
yet to report (possibly early in 2009).
Conclusion
In summary, it appears that before
IMES can make signifcant inroads into
indoor positioning, a series of hurdles
must frst be overcome. IMES is not a
magic bullet for the current hot problem
of ubiquitous or seamless positioning.
Its just another option to add to the list,
with its own set of strengths and weak-
nesses.
Japan Aerospace Exploration Agency, Quasi-
Zenith Satellite System Navigation Service Inter-
face Specication for QZSS (IS-QZSS) V1.0, June
17, 2008 (available on-line at <http://qzss.jaxa.
jp/is-qzss/index_e.html>)
NAVSTAR Global Positioning System Interface
Specication IS-GPS-200, Revision D, IRN-200D-
001, Navstar GPS Space Segment/Navigation User
Interfaces, GPS Wing, Space and Missile Systems
Center, Los Angeles AFB, California, USA, March
7, 2006,
Glennon, E., and R. Bryant, A. Dempster, and P.
Mumford, Post Correlation CWI and Cross Corre-
lation Mitigation Using Delayed PIC, Proceedings
of ION GNSS 2007, Fort Worth, Texas, US, Septem-
ber 26-28, 2007
Manandhar, D., and K. Okano, M. Ishii, M. Asako,
H. Torimoto, S. Kogure, and H. Maeda, Signal
Denition of QZSS IMES and Its Analysis, Pro-
ceedings of ION GNSS 2008, Savannah, Georgia,
USA, September 1619, 2008
Manandhar, D., and K. Okano, M. Ishii, M. Asako,
H. Torimoto, S. Kogure, and H. Maeda, Develop-
ment of Ultimate Seamless Positioning System
for Global Cellular Phone Platform Based on QZSS
IMES, Proceedings of ION GNSS 2008, Savannah,
Georgia, USA, September 1619, 2008
Manandhar, D., and S/ Kawaguchi, M. Uchida, M.
Ishii, and H. Torimoto, IMES for Mobile Users:
Social Implementation and Experiments based
on Existing Cellular Phones for Seamless Position-
ing, Proceedings of International Symposium on
GPS/GNSS, Tokyo, November 1114, 2008
Martin, S., and H. Kuhlen, and T. Abt, Interference
and Regulatory Aspects of GNSS Pseudolites,
Journal of Global Positioning Systems, vol. 6, no.
2, pp. 98-107, 2007
Author
Andrew Dempster is
Director of Research in
the School of Surveying
and Spatial Information
Systems at the Univer-
sity of New South Wales
(UNSW), Sydney. He has
a BE and MEngSc from UNSW and a Ph.D. from
University of Cambridge. He was project manager
and system engineer on the first GPS receiver
developed in Australia in the late 1980s and is
named on three patents arising from that work. He
has taught GPS-related topics for more than a
decade at the University of Westminster, London,
and at UNSW. His current research interests are
GNSS signal processing, software-dened receiv-
er design, and new location technologies.
QZSSS INDOOR MESSAGING
______________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
cess of language arts commonly likened to sausage-making.
Te process was probably more revealing than the results,
which can be viewed in the ofcial report issued following
the meeting. A copy of the report can be downloaded as a
PDF on-line at <www.insidegnss.com/auto/ICG-3_Joint_
Statement_&_PF_Report.pdf>.
Te nuances refected underlying issues between indi-
vidual GNSS providers, including some under discussion in
bilateral talks and others have yet to be raised. Tese issues
Chinas plan to transmit Compass signals on the same fre-
quencies as Galileos Public Regulated Service (PRS) and
GLONASSs possible selection of a BOC (2,2) design for its
CDMA signal at L1.
As an example of the tendency toward divergence, con-
sider the various approaches to defning compatibility and
interoperability taken up by Working Group A charged with
drafing appropriate language on the subject for the ICG
workplan.
Spectral Separation and Other Barriers
At the Bangalore meeting, wording for the compatibility
principle suggested that Compatibility should also involve
spectral separation between each systems authorized service
signals and other systems signals. However, in Pasadena
several representatives pointed out that spectral overlap
already exists in signals for a number of authorized or
restricted services.
Ultimately, the group agreed on the following language:
Compatibility should also respect spectral separation
between each systems authorized service signals and other
systems signals. Recognizing that some signal overlap may
be unavoidable, discussions among providers concerned will
establish the framework for determining a mutually accept-
able solution.
Interoperability posed an even more complex challenge
for reconciliation. Although European and Indian delega-
tions remained satisfed with the language worked out in
Bangalore, the intervening months appear to have brought
second thoughts and diverging analyses by other providers.
In a U.S. efort to fesh out the concept, Harrington pre-
sented a lengthy set of technical characteristics that he sug-
gested underlie interoperability. Tese included common
time and reference frames or broadcast of the ofsets between
systems, common carrier frequencies, similar spreading mod-
ulation spectra, common minimum and maximum power
levels, common spreading code lengths and common code
family, and common data message structure and encoding.
Professor Grigory Stupak of the Russian Institute for
Space Device Engineering (RISDE) introduced a formula in
which interoperability could be quantifed using other crite-
ria for cost (price, size, weight, power consumption) and per-
formance (accuracy, availability, reliability, responsiveness).
China frst proposed that interoperability simply allows
navigation with signals from diferent systems with minimal
additional receiver cost or complexity.
Eventually, the working group agreed that, in addition
to the Chinese description, for purposes of interoperability,
Geodetic reference frames realization and system time
steerage standards should adhere to existing international
standards to the maximum extent practical.
Te ICG-3 language also acknowledged the benefts of
multiple constellation in providing improved accuracy and
availability of PNT services.
In the preface to the revised document, delegates added,
For many applications, common carrier frequencies are
essential to interoperability, and commonality of other signal
characteristics is desirable. In some cases, carrier frequency
diversity may be preferable to improve performance.
Te workplan also included language that providers
should develop proposals to widely monitor the perfor-
mance of their open signals and provide timely updates to
users with a focus on potential cooperation in the develop-
ment of the necessary ground infrastructure to monitor sig-
nal and service performance for open services. . . .
NarrowRole of the ITU
Another point of contention revolves around the role of the
International Telecommunications Union (ITU), a UN-afli-
ated body that allocates radio spectrum in world radiocom-
munication conferences for particular uses regionally and
globally. Compliance is voluntary, but generally efective
among ITU members.
However, the ITU is essentially limited to addressing a
narrow defnition of compatibility, based on frequency allo-
cations and attempts to limit cumulative increases in the RF
noise foor within radio bands as a result of new systems and
signals. Adding signals can eventually create interference to
existing services within a frequency allocation.
China appears inclined to rely more heavily upon the ITU
process, while others are less sanguine about the organiza-
tions capability for reconciling broader diferences among
programs.
Among these diferences is the matter of spectral separa-
tion of military or authorized services, such as the GPS M-
code and Galileo PRS, which have been the subject of a series
of bilateral talks thus far inconclusive between China
and the United States and the European Union in recent
years.
GPS as a Legacy System
Apparently, the Chinese have acknowledged GPS as a legacy
system that has a well-established presence in certain bands
and will avoid conficts with it. However, they consider Gali-
leo, as a system at a comparable stage of development with
Compass and, therefore, with comparable claims to the spec-
trum. Te launch of the frst Compass MEO in April 2007
provided a foothold, a claim that they could well expand with
the additional launches of satellites over the next two years.
GNSS WORLD
ICG-3 continued from page 13
____________________
___________________________
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
42
Program Updates
GLONASS
Program updates revealed steady progress by Russia, with a delegation
headed by Yury Nosenko (deputy director of the Russian Space Agency
or Roscosmos), which expects to have 24 GLONASS satellites or space
vehicles (SVs) on orbit by the end of 2009.
Sergey Revnivykh, director of the Satellite Navigation Department in
the Roscosmos Mission Control Center, noted that ight test of the rst
GLONASS-K satellite toward the end of 2010 will allow Russia to evaluate
new CDMA signals that it plans to broadcast at L1 and possibly L3 frequen-
cies.
GLONASS will continue transmitting its existing FDMA signals for the
unlimited future to provide backward compatibility, Revnivykh added.
Compass
China announced plans to launch 3 to 4 Compass (Beidou 2) spacecraft in
2009 and 11 over the next two years.
Jun Lu, representing the China Satellite Navigation Project Center, said
that results from 18 months of observations of the rst Compass middle
earth orbiting (MEO) satellite launch in April 2007 indicated a precise
orbit determination of better than 5 meters as well as the ability to predict
future orbits of better than 10 meters over a 24-hour period.
The four rubidium atomic frequency standards on board the MEO-1 space-
craft are providing a synchronization with Compass system time of about
two nanoseconds, she said.
GPS
Presentations by U.S. representatives focused on recent publication of
new performance standards, progress in U.S. space- and ground-based
augmentation systems, and GPS interference detection and mitigation.
They also described GPS modernization plans scheduled to bring fully
operational capability (FOC) with 24 GPS SVs broadcasting the second civil
signal (L2C) by 2016, L5 by 2018, and the new L1 civil signal by 2021.
U.S. presenters included Mike Shaw, director of the National Coordina-
tion Ofce for Space-based PNT, Leo Eldredge, GNSS program director at
the Federal Aviation Administration, Col. David Goldstein, chief engineer
for the GPS Wing at Los Angeles Air Force Base, and John Merrill, with
the Ofce of Applied Technology, Geospatial Management Ofce, in the
Department of Homeland Security. Lt. Col. Patrick Harrington, from the
Ofce of the Under Secretary of the Air Force Directorate of Space Acquisi-
tion, headed the U.S. delegation.
Galileo
Paul Verhoef, head of the Galileo unit for the European Commission Direc-
torate-General for Transport and Energy, said that the EU expects to have
contracts for building the Galileo FOC system in place by next summer.
He noted that the target date for completion is 2013 had been given to
prospective vendors in the Galileo invitations to tender (ITTs). Well see if
and how they will make it.
The European GNSS Supervisory Agency (GSA), which had been
charged with overseeing development and operation of Galileo under the
abandoned public-private partnership, will be given a different name next
year to reect its altered responsibilities under the public procurement
now led by the EC and the European Space Agency.
Verhoef added that about 30 GSA staff members will soon be trans-
ferred to the ECs Galileo unit.
Meanwhile, despite pointed industry encouragement, Verhoef did
not say when a nal interface control document (ICD) would be available.
Manufacturers need the commercial terms and technical specications in
order to build and sell Galileo-capable equipment with condence.
GAGAN
The Indian government has approved the follow-on plan for its satellite-
based augmentation system the GPS Aided, Geo Augmented Navigation
(GAGAN, which stands for space in Sanskrit).
India plans to launch the rst payload on its GSAT-4 satellite next April
or May and a second GAGAN payload on GSAT-8 during the last quarter
of 2009, according to Suresh Kibe, satellite navigation program manager
at the Indian Space Research Organization (ISRO). A third GAGAN GEO is
scheduled for launch on GSAT-9 or -10 in 2010.
Kibe says development continues for the Indian Regional Navigation
Satellite System (IRNSS), with seven to nine satellites in both geosyn-
chronous and non-geosynchronous orbits. It will include an open standard
positioning service (1Mhz, biphase skip keying signal) and a restricted
service BOC (5,2) signal at 1176.45 MHz (L-band) and 2492.08 MHz
(S-band).
India is studying the signal structures of GPS L2C, L5, and L1C as the
basis for the IRNSS signal design.
QZSS
The Japan Aerospace Exploration Agency (JAXA) will launch the rst satel-
lite in its Quasi-Zenith Satellite System (QZSS) in the summer of 2010,
followed by a year of in-orbit validation.
In addition to several open signals closely aligned with GPS, QZSS will
include two additional signals an L1-SAIF (submeter-class augmenta-
tion with integrity function) and an experimental signal (LEX) at 1278.75
MHz. These may be implemented with fees charged by a private operator,
according to Koji Terada, QZSS program manager for JAXA.
GNSS WORLD
Reportedly, China has posed a long list of requirements to
EU negotiators, including some unrelated to GNSS, as condi-
tions for abandoning or reducing their frequency plans for
Compass.
Te ICG itself has yet to demonstrate whether it will
tangibly and practically improve the prospects for creating
a global GNSS system of systems with true compatibility
and interoperability among separate GNSSs. Or whether it
is merely a debating society or paper tiger, as some have sug-
gested.
But for the time being, the ICG appears to ofer the only
alternative to continuing rounds of bilateral negotiations that
may achieve individual agreements but may have less success
in coordinating uniformity among GNSS providers. If noth-
ing else, the ICG ofers a more open forum in which moral
suasion compels a semblance of cooperation.
ICG-4 in St. Petersburg
Te next full meeting of the ICG ICG-4 is scheduled for
September 1418, 2009, in St. Petersburg, Russia.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
I
n the November/December issue of
Inside GNSS, the frst part of this
column described the upcoming
scientifc missions that fy two or
more smaller satellites in close forma-
tion to create large spaceborne instru-
ments. The final part of this series
explains the GNSS techniques and
technologies employed to achieve very
accurate relative positioning and orien-
tation of the spacecraf at lower altitudes
as well as a similar approach used at
higher altitudes for relative positioning
by means of RF carrier phase measure-
ment techniques.
FF Missions Metrology
Requirements
Table 1 summarizes the needs of the
formation fying (FF) missions in terms
of metrology requirements. In general,
all the proposed scientific missions
discussed here have FF elements with
closed-loop formation control in non-
Keplerian orbits, typically at L2 (the sec-
ond Lagrange Point about 1.5 million
kilometers from the Earth) or HEO. Tey
all have demanding accuracy require-
ments and are sufciently consolidated.
The proposed EO missions are
mostly in LEO orbit with less demand-
ing navigation accuracy requirements
in real time (very ofen the processing
is done on the ground). At the moment,
requirements for the EO FF missions are
less well defned than the space science
FF missions. PRISMA and CanX-4/5 are
short-term missions with launches ten-
tatively scheduled for 2009.
From the overview of the FF mission
metrology requirements, we can iden-
tify four main development lines in the
frame of spacecraf formation fying.
Earth Observation Missions. These
missions, in LEO orbit, will respond to
the demand for highly accurate Earth
models on a global space and time scale.
Two or more satellites of identical type
and build are fown at close distances to
synthesize three-dimensional baselines
between the satellites that can be recon-
fgured during the mission lifetime.
Te relative orbit control accuracy
required for such formations is relatively
coarse (~100 meters) and may drive the
need for real-time onboard relative navi-
gation accuracies at a 110 meter level.
High precision (submillimeter) post-
facto reconstruction of the three-dimen-
WORKING PAPERS
The idea of ying two or more satellites in controlled formations with complementary sensors
on board to execute scientic research missions will receive a lot of practical attention in
the near future. Part 1 of this two-part series reviewed some of those missions and their
objectives. In some cases, GNSS signals will be used to meet the high-precision positioning
and orientation requirements for this application. Other, higher-orbit missions will employ
RF-based techniques similar to those used to make GNSS carrier phase measurements. This
second and nal part of the series describes in greater detail the technologies and techniques
used in this scientic research.
GNSS in Space
Part 2 Formation Flying
Radio Frequency Techniques
and Technology
THOMAS GRELIER
CENTRE NATIONAL DETUDES SPATIALES (CNES)
ALBERTO GARCIA
EUROPEAN SPACE AGENCY (ESA)
ERIC PRAGIN, LAURENT LESTARQUIT, JON HARR,
DOMINIQUE SEGUELA , JEAN-LUC ISSLER
CNES
JEAN-BAPTISTE THEVENET, NICOLAS PERRIAULT,
CHRISTIAN MEHLEN, CHRISTOPHE ENSENAT,
NICOLAS WILHELM, ANA-MARIA BADIOLA
MARTINEZ
THALES ALENIA SPACE
PABLO COLMENAREJO
GMV
PRISMA satellite
Photo courtesy of Swedish Space Corporation (SSC) and Intespace
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
44
of about 30250 meters. Autonomous
formation fying capabilities are driv-
en by the telescope optical design and
should allow uninterrupted science
observations. Tis translates into com-
bined attitude/orbit control systems
with required navigation accuracies at
(sub)centimeter level.
A clear constraint arises from the
need to implement a navigation system
at altitudes above the GNSS constella-
tions. Provided that the GNSS receivers
can acquire and track the very weak side
lobes of the broadcast signals, real-world
simulations have demonstrated centime-
ter-level relative navigation accuracies in
LEO (~5 centimeters) and HEO at radial
distances up to 17 times the Earth radi-
us. Only self-contained relative (inter-
satellite) navigation sensors (i.e., radio
frequency and optical) can fulfil the
requirements of autonomous formation
fying at even higher altitudes.
Multi-Spacecraft Telescopes. Te third
type of application addresses the use of
multiple spacecraf telescopes. Research-
ers have identifed interferometry in the
infrared and visible wavelength regions
as the key technology to support new
astrophysical discoveries and the direct
search for terrestrial exoplanets.
To achieve that purpose, clusters of
three or more units need to fy in -mil-
limeter-precision-close formations with
interspacecraf navigation accuracies at
the submillimeter level. Only optical
sensors and laser interferometers can
provide the required formation fying
metrology performances. RF metrology
is needed for deployment and maneuver
control.
Long-Range and RdV Missions. Te last
type of application involves long-range
and rendezvous (RdV) missions. Tese
types of missions require an RF sensor
technology, combined with the navi-
gation algorithms of the GNC system,
during the long-range phase while the
satellites are far apart. Te chaser vehicle
must be able to detect, acquire, and track
the relative position of the target space-
craf to close on, and then perform, the
fnal approach and docking.
Examples are post-ATV (Crew Space
Transportation System or CSTS), Next-
Mars, and Mars Sample Return (MSR)
missions. Long-range metrology used
in HEO activities include the Magne-
tosphere Constellation (MagCON),
Magnetospheric Multiscale (MMS), and
Cross-Scale missions (all studies of the
Earths magnetosphere). Tese missions
also require a capability for long-range
operations across thousands of kilo-
meters, which makes this type of FFRF
technology very attractive.
The RF Metrology
Subsystem
On FF missions, the FFRF subsystem is
responsible for the relative positioning
of two to four satellites. It generates rela-
tive position, velocity, and line-of-sight
(LOS) data as inputs to a GNC subsys-
tem for which it provides coarse mea-
surements.
Te overall FF metrology system has
both coarse and fne modes. Te coarse
mode provides accuracy of one meter
and 20 degrees for LOS with omnidi-
rectional coverage in order to provide
inputs for pointing the telescope, imag-
ing sensor, or other instrumentation of
the spacecrafts primary mission. For
fne mode, the accuracy is expected to
be one centimeter for distance and one
degree for LOS measurement in a cone
of a few degrees of aperture.
sional relative motion may be needed for
some missions.
Discussions about future gravity
feld satellite missions are underway to
overcome the intrinsic limitations of
gravimeters such as CHAMP (Chal-
lenging Minisatellite Payload), GRACE
(Gravity Recovery and Climate Experi-
ment), and GOCE (Global Ocean Cir-
culation Experiment). The GRACE
geodetic observables, for example, are
inherently non-isotropic, as a result of
the permanent along-track orientation
of the laser link and its scalar charac-
ter. To enhance the spectral content,
future geodetic satellite missions (post-
GRACE, post-GOCE) would make use
of autonomous formation flying with
multiple baselines.
Dual Spacecraft Telescopes. These
instruments aim at spectral investiga-
tion of sources that are too faint for study
with the current generation of observa-
tories (e.g., Chandra, XMM-Newton).
Te typical mission profle seeks orbits
with a low level of perturbations, stable
thermal environment, lack of eclipses,
and wide sky visibility.
In contrast to the unfavorable LEO
environment, in this context, GEO,
HEO, and Lagrange points of the Sun-
Earth system ofer optimum conditions.
Typical separations aim at focal lengths
FF Mission Needs Non-FF
Earth Observation Dual Spacecraft
Telescope
Multi Spacecraft
Telescope
Long-Range & RdV
Missions PRISMA, GRACE,
PostGOCE, postGRACE,
TerraSAR/TanDEMX,
NanoForm, SABRINA,
Romulus
PROBA-3, GRL, Xeus,
Simbol-X, MAX
Darwin, PEGASE,
SPECS, TPF, New
Millenium
NextMars, MSR, ATV,
CSTS-ISS (LEO), CSTS-
Expl (Moon), SMART-
OLEV, MagCON, MMS,
ALFA, MAXIM
Orbit Low earth orbit (LEO) High earth orbit,
HEO (or Lagrange
point)
Lagrange point (or
HEO)
HEO, moon, Mars, LEO
(ISS), GEO (OLEV)
Number of
S/Cs
2 2 3 Long-range: 3
RdV: 2
Typical
separation
100 m1000 km 30250m 101500m Long-range: 100m
3000km
Navigation
accuracy
110 m
(1 mm post-facto)
0.011cm 1100mm 10100m
Technology GNSS space receivers
(or integrated with
RF ISD)
RF metrology RF long-range metrology
(integrated with
GNSS Rx for LEO)
TABLE 1 FF Metrology needs
WORKING PAPERS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
As the frst element in the FF metrol-
ogy system chain, the FFRF sensor
provides initial relative positioning for
the subsequent optical metrology sub-
systems that could be used if a higher
accuracy is required (coarse optical lat-
eral metrology, fne optical metrology,
and fne longitudinal metrology).
A dedicated computer/RF terminal
and up to four sets of antennas on each
satellite in the constellation comprise the
FFRF subsystem. A set of antennas can
be either a triplet one receiver/trans-
mitter (Rx/Tx) master and two Rx slaves
or a single Rx/Tx antenna. Te maxi-
mum number of triplets is two, and the
total maximum number of antennas is
eight.
Te FFRF terminal operates with a
dual-frequency S-band ranging signal,
and each terminal transmits and receives
signals to and from all the other satellites
in a time division multiple accessbased
pattern. Ranging and angular measure-
ments are extracted from received sig-
nals and used for computing relative
position, velocity, and LOS. In addition
to providing relative navigation mea-
surements, the FFRF subsystem also
provides an intersatellite data link (data
ISL) as auxiliary functionality.
Te FFRF equipment developed for
PRISMA thanks to the European Space
agency (ESA), CNES, and CDTI (Centro
para el Desarrollo Tecnolgico Industri-
al Spain) is generic fexible in terms
of frequency plan, number of antennas,
and antenna accommodations. The
FFRF functionalities planed for PROBA3
that are not present on PRISMA include,
for instance, an extended intersatellite
range up to 100 kilometers instead of
30 kilometers and higher data rates for
short intersatellite distances, thanks to
a current ESA predevelopment made in
cooperation with CNES.
Frequency Band. A study of Inter-
national Telecommunications Union
(ITU) regulation showed that the only
portion of RF spectrum allowing FFRF
operation (provided with reasonable
transmitted power to comply with the
omnidirectional coverage requirement)
was the space-to-space communication
band sharing the tracking, telemetry,
and control (TTC)
al l ocat ion i n S-
band (20252110
MHz and 2200
2290 MHz). The
challenge is to fnd
frequencies that are
compat ible wit h
the platform TTC,
which is also in S-band. To ensure com-
patibility with all future FF missions, the
FFRF frequencies can be programmed
during the manufacturing stage at sev-
eral frequencies located in the TTC S-
band. Te TTC frequencies of missions
like PRISMA, PROBA3, and Simbol-X
are in S-band, but X-band could also be
used for telemetry of certain missions,
such as one at a Lagrange point, for
instance.
For the PRISMA mission, Tales Ale-
nia SpaceFrance (TAS-F) performed
a thorough study to reduce the risk of
disturbance between the FFRF and the
TTC subsystems. Te result is a frequen-
cy plan that takes into account the fol-
lowing constraints and hypotheses:
maximize the frequency separation
between the telecommand (TC)
band and S2 and between the telem-
etry (TM) band and S1.
maximize the RF distance between
FFRF and deep space research (DSR)
bandwidth (22902300 MHz).
provide a reasonable FFRF interfre-
quency separation to optimize the
integer ambiguity resolution (IAR)
functionality.
Intermodulation (IM) products
between FFRF S1 and S2 and TM could
result from TM signal leaking into the
switch of the FFRF RF module; these
IM products will not fall into the TC
band.
For PRISMA, this led to a frequency
plan in which FFRF frequencies are
located in the upper part of TM and TC
bands: S1 = 2275 MHz and S2 = 2105
MHz. TM frequency is 2214 MHz, while
TC frequency is equal to 2035 MHz. Tis
frequency plan guarantees sufficient
separation between S2 and TC. More-
over, the signal at S2 has reduced power
compared to S1, which limits further
interference with TC.
Signal Characteristics. Measurements
are made thanks to signals based on GPS
technology. Using multi-antenna bases
(triplets) and TDMA sequencing, each
terminal transmits and receives a GPS C/
A-code navigation signal modulated on
two S-band carrier frequencies. A data
link is provided on the frst frequency
with data bits modulated in quadrature
of the navigation signal.
Table 2 outlines the full set of FFRF
signal characteristics.
Measurement Principle and
Factors
To allow the determination of relative
position and relative speed, the FFRF
subsystem provides the following fne-
mode information every second:
intersatellite scalar distance (speci-
fed precision = one centimeter)
intersatellite velocity with a precision
of a few millimeters
azimuth and elevation of line of sight
between two satellites (specified
accuracy = one degree)
azimuth and elevation variations
clock bias between the two satel-
lites.
Every terminal is equipped with both
a transmitter and a receiver; so, each sat-
ellite is able to make ranging and LOS
measurements with every other satellite.
First, the system makes coarse measure-
ments using the ranging signal from the
C/A code and then performs fne mea-
surements with centimeter accuracy
using carrier phase measurements.
The dual-frequency S-band con-
fguration allows the system to perform
carrier ambiguity resolution using a
wide-lane technique, while two-way
measurements account for the relative
clock drif of the platforms. LOS mea-
surements are made by measuring the
carrier phase difference between the
Carrier
frequency
S1 : 2275 MHz
(in TM band)
S2 : 2105 MHz
(in TC band)
Modulation QPSK BPSK
Channel I (ranging signal) Q (data signal) I (ranging signal)
Ranging code C/A (same as GPS) - C/A (same as GPS)
Average data rate - 7.2 or 16.8 kb/s -
TABLE 2 FFRF signal characteristics
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
46
master and slave antennas on the triplet
antenna base. Attitude, though, comes
from an external attitude sensor, such as
a star tracker.
On PRISMA, the Target satellite has
only three Rx/Tx antennas on diferent
facets to ensure full space coverage. Tus,
the Main satellite, which is the only one
equipped with a full
antenna triplet, will
perform the posi-
tion-velocity-time
(PVT) algorithms
to determine the
precise positioning
of the Target satel-
lite.
For formations
with more than two
satellites, a central-
ized PVT algorithm
uses all the measure-
ments made by all
the satellites to give
the relative position
of each satellite. See
Figure 1.
Ambiguity Reso-
lution. Carrier phase, used for LOS and
distance computation, is measured
modulus 2 pi. Tus, an integer number
of cycles remains unknown and must be
solved to reach nominal precision. Tis
operation is performed by the IAR func-
tion that is responsible for solving both
LOS and distance phase ambiguities.
In this scenario, retained precise
knowledge of LOS is necessary to cor-
rectly solve the ambiguity on distance.
Tus, IAR has to be performed frst on
LOS, then on distance. IAR on LOS
requires a satellite rotation around
antenna direction of about 40 degrees
magnitude, with the rotation charac-
teristic provided to the network proces-
sor unit sofware. Tis rotation helps to
mitigate the multipath error bias.
Te IAR on distance is done in two
steps. First, the system forms a carrier
widelane and removes its ambiguity
using code measurement smoothed over
time. Ten, in a second step, the carrier
ambiguity on S1 is removed using the
filtered widelane measurement. The
sensitivity of the latter step to multipath
is critical and requires multipath cali-
bration
Multipath Errors. A signifcant source
of errors on the LOS and distance mea-
surements comes from signal multipa-
ths created by the satellite structure sur-
rounding FFRF antennas. Tese errors
can reach several centimeters on phase
measurementsresulting in a signif-
cant degradation of FFRF precision and
potentially causing the carrier ambi-
guity resolution to fail. Indeed, when
forming the carrier phase widelane, the
carrier multipath error is amplifed. Te
IAR can be successful only if the carrier
phase error can be reduced to a few mil-
limeters on both frequencies. Current-
ly, a calibration method of multipath
errors in an anechoic chamber is under
study at CNES. Te new method maps
multipath errors in function of LOS to
provide calibration tables for in-fight
correction.
Terminal Architecture
The FFRP terminals architecture is
largely based on results of an ESA Tech-
nology Research Program study and
CNES architecture studies. It reuses
some of the components and sofware
from a 12-channel, L1 spaceborne GPS
receiver.
The following hardware sections
comprise the terminal (see the block
diagram in Figure 2 and box design in
Figure 3):
WORKING PAPERS
FIGURE 1 Two-satellite FFRF S/S conguration
FIGURE 2 FFRF terminal block diagram (generic case)
FIGURE 3 FFRF box preliminary design (TAS-F)
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
RFfront-end(RFE), whichmakes the
switchbetweenthe eight antennas on
one side and the Tx signal, the master
Rx signal and the slave Rx signal on
the other side
transmitter RF module, which gener-
ates the dual-frequency signal to be
transmitted by the Rx/Tx antennas
RF receiver (Rx) module, which per-
forms the frequency conversion and
digitization of the dual-frequency
received signal (There are two Rx
modules, one for the master signal,
one for the slave signal.)
digital (or processing) section, which
acquires and tracks signals, produc-
es ranging and angular measure-
ments, computes relative position/
velocity/line of sight, synchronizes
with TDMA sequence, controls the
transmitter section, and generates Tx
baseband signals
10.00 megahertz oven controlled
crystal oscillator (OCXO), which
provides the reference clock of the
terminal. (Tere is currently an open
trade-ofbetweenthe OCXO, already
used for the 12-channel, L1 space-
borne GPS receiver and the OCXO
that is currently being qualifed.)
Power supply section: provides sec-
ondary supply voltages fromsatellite
power bus
All the modules are then stacked
together, except RF flters (Rx and Tx),
which can be separate.
In the PRISMA case, there are two
configurations for the FFRF terminal
one for the Main spacecraf and one
for the Target. On the Main craf, the
FFRF terminal con-
nects to a unique
triplet antenna set.
As a consequence,
at module levels,
di f ferences wit h
the generic case are
limited to front end.
On the Target, the
FFRF terminal con-
nects to three, single
Rx/Tx antenna sets.
As a consequence,
at module levels,
diferences with the
generic case occur:
in the front-end
in the Rx RF module, where only one
receiver chain is present (master Rx
signal only)
in the digital processing module,
where only one digital ASICis need-
ed, because only a master Rx signal
is processed.
Data Link (ISL). Te FFRF subsystem
includes an intersatellite data link capa-
ble of transferring data to or from any
visible satellite in the formation. Two
kinds of data are exchanged:
navigation ISL: For positioning, the
RF sensor works on a cooperative
basis, meaning that RF signals are
transmittedandreceivedto andfrom
each vehicle to measure distance and
line of sight. Each sensor calculates
the results of these measurements,
then the sensors exchange the navi-
gation data to allow PVT computa-
tion for each vehicle.
transparent ISL for the on-board
computer (OBC): The RF sensor
allows data to be transferred in a
transparent manner between the
OBC of each satellite. For all FF
missions, this data exchange link is
necessary to transfer TC and TM
between master and slave satellites
for command and control purposes.
Tis OBC ISL operates with a dual-
bit rate of either 4 or 12 kilobytes.
Multipath Calibration
Signal reflections caused by the satel-
lite structure surrounding FFRF anten-
nas will be the major source of error on
FFRF LOS and distance measurements.
Tese multipath errors can reach several
centimeters on phase measurements
resulting in a signifcant degradation of
precision.
To reduce multipath efects, CNES
has set up a method of multipath-error
calibration in an anechoic chamber. (See
Figure 4.) Tis method is inspired by pre-
vious work done at CNES to improve
attitude determinationby GPS. Te mul-
tipath calibration method is depicted in
the following section.
Stanford University and NASAhave
also performed accurate multipath cali-
brations for GPS attitude determination
applications.
T. Grelier and associates performed
the first measurement campaign at
CNES in 2006, and those results can be
found in their publication listed in the
Additional Resources section at the end
of this article.
CNES performed a secondmultipath
measurement and calibration campaign
in 2007, and those results are presented
here. For this campaign, the set-up was
diferent than the previous one, and it
represents the fnal calibration that will
provide error correctiontables to be used
in 2009 prior to PRISMAs launch.
Calibration Principle. A wave front
originating froma given direction with
respect to the satellite will always refect
and difract of the various surfaces of
the spacecraf in exactly the same way.
Given such a repeatability of the wave
front scattering, we can calibrate mul-
tipath errors as a function of the LOS
signal by premeasuring these errors.
Te calibration systemmakes multipath
error maps depicting carrier phase and
interferometric phase residual errors as
a function of the azimuth and elevation
of the signal.
Experimental Set-Up. Measurements
were made in the CNES anechoic cham-
ber to avoid the creation of external
multipath to the mock-up satellite and
to create space-like conditions.
For this experiment, two radioelec-
trical mock-ups of PRISMA satellites
were built (see accompanying photos).
Te design called for three antennas to
be mounted on the Main craft on the
FIGURE 4 CNES compact range conguration, anechoic chamber
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
48
same face, allowing distance and phase
interferometric measurements. Te Tar-
get had only one antenna, allowing dis-
tance measurement.
The 2006 campaign used a FFRF
breadboard, developed in 2004 under
a joint ESA and CNES contract. This
breadboard three terminals with an
integrated 12-channel L1 receiver. Te
terminals differed from FFRF in that
signal transmission occurred in L-band
and single frequency, instead of S-band
and dual frequency.
Tis frst campaign provided prelim-
inary information about the multipath
magnitude that could occur on PRISMA
satellites. However, no information was
available for S-band multipath magni-
tude, and more important, the campaign
failed to provide information on the
magnitude of the widelane multipath
magnitude information, which is critical
for IAR. So, a new measurement cam-
paign was necessary.
As no FFRF equipment or bread-
board was available in S-band, FFRF was
replaced by a network analyzer (NA).
Te previous campaign had shown that
differences between phase measure-
ments made by FFRF and a network
analyzer were negligible.
Moreover, in this second experiment,
L-band patch antennas were replaced by
S-band helix antennas, the same ones
that will be used on PRISMA satellites
(see accompanying photos of the radio-
electrical mock-ups in anechoic cham-
ber). Multipaths measured on the mock-
ups should be representative of what will
occur for PRISMA.
Final calibration in 2009 will use
engineering models of FFRF and the S-
band helix antennas mounted on new
radioelectrical mock-ups of both satel-
lites.
Calibration Process. To perform
error mappings, the anechoic chamber
was used in its normal configuration.
A network analyzer was connected to
an S-band horn antenna located in the
beam source room and to the S-band
Saab antennas on the mock-ups (either
Main or Target) fxed to the positioning
mast.
This configuration simulated far-
feld conditions because the parabolic
refector created a perfect plane wave.
Successive mappings of Main and Target
spacecraf were made.
Error mappings took place through
continuous measurements, which con-
sist of scanning the meridians (fixed
azimuth, elevation: from -90 to +90
degrees) with a given step in azimuth
angles (two degrees). A very low (one
degree per second) rotation velocity of
the positioning mast to avoid dynamic
stress errors.
Each successive position taken by the
positioning mast recorded carrier phase
and interferometric phase measure-
ments. A postprocessing tool was used
to compute the expected measurements
(from the positioning mast positions)
and then obtain the residual errors by
subtracting the expected phase from NA
phase measurements. Complete error
maps were achieved in eight hours.
Next, multipath mapping took place,
consisting of mapping the carrier phase
errors and two of the interferometric
phase errors for the Main craft and a
carrier phase error map of the Target
craf. (Only carrier phase error mapping
was done for the Target craf because it
makes only absolute phase measure-
ments which could be diferentiated
but not direct interferometric mea-
surements.)
Calibration Method Assessment. To
evaluate the efciency of the calibration
method, a second configuration was
used with the Main craf mounted on
the positioning mast while the Target is
fxed on a second mast added in a corner
of the anechoic chamber (see accompa-
nying photo).
Measurements were performed for
diferent positions of the frst mast. Ten
residual errors were computed and com-
pared to the residual errors taken from
satellite multipath mappings.
For carrier phase measurements the
applied correction consists of the sum
of both Target and Main carrier phase
corrections. Subtracting corrections
from errors produced a calculation of
the residual error afer multipath correc-
tions. Te efciency in terms of percent-
age of correction is then computed.
Measurement Results. Following are
the key results of the calibration cam-
paigns of 2006 and 2007:
Multipath errors on carrier phase
measurements can reach several
centimeters. Statistically, multipath
error increases with elevation of
received signal (elevation is the angle
between antenna boresight and sig-
nal direction of arrival).
PRISMA radioelectrical mock-ups with S-band helix antennas in CNES anechoic chamber (2007 campaign)
WORKING PAPERS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Te standard deviation of multipath
error residual is half that of initial
multipath error, that is, multipath
calibration enables the reduction of
error by 50 percent.
Modifcation of the multi-layer insu-
lation (MLI) configuration (which
could happen because of launch
vibrations) changes the multipath
geometry confguration, thus reduc-
ing the performance of multipath
calibration. Indeed, measurements
performed before and after modi-
fcation of MLI showed that difer-
ences in multipath errors of up to
fve millimeters can occur between
measurements. So, multipath correc-
tion tables created through calibra-
tion campaigns on the ground will
probably not be fully representative
of an in-fight situation.
We concluded that, because of poten-
tially large values of multipath errors
for high elevations and reduced per-
formance of multipath correction, the
integer ambiguity resolution process
will have to be performed using reduced
values of elevation (LOS measurement
cone elevation less than 20 degrees).
Alignment in such a cone could be made
using the FFRF coarse mode.
Future Trends
As we have seen, a high number of
developments made for FFIORD will
be directly reusable for other FF mis-
sions.
For CNES, one important FF mis-
sion within a mediumtime frame is the
Simbol-X mission an astrophysical
X-band observatory (covering a con-
tinuous energy range from0.570 keV)
to be launched in 2013 in a HEO orbit.
Te CNES mission, which is a coopera-
tive efort with Italy and Germany and
France, will consist of two satellites (one
Mirror and one Detector) in close forma-
tion:
intersatellite distance = 30 meters
lateral relative position requirement
= 1 centimeter
longitudinal relative positionrequire-
ment = 10 centimeters
In this formation, the Mirror will be
the master satellite, and the Detector will
act as the slave. Dependent on accurate
formation control, Simbol-X will carry
the FFRF sensor as a main coarse and
fne metrology subsystem, which will be
fully redundant. Both GNC/formation
fight management and FFRF develop-
ments andvalidationperformedthrough
the FFIORD experiment will therefore
be directly reusable for Simbol-X.
Moreover, ESAs FFRF R&D plan
includes the current development of a
dynamic FFRF test bench, using true
hardware equipment and a dynamic
multipath and propagation channel
generator. This hardware is associ-
ated with GNC algorithms embedded
within LEON boards. The dynamic
avionics test bench aims at simulating
the behaviors of two spacecraf fying
in formation.
Te RF-conducted network is made
up of three dynamic phase shifers pro-
viding a 30-centimeter range capability
on the line-of-sight and on the distance,
plus static attenuators. Te phase shif-
ers are controlled in coherency with the
environment thanks to a Dspace box in
charge of propagating the relative states
of the satellites.
The FFRF R&D plan of ESA also
includes the upgrade of the FFIORD
third Engineering Model (EM) to extend
its fight domain up to a distance as long
as 100 kilometers between the spacecraf
without increasing the transmitted
power. Tis third EMwill also include
increased data rates (up to about 100
kilobytes per second), which could be
used for short distances between the
spacecraft. CNES is participating in
these ESA activities.
CNES is also studying the possibility
of performing line-of-sight IARwithout
any maneuvering of the spacecraf, as
well as the possibility of improving the
distance IAR thanks to the simultane-
ous use of four or, more probably, fve
dual-frequency S-band antennas and to
improved IAR algorithms. CNES will
also contribute to the evolution of the
ESA FFRF test bench with the adapta-
tion of the bench to the S-band FFRF
engineering model and with the intro-
duction of dynamic multipath. Finally,
CNES will predevelop a multipath 10-
decibel attenuating MLI, with the help of
the Commissariat lEnergie Atomique
(CEA).
Some of these R&D efforts might
beneft the Simbol-XandPROBA-3 proj-
ects. In the long term, for post-PROBA-
3/Simbol-X missions, we envisage min-
iaturization of the FFRF technology
and some simplifed designs for future
missions. One such mission could be the
return of samples fromthe planet Mars,
requiring FFRF for the recovery of the
canister containing the samples.
In low earth orbit, the future of
formation fying, including controlled
approaches and rendezvous between
manned spacecraft, relies mainly on
multistandard GNSS receivers, in L1/
E1/B1-BOC, or L5/E5a, or E5b/L3/B2
combinations, for instance. The use
of several interoperable constellations
provides more robustness and more
accuracy, especially for spacecraf with
GNSS antennas not pointed toward the
zenith, that is, those craf lacking vis-
ibility of GNSS satellites when a single
GNSS constellation is used.
WORKING PAPERS
Conguration with two satellites in anechoic
chamber (2006 CNES campaign)
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
50
Conclusion
GNSS constellations are the most practi-
cal way to perform RF formation fying
in LEO, and autonomous two-way trans-
mission of GNSS-like S-band signals is
a better way to perform FFRF in HEO
or within the Lagrange points. PRISMA
is a unique opportunity in Europe, both
technically and programmatically, to
validate under real conditions the basic
feature of any non-LEO future FF mis-
sionthe RF-based autonomous metrol-
ogy, using GPS-C/A-like signals and
techniques.
By early 2009, an autonomous RFFF
sensor shall be fying onboard the PRIS-
MA satellites. Tis sensor will use GPS-
like signals in S-band. Later, in 2012,
the ESA PROBA-3 and CNES Simbol-X
spacecraf will demonstrate the technol-
ogy in scientifc missions in HEO orbit.
However, to achieve the required
accuracy, IAR on carrier phase will be
needed. For this to succeed, multipath
errors will have to be mitigated by cali-
brating the multipath on the ground to
make in-fight corrections.
Manufacturers
PRISMA is a Swedish National Space
Board (SNSB) mission, undertaken as
a multilateral project with additional
contributions from CNES, the German
DLR, and the Danish DTU. Te prime
contractor is the Swedish Space Corpo-
ration (SSC), Solna, Sweden, responsible
for design, integration, and operation of
the space and ground segments, as well
as implementation of in-orbit experi-
ments involving autonomous forma-
tion f lying, homing and rendezvous,
and three dimensional proximity opera-
tions. It employs Phoenix GPS receivers
developed by DLR that incorporates the
GP4020 chip from Zarlink Semicon-
ductors, Ottawa, Ontario, Canada.
The FFRF subsystem development
is currently in phase C/D, with Tales
Alenia Space-France, Toulouse, France,
as the prime contractor on both the sub-
system and FFRF terminal level. In turn,
TAS-F is relying on the following sub-
contractors:
Tales Alenia Space Espaa (TAS-
E, Madrid, Spain) for development
of the RF modules of the FFRF ter-
minal (RF front end, RF transmitter
section, RF receiver section), which
incorporate a digital technology
and sofware coming from the TAS
TOPSTAR 3000 spaceborne GPS
receiver.
GMV (Madrid, Spain) for develop-
ment of the navigation software,
including implementation of PVT
algorithms
Tales Avionics (France) for devel-
opment of the FFRF terminal signal
processing sofware
Saab Space (Gteborg, Sweden) for
the S-band helix antennas
an OXCO from TES Electronic Sys-
tems, of Bruz, France, is currently
being qualified; an OXCO from
Composants Quartz et Electro-
nique (Temex), Mougins, France is
also being used.
[1] Bourga, C., et al., Autonomous Formation Fly-
ing RF Ranging Subsystem, Proceedings of ION
GNSS 2003, Portland, Oregon, USA, September
2003
[2] Cledassou, R., Ferrando, P., Simbol-X: An
Hard X-Ray Formation Flying Mission, Focusing
Telescope in Nuclear Astrophysics Gamma Wave
Workshop, Bonifacio, France, September 2005
[3] Garcia-Rodrguez, A., Formation Flight (FF)
Radio-Frequency (RF) Metrology. ESA/ESTEC
Technology Dossier, Issue 1.2, Noordwijk, the
Netherlands July 2008
[4] Godet, J. et al., Improving Attitude Determi-
nation of Satellites, Internationnal Workshop on
Aerospace Applications of the GPS, Breckenridge,
Colorado, USA, February 2000
[5] Harr, J., et al., The FFIORD Experiment: CNES
RF Metrology Validation and Formation Flying
Demonstration on PRISMA, 3rd International
Symposium on Formation Flying, Missions and
Technologies, Noordwijk, the Netherlands, April
2008
[6] Issler, J., et al., Lessons Learned from the
Use of GPS in Space; Application to the Orbital
use of GALILEO, Proceedings of ION GNSS 2008,
September 2008
[7] Lestarquit, L., et al., Autonomous Formation
Flying RF Sensor Development for the PRISMA
Mission, Proceedings of ION GNSS 2006, Fort-
Worth, Texas, USA, September 2006
[8] Persson, S. et al., PRISMA: An Autonomous
Formation Flying Mission, Small Satellite Systems
and Service Symposium, Chia Laguna, Sardinia,
Italy, September 2006
[9] PROBA-3 <http://www.esa.int/techresosur-
ces/index.html>
Authors
Thomas Grelier has
been a navigation engi-
neer in the Transmission
Techniques and Signal
Processing Department
at CNES since December
2004. He graduated from
the French engineering school Supelec and
received an M.S. in electrical and computer engi-
neering from Georgia Tech (USA). Galileo signal
processing is one of two main areas of research. He
analyzed GIOVE-A, GPS IIR-M, Beidou-1 S-band,
and modernized GLONASS signals. Grelier has also
developed various Galileo #5 ALTBOC tracking tech-
niques and analyzed their theoretical performanc-
es. Thomas Grelier is also technical responsible of
the FFRF equipment of the PRISMA satellites.
Al ber t o Gar ci a-
Rodrguez is a radio-
navigation systems
engineer of the European
Space Agency. He is
involved since 2000 in
activities related to GNSS
space receivers, GNSS applications and Formation
Flying. He worked previously at GMV (Spain) per-
forming GNSS system studies for the EGNOS and
Galileo projects. He has a degree in Telecommuni-
cations Engineering from the Universidad Politc-
nica de Madrid (T.U. Madrid).
Eric Pragin is a TMTC
expert in the CNES Trans-
mission Techniques and
Si gnal Pr ocessi ng
Department having
worked in this area for
about 20 years. He
designed the high performance Rx/Tx used on
board the Myriade microsatellite family, also used
for the intersatellite link of cometarian probes
Rosetta (ESA/CNES/DLR) and Deep Impact (NASA).
He was involved in the Martian Data Relay equip-
ment supplied by CNES for Mars 96 Russian probe,
and the U.S. probes Mars Observer and Mars Glob-
al Surveyor. He integrated a GPS receiver on the
HETE1 spacecraft of MIT and NASA. He is respon-
sible of the intersatellite link of ROSETTA and RF
expert for the FFRF equipment of PRISMA. He is
also in charged of predevelopment activity for
CNESs next generation of high performance TMTC
susystems in S- and X-band for small platforms.
Laurent Lestarquit is a navigation signal expert
at the CNES Transmission Technique and Signal
WORKING PAPERS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Processing Department
(TT). He was a member
of theGalileoSignal Task
Force andcontributedto
the definition of the
Galileo signal and pro-
vided support to the
2004 US-EU agreement on GPS and Galileo. He
invented the constant envelope four-code ALT-
BOC modulation.
Jon Harr works at CNES
i n t he Et udes de
l Uni ver s pr oj ect
department. He is cur-
rently responsible for
coordinating the CNES
participation on the
PRISMA mission, as project manager for the
FFIORDexperiment. Hehas backgroundas aradio
and remote sensing engineer, with several years
of experience as specialist onradar altimetryand
TTCsystemsonboardCNESsmall satellitemissions
such as Jason, Demeter, and Parasol.
Dominique Seguela is
inthe CNES R&Ddepart-
ment. Sheis responsible
for several R&D areas,
including onboard-
ground processing and
hardware, andformation
ying, and supervises the CNES mission group of
PRISMA. Shewas involvedinsystemandsatellite
activities for the SPOT spacecraft family.
Jean-Luc Issler is in
charge of the Transmis-
sionTechniquesandSig-
nal Processing Depart-
ment at CNES. He is
involvedinthedevelop-
ment of several types of
GNSS, FFRF, and TMTC equipment in Europe for
space or ground users. He is a French delegate to
the GALILEO Signal Task Force. Issler received in
2004 the Astronautic Prize of the AAAF (French
aeronautical andspaceassociation), andin2008
the EADS Prize of science and engineering deliv-
ered by the French Academy of Sciences, mainly
for his technical work on Galileo signals and
spaceborne GNSS equipment.
Jean-Baptiste Thevenet was a doctoral and
postdoctoral researcher in numerical optimiza-
tion at ONERA and CNES for five years. He has
beena navigationsystemengineer at Thales Ale-
nia Space since early 2006, mostly in charge of
systemaspects for FFRF subsystems.
Christophe Ensenat graduatedfrom E.N.S.E.E.H.T
Toulousewithanengineeringdegreeinelectron-
ics. He serves now at Thales Alenia Space, Tou-
louse, as equipment project manager, within the
digital section of the Industrial Unit. His area of
expertiseincludes thedigital units of radar altim-
eters, GPS receivers, and RF sensors.
Nicolas Wilhelmhasbeen
working on GNSS space
equipment at Thales
Alenia Space for more
then 10 years. He has
beenresponsiblefor the
development of theTOP-
STAR 3000 space receiver in late 1990s and
involvedinformationyingRFstudies anddevel-
opment withESAandCNESsincepreliminarystud-
ies in 2002.
Ana Maria Badiola is the
technical responsiblefor
the RF modules of the
FFRF terminal at Thales
Alenia Space Espaa.
She has participated in
development of theFFRF
terminal modules since 2004, including prelimi-
nary studies, phase B, and the current phase C/D
of the terminal. She has a wide experience in S-
band transponders (for LEO applications of the
PROTEUS satelliteplatform, ATVandGalileo) and
X/C/Ku-bandtelemetrytransmitters as technical
responsible and RF design engineer.
Pablo Colmenarejo works
for more than 10 years
for GMV (Spain), where
he is responsible of the
GNC Division. With an
Aeronautical Engineer-
ing academic back-
ground, he has developed most of his profes-
sional experienceinprojects relatedtoformation
ying and rendezvous and docking and on GPS
and radio frequencyrelated technologies. He is
actively involvedinthe FFRF metrology develop-
ment anduse for missions suchas DARWIN, Mars
Sample Return, PRISMA, and PROBA3.
Relations with China overall will
pose a sizable and complex challenge
to the new U.S. president. Te security
issues associated with GNSS technol-
ogy make this a particular prickly
subject. And the lack of transparency
in Chinas Compass program, its efec-
tive control by the Peoples Liberation
Army (Chinas defense establishment),
and unexpected events such as Chinas
January 2007 anti-satellite test (ASAT)
have combined to limit face-to-face
talks on the subject.
Te constraints on bilateral
approaches to China have made the
success of the ICG thus far all the more
important and noteworthy. Although
the most recent ICG meeting in Pasa-
dena, California, in early December
marked the return of a more cautious
mode, it probably refected a desire
to digest and sort out the surprising
advances made in the previous meeting
in Bangalore, India.
How far this internationalist
impulse might carry GPS is yet to
be determined. On the one hand, it
might go as far as the joint secretariat
established for the COSPAR/SARSAT
with voluntary contributions to handle
paperwork (glossary, terms of refer-
ence) and exchange information on
system operations. On the other, it
could remain a series of arms length
relationships that manage to reduce the
inevitable conficts that arise among
systems operating in a common physi-
cal and technological space.
A New Regime
Overall, GPS is in a good place, with a
sound foundation from the past eight
years of work. But much of the current
policy is designed for old situations and
problems.
Management and advocacy of the
system is divided in all ways: civil,
military and within agencies. Markets
and applications are exploding, under-
scoring the systems status as a critical
infrastructure as does the geopoliti-
cal situation. And competition from
other GNSS providers is growing.
Tis is all a good thing a critical
infrastructure on solid footing, but a
little behind the times, with competi-
tors goading the U.S. leadership (so, its
not a good time to relax).
What is needed now is a new,
strong advocate, a GPS/PNT champion
who can connect the dots and make
sure the Obama administration knows
whats at stake.
Te situation is a little like that at
the beginning of the biblical Exodus
story: Now a new king arose over
Egypt, who did not know Joseph.
Somebodys got to go tell the King.
REGIME CHANGE continued from page 27
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
52
Te European Space Agency (ESA) has
awarded Tales Alenia Space Italia (TAS-
Italia) two contracts for development of
Galileo ground station receiver equipment.
One contract is for Galileo In-Orbit
Validation Element (GIOVE) phase A/B
ground station receivers capable of track-
ing the multiplex binary ofset carrier
(MBOC) signal. MBOC is common to
both the Galileo Open Service (OS) and
the new GPS L1 civil signal, which will
be transmitted beginning with the GPS
III generation of satellites.
Te other TAS-Italia contract is for
the development of a receiver bread-
board version of a new generation of
Ranging and Integrity Monitoring Sta-
tion (RIMS) receivers.
Te GIOVE MBOC receiver will
be capable of managing GPS channels
on L1 and L2P, Galileo channels on L1,
E5a, E5b and E6, and GIOVE channels
including the MBOC modulation on L1.
Te receiver will incorporate confgu-
rability features for test purposes.
TAS-Italias MBOC receivers will be
installed in an existing Galileo Experi-
mental Sensor Station (GESS) at ESA/
ESTEC in Noordwijk, Netherlands, and
another in Italy on a site to be defned
by the Italian Space Agency (ASI). Te
interest in this development is linked to
the possibility of carrying out early feld
tests of receiver components that will
then be part of the Galileo Ground Mis-
sion Segment (GMS) Galileo Reference
Chain (GRC) installed in each Galileo
Sensor Station.
ASI is supporting this development
in close cooperation with ESA. Tales
Alenia Space Italia will be completely
responsible for the defnition, design,
manufacturing, and test of the receivers.
In addition to the deployment of the
GIOVE MBOC receivers, the contract
also includes a parallel development of an
RF application specifc integrated circuit
(ASIC) prototype specifcally designed
for the demanding performance require-
ments of ground reference stations.
Development of the new generation
of RIMS receiver is part of the ESA GNSS
Evolution program designed to pave the
way for the European navigation infra-
structure. Te new RIMS receiver will be
capable of managing GPS channels on
L1, L2, and L5; Galileo channels on L1,
E5a, E5b, and E6; GLONASS channels
on L1; and satellite-based augmentation
system (SBAS) geostationary satellite
channels on L1 and L5. Tales Alenia
Space Italia will lead a consortium of
companies from France, Spain, and Italy.
Ventures
Blaupunkt Picks IfEN
NavX-NCS to Test Car Nav
Blaupunkt GmbH, a Hildesheim, Ger-
manybased automotive electronics
manufacturer, has selected the NavX-
NCS multiconstellation, multifrequency
GNSS simulator from IfEN GmbH
(Poing, Germany) as the GPS reference
for production testing of Blaupunkts car
navigation systems.
Te NavX-NCS RF constellation sim-
ulator supports all present and planned
Galileo and GPS frequencies (E1/L1,
L2C, E5AB/L5, E6) in one box, now
including the composite binary ofset
carrier (CBOC) signal structure for the
Galileo Open Service on E1.
Blaupunkt, which launched the frst
European car navigation system in 1989,
will use the IfEN simulator for fnal
production testing of the new original
equipment that use GNSS signals to
provide positioning for map-based and
turn-by-turn driver assistance systems.
PCTEL Acquires Wi-Sys
Illinois-based PCTEL, Inc., announced
January 5 that it has acquired Wi-Sys
Communications Inc., an Ottawa,
Ontario, Canadabased company that
specializes in GPS antenna and receiver
technology. PCTEL will pay $2.1 million
for Wi-Sys and plans to fully integrate
the latter companys operations into its
Antenna Products Group (APG).
In 2008 Wi-Sys proprietary products
and technology generated approximately
$2.2 million in revenue. Te companys
low-noise and low-current products
include a family of magnetic and fxed-
mount GPS antennas, OEM antennas, and
integrated GPS receiver/antennas targeting
the diversifed professional GPS market.
Wi-Sys products are used in covert
antenna applications, commercial vehicle
tracking, and precision agriculture. Its
product line includes antennas for GPS
L1/L2 and for Iridium and GlobalStar
satellite communications systems.
With the acquisition, PCTEL will
extend its product ofering and technol-
ogy base, which includes timing, avia-
tion, land-vehicle, and WAAS-capable
antennas. Te Wi-Sys product line will
also be sold through PCTELs existing
distribution channels.
People
U.S. Space Commerce
Ofcial Joins ITT
Edward M. Morris, formerly execu-
tive director for the U.S. Ofce of Space
Commercialization, Department of
Commerce, has joined ITT Space Sys-
tems Division (SSD) in Rochester, New
York, as executive director of strategic
business development.
In his new position, Morris will be
responsible for strategic program and
business development of GPS navigation
systems and additional space-related
capabilities. ITT SSD has designed and
built the navigation payloads fown on
every GPS satellite ever launched more
than 60 spacecraf. ITT is also part of a
team led by Raytheon Company compet-
ing for a contract to build the next-gen-
eration GPS ground control segment.
In his role at the U.S. Department of
Commerce, Morris served as the excu-
tive branch industry liaison where he
helped the U.S. government optimize the
use of space goods and services. He was
also responsible for implementing PNT,
commercial remote sensing, and related
space policies.
INDUSTRY VIEW
Thales Alenia Space Italia Wins
ESAGalileo Receiver Contracts
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
GNSS Receivers
u-blox 1.8-Volt GPS Module
u-blox has released a new low-voltage
GPS module the NEO 5D. Operating
at 1.8 volts, the module requires at least
40 percent less power compared to its
predecessors in the 50-channel u-blox
5 engine product line: 72 mw @ 1.8V in
the Eco mode and 77 mW @ 1.8V in
the maximum performance mode. Te
module comes in a 12162.4-millimeter
package and employs u-bloxs Super-
Sense acquisition and tracking technol-
ogy with a reported reacquisition and
tracking sensitivity down to 160 dBm.
Cold-start acquisition sensitivity is -143
dBm with a time to frst fx of 32 seconds.
Te ROM-based architecture does not
require an external Flash EPROM, mak-
ing it suitable for high-volume consumer
products. Like all of u-bloxs modules,
the new NEO 5D and NEO 5G support
u-blox AssistNow, the freely available
service that reduces the time to frst fx to
one second by providing satellite ephem-
eris and almanac position data, as well as
accurate time and other satellite infor-
mation over a mobile network, according
to the company. u-blox, Talwil, Switzer-
land. PRODUCT 122.
Dual-Frequency RTK GPS
Hemisphere GPS ofers the dual-fre-
quency R220 GPS receiver, a 39-channel
that ofers real-time kinematic (RTK)
operation with 12 channels of L1 C/A
code, 12 channels of L1 P-code, and 12
channels of L2 P-code tracking. Te
other three channels can be used for
tracking satellite-based augmentation
systems including OmniSTARs com-
GNSS INSIDE
Visit the Inside GNSS website at www.insidegnss.comfor click-through viewing of additional
information about the new products and manufacturers described here.
process requiring a single 1.8V power
supply with a typical power consumption
of less than 20 mW. QHx220 devices are
ultra-small (less than one square milli-
meter) and available in a wafer level chip
scale package (WLCSP) or in 3x3-milli-
meter quad fat no-lead (QFN) packages.
Quellan Inc., Santa Clara, California,
USA. PRODUCT 124.
Antenna
GPS Quadrilar Antenna
Maxtena ofers its RadioMax M1575CQA
GPS quadriflar antenna. Its compact
size and groundplane independence
is designed to simplify integration into
handsets. Te M1575CQA directly con-
nects to an RF transceiver chip without
baluns or matching networks. RadioMax
technology reduces currents that are
driven onto the ground plane, making
the antenna more resistant to the efects
of human exposure (for better perfor-
mance in handsets). According to the
company, RadioMax provides efciency
of 38 percent (total spherical) and gain
of -1.3 dBic. Te beamwidth (3dB) is 140
degrees (both axes) and bandwidth is
10 MHz (1570.421580.42 MHz). Cross
Pole rejection is 15 dB; axial ratio, 0.2 dB;
front to back ratio, 15 dB; VSWR 1.3: 1;
and impedance (diferential) 100 Ohms.
Overall dimensions are 10 millimeters
(diameter) by 20 millimeters (length).
Te Maxtena FilterMax technology is
available if more out-of-band rejection
of undesired signals is needed. Maxtena,
Inc., Blacksburg, Virginia, USA. PRODUCT
125
mercial HP and XP diferential correc-
tion services or three additional chan-
nels of L1 C/A-code tracking. Te unit
measurements are: height, 45 millime-
ters (1.77 inches), width 114 millimeters
(4.49 inches), length 160 millimeters (6.3
inches), and weight 0.54 kilograms (1.19
pounds). Te unit is intended for pro-
fessional mapping, guidance, machine
control, and navigation applications.
Integrated L-band tracking powers
down when not in use, and OmniSTAR
subscriber access permits remote activa-
tion via satellite. Te R220 incorporates
Hemisphere GPSs COAST technology,
which maintains accurate solutions for
40 minutes or more afer loss of a dif-
ferential signal, according to the com-
pany. A position output rate of 10 Hz is
standard but up to 20 times per second is
available, as is raw GPS data output. Te
unit uses a standard USB Flash drive for
data logging. Hemisphere GPS, Calgary,
Alberta, Canada. PRODUCT 123.
GNSS Hardware
RF Noise Canceller for GPS
Quellan ofers a GPS Noise Cancellation
Starter Pack for its QHx220 RF noise
canceller: an IC is incorporated into a
handset to cancel electromagnetic inter-
ference (EMI) and improve signal integ-
rity. Typical applications are a sensitive
GPS receiver co-located with a GSM/
CDMA transmitter or a GPS system
corrupted by digital noise. Te QHx220
reduces EMI by taking the sampled
interference source as an input and
fltering the sampled signal to emulate
and subsequently remove the incurred
interference on the received victim sig-
nal. Te QHx220 is composed of a chan-
nel emulator (CE) and two low noise
amplifers (LNAs) in the sampling path.
Te CE provides phase shif and attenu-
ation to tune out the unwanted noise
at the correction point. Te QHx220 is
designed in a standard 0.18 m CMOS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
54
MARCH 18-19
CERGAL 2009
Oberpfaffenhofen, Germany
GNSS systems certication
conference will focus on GNSS
simulation, testing tools and
infrastructure. < www.dgon.de/>
MARCH 18-19
SPIRENT FEDERAL GPS
TRAINING CONFERENCE -
EAST
Bethesda, Maryland USA
Identical GNSS test systems
training sessions on each coast.
< www.spirentfederal.com/GPS/
Training/>
MARCH 23-APRIL 9
SATELLITE NAVIGATION
SCIENCE & TECHNOLOGY FOR
AFRICA
Trieste, Italy
Focus on space weather and
ionospheric research activities
to help initiate space science
March
MARCH 3-5
MUNICH SATELLITE
NAVIGATION SUMMIT 2009
Munich, Germany USA
Features include updates on all of
the GNSSes, new applications, a visit
to the Galileo Test Environment and
the European Satellite Navigation
Competition 2009 kickoff. <www.
munich-satellite-navigation-
summit.org>
MARCH 3-9
CEBIT HANOVER 2009
Hanover, Germany
Massive trade fair with a dedicated
Telematics and Navigation section.
<www.cebit.de/53659>
MARCH 5-6
NORDIC INSTITUTION
OF NAVIGATION E-NAV
CONFERENCE
Bergen, Norway
Discussions on IMOs maritime e-
navigation vision plus space, land,
and air topics. <www.nornav.org>
MARCH 7-14
IEEE/ AIAA AEROSPACE
CONFERENCE 2009
Big Sky, Montana USA
Of particular interest to Inside GNSS
readers is Track 4: Communication
and Navigation Systems and
Technologies. Website: < www.
aeroconf.org/>
MARCH 10-13
GEOFORM + 2009
Moscow, Russia
Four conferences will take place
under the GeoForm umbrella,
including GeoWAY, covering
intelligent transportation and
satellite navigation.
< www.geoexpo.ru/2009/eng/
about/>
GNSS TIMELINE
February
FEBRUARY 3-5
AUVSI UNMANNED SYSTEMS
PROGRAM REVIEW 2009
Washington DC USA
Government UVS programs review
- air, ground and maritime. <www.
auvsi.org/programreview/index.cfm>
RESCHEDULED
GPS-WIRELESS 2009
San Francisco, California
Previously set for February 45, this
13th annual conference for mobile
information companies,converging
navigation, location-based services,
and eet and asset management
for enterprise markets has been
rescheduled to April 2829. < www.
gps-wireless.com>
FEBRUARY 12-13
LOCATION SUMMIT 2.0
Hyderabad, India
A conference on positioning and
navigation with the theme Towards
Collaborative Models. <www.
locationsummit.com>
FEBRUARY 16-19
GSMA MOBILE WORLD
CONGRESS
Barcelona, Spain
Mobile industrys largest exhibition
with 50,000 attendees, awards for
best new technology and more.
Formerly 3GSM World Congress.
<www.mobileworldcongress.com>
FEBRUARY 23-27
NAVTECH SEMINARS 2009
San Diego, California USA
A new course on Assisted GNSS joins
many others at Doubletree. < www.
navtechgps.com>
FEBRUARY 23-25
TRIMBLE DIMENSIONS 2009
Las Vegas, California USA
Positioning conference with 300
sessions and a keynote by the Mars
Exploration Rover Project principal
investigator. < www.trimbleevents.com>
ADVERTISERS INDEX
Company Page Number
746th Test Squadron 25
CAST Navigation 13
CTIA Wireless 2009 26
GPS Networking 15
GPS Source 27
Hemisphere GPS 7
IfEN 9
L3/Interstate Electronics Inside Front Cover
NavCom Technology Inc. Inside Back Cover
NavTechGPS 21
NovAtel, Inc. 23, Back Cover
Pacic Crest 17
PCTEL 29
Septentrio 5
Spirent Communications 3
u-blox 19
research programs in African
universities. <http://cdsagenda5.
ictp.trieste.it/>
MARCH 24-25
SPIRENT FEDERAL GPS
TRAINING CONFERENCE -WEST
Anaheim California USA
Identical GNSS test systems training
sessions on each coast. < www.
spirentfederal.com/GPS/Training/>
MARCH 31-APRIL 3
CTIA WIRELESS 2009
Las Vegas, Nevada USA
Nobel prizewinner and former U.S.
Vice President Al Gore will keynote
the big wireless industry association
show, with education sessions
April 1 and 2. The Blueprint
for Convergence track targets
engineers and other technology-
oriented participants. (See ad on
page 26)
<www.ctiawireless.com/
Need more information?
Its online!
<WWW.INSIDEGNSS.COM/EVENTS>
___
___
_________
___
__________
___
________
________________
___
_______
___
___
_______________
______
_____
___
_______
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 20090101 Jan 2009

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

GNSS 20090101 Jan 2009

Transféré par

Droits d'auteur :

Formats disponibles

_____________________________________________________

Vous aimerez peut-être aussi