Inside GNSS Magazine: Off The Shelf and Into Space, Volume 9 Number 4 July/August 2014

July/August 2014 | Engineering Solutions from the Global Navigation Satellite System Community | www.insidegnss.
com
OFF THE SHELF
AND INTO
SPACE
ENVIRONMENTAL SENSING | GNSS Revolution
GNSS SDR | A Toolbox for All Systems
WASHINGTON VIEW | New GPS Leadership
Contents | Zoom in | Zoom out Search Issue | Next Page For navigation instructions please click here For navigation instructions please click here
READ IT FIRST IN...
E-news and analysis twice a month,
with exclusives by Washington
correspondent Dee Ann Divis and
Inside GNSS editor Glen Gibbons
Subscribe now
www.insidegnss.com/enews
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
July/August 2014 | Engineering Solutions from the Global Navigation Satellite System Community | www.insidegnss.com
OFF THE SHELF
AND INTO
SPACE
ENVIRONMENTAL SENSING | GNSS Revolution
GNSS SDR | A Toolbox for All Systems
WASHINGTON VIEW | New GPS Leadership
Gain perspective
in real-world
GNSS simulation
The GNSS simulator in the SMBV100A vector signal generator
Etersive, irfetible sirulatior of ChSS scerarios is a tlir of tle ast.
how ]ou car easil] ard cost-effectivel] test ]our satellite receivers urder
realistic corditiors.
t Cores witl a variet] of redefred ervirorrert rodels sucl as
"rural area', "urbar car]or', "bride' ard "lilwa]'
t Allows fetible corfuratior of realistic user ervirorrerts ircludir atro-
sleric rodelir, obscuratior, rultiatl, arterra claracteristics
ard velicle attitude
Ile RSMBV1JJA ererates all relevart corruricatiors ard broad-
castir stardards sucl as LIE, hSPA+, wLAh, h0 Radio' ard FM stereo.
Io frd out rore, o to
www.rohde-schwarz.com/ad/smbv-gnss
SMBV-GNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Price and performance
in the palm of your hand
KVH Industries, Inc. World Headquarters: 50 Enterprise Center | Middletown, RI 02842 U.S.A. | info@kvh.com | +1 401.847.3327
2014 KVH Industries, Inc. KVH is a registered trademark of KVH Industries, Inc.
Bias instability of 0.1/hr, 1
Angle Random Walk (ARW) of 0.012//hr
3 axes of DSP-1750 FOGs combined with 3
MEMS accelerometers
Easy integration with other sensors (GPS,
DVL, ODO, etc.)
Compact, precise & accurate
Commercial-grade
Breakthrough price & performance
Available immediately
KVH 1750 IMU
Three of KVHs high-performance
DSP-1750 gyros are at the
heart of this advanced new system.
The Power Within
Download the White Paper:
kvh.com/1750IMU
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_______
4 InsideGNSS J ULY/AUGUS T 2014 www.insidegnss.com
ON THE COVER
48
GPS Receiver
Performance On
Board a LEO Satellite
Te Navigation and Occultation
eXperiment
Andr Hauschild, Markus
Markgraf, and Oliver
Montenbruck
A test and evaluation program demonstrates that a commercial
GPS receiver can operate as a spaceborne research tool.
TECHNICAL ARTICLES
36
Environmental Sensing
A Revolution in GNSS Applications
Kristine M. Larson, Eric E. Small, John J. Braun,
and Valery U. Zavorotny
Increasingly sophisticated uses of GNSS observables
have led to a new era in remote sensing. A team of
researchers describe the results of the applications
of interferometric refectometry to measure snow
depth, vegetation water content, and soil moisture.
58
A Universal GNSS Software
Receiver Toolbox
For Education and Research
Sanjeev Gunawardena
GNSS sofware defned radio receivers and associated research
are proliferating rapidly as computer processing power increases
and costs decline. Te author describes the latest version of a
universal GNSS SDR processing toolbox that is distributed as a
plug-in for high-level algorithm development.
COLUMNS
22
Washington View
New Leaders at the GPS Helm
Dee Ann Divis
Changes across the board as new
threat to spectrum emerges.
TABLE OF CONTENTS
JULY/AUGUST 2014
VOLUME 9 NUMBER 4
TOC BY THE NUMBERS
12 Tinking
Aloud
14 360 Degrees
15 GNSS Hotspots
ARTICLES
22 Washington
View
30 GNSS
Solutions
32 GNSS &
Geohazards
36 Environmental
Sensing
48 GPS Receiver
on a LEO
Satellite
58 GNSS SDR
Toolbox
68 Working
Papers:
Assessing
GNSS Signal
Acquisition
DEPARTMENTS
80 Industry View
82 Advertisers
Index
82 GNSS Timeline
Cover photo courtesy of DLR
German Aerospace Center
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Copyright 2014, Trimble. All rights reserved. All other trademarks are the property of their respective owners. TPC095 (06/14)
www.trimble.com/gnss-inertial
Trimble GNSS OEM
Simplied Integration
with single-board GNSS + Communications.
By integrating UHF or WiFi/Cellular communications on the
same module, the task of receiving RTK corrections is signicantly
simplied speeding integration and development times.
Trimble BD920-WG3
60mm x 55mm x 19mm
GNSS RTK receiver module
with WiFi/Cellular modem
Trimble BD930-UHF
60mm x 55mm x 15mm
GNSS RTK receiver module
with UHF receiver
Booth H
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
With the new Multi-Frequency, Multi-GNSS RF Constellation
Take GNSS Testing to
Simple signal selection
Sky plot
Comprehensive constellation editor
Real-time remote 6-DOF trajectory and
simulation control
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Simulator from Spirent
a New Level
ULTIMATE exibility
SUPREME performance
COMPREHENSIVE capability
Antenna pattern and multipath mask editors
Import or dene orbits
Flexible vehicle models
Global: spirent.com/positioning
US Government & Defense: spirentfederal.com Our best thinkings done inside the box
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
DEPARTMENTS
12 Thinking Aloud
Te Next Big Mac
Glen Gibbons
14 360 Degrees/ GNSS
Hotspots
News from the world of GNSS
80 Industry View
82 Advertisers Index
82 GNSS Timeline
Calendar of Events
30
GNSS Solutions
How do measurement errors propagate into GNSS position estimates?
Mark Petovello
68
Working Papers
Assessing the Performance of GNSS Signal Acquisition:
New Signals and GPS L1 C/A Code
Myriam Foucras, Bertrand Ekambi, Fayaz Bacard,
Olivier Julien, and Christophe Macabiau
TABLE OF CONTENTS
www.labsat.co.uk/trial
FREE 14 day trial
One touch record/replay of RF signals
Recreate real world conditions
Single or dual channel
GPS, GLONASS, Galileo, Beidou, QZSS and SBAS
Signal simulation software available
Free library of worldwide recordings & simulations
LabSat3GPS simulator
record replay simulate
ION GNSS+ 2014
Visit us at
Booth 108
GNSS Hot Spots,
pages 1415
SPONSORED SECTION: THOUGHT LEADERSHIP SERIES
32 GNSS & Geohazards Ken Hudnut
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
www.insidegnss.com J ULY/AUGUS T 2014 InsideGNSS 9
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
July/August 2014 Volume 9/Number 4
EDITORIAL
Editor & Publisher Glen Gibbons glen@insidegnss.com
Art Director Gwen Rhoads
Contributing Editor for Working Papers
Gnter Hein Guenter.Hein@unibw-muenchen.de
Contributing Editor for GNSS Solutions
Mark Petovello mark.petovello@ucalgary.ca
Contributing Editor for Washington View
Dee Ann Divis deeann@insidegnss.com
Contributing Editor for Brussels View
Peter Gutierrez peter@insidegnss.com
Staff Writer/Editor Eliza Schmidkunz
Technical Editor Fiona Walter
Cover Design: Christine Waring
Web Designer/Webmaster Mike Lee
Web Editor Sierra Robinson
Circulation Director Peggie Kegel
ADVERTISING SALES AND BUSINESS DEVELOPMENT
Director Richard Fischer richard@insidegnss.com
Mobile: 609-240-1590
Ofce: 732-722-7506 richard@insidegnss.com
PUBLISHED BY GIBBONS MEDIA & RESEARCH LLC
Managing Partner Glen Gibbons glen@insidegnss.com
Director/Partner Eliza Schmidkunz eliza@insidegnss.com
1574 Coburg Road No. 233
Eugene, Oregon, 97401-4802 USA
Telephone: 408-216-7561
Fax: 408-216-7525
Copyright 2014 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 Internet,
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) (Online version ISSN 2329-2970) is a controlled circulation magazine,
published six times a year (bimonthly). 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
Chungnam National University, Daejon, Korea
JULES MCNEFF
Overlook Systems Technologies, Inc., Vienna, Virginia, USA
PRATAP MISRA
Tufts University, Medford, Massachusetts, USA
BRAD PARKINSON
Stanford University, Palo Alto, California, USA
TONY PRATT
Professor and Consultant, United Kingdom
SERGEY G. REVNIVYKH
ISS Reshetnev, Zheleznogorsk, Russian Federation
MARTIN RIPPLE
berDash Pty Ltd, Australia
CHRIS RIZOS
University of New South Wales, Sydney, Australia
TOM STANSELL
Stansell Consulting, Rancho Palos Verdes, California, USA
JACK TAYLOR
The Boeing Company, Colorado Springs, Colorado USA
JRN TJADEN
European Space Agency, Noordwijk, The Netherlands
A.J. VAN DIERENDONCK
AJ Systems, Los Altos, California, USA
FRANTISEK VEJRAZKA
Czech Technical University, Prague, Czech Republic
PHIL WARD
Navward Consulting. Garland, Texas, USA
CHRISTOPHER K. WILSON
Vehicle Data Science Corporation, 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 to the print and/or digital versions are available to
qualied readers who work in GNSS-related companies, organizations, research
institutes, government agencies, and the military services.
You may also change your address, renew, or unsubscribe online:
WWW.INSIDEGNSS.COM/SUBSCRIPTIONSERVICES
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
_______________
H
eres the coolest technology-
meets-ingenuity-meets-sus-
tainable-economics story that Ive
heard in a long time: the International
Sun-Earth Explorer-3 (ISEE-3) Reboot
Project, a crowd-funded rescue mis-
sion to repurpose a 36-year-old NASA
spacecraf.
Operating out of an abandoned Mc-
Donalds restaurant near NASAs Ames
Research Center in California, a team
led by former NASA employee Keith
Cowing and space technologist Den-
nis Wingo cut a deal with the U.S. space
agency. Te group would try to wake up
the ISEE-3s onboard systems, refre its
engines, bring the craf back into Earth
orbit, and put the ancient mariner to
work on new tasks.
Tats probably the best use of an
abandoned McDonalds hamburger
stand that Ive heard. (Actually, its
the only one Ive heard of. Who could
imagine a place and people without
the need for a Big Mac?)
Perhaps the Golden Arches caught
their attention, but in place of the icon-
ic curved brandmark, these visionaries
saw the arcs of possible trajectories,
intersections in space and time.
In any case, afer digging through
old documents in the basements of
retired NASA engineers, the reboot
team recreated the command language
and began executing their mission.
To their surprise they discovered that
many of the ISEE-3s sensors were still
operable afer all these years and hav-
ing absorbed fve times their design
level of space radiation. And there was
even a little gas lef in the tank!
Why did they do it, Cowing asked
rhetorically in a July 19 New York Time
op-ed article? First, because we could:
Recycling this piece of space hardware
seemed cool and fun. And second, be-
cause it might generate useful scientifc
data and we could take people all
over the world along for the ride, he
answered.
For me, the storyline of the ISEE-3
Reboot Project resurrected not only
an aging spacecraf, but the fagging
narrative of American can-do, imagi-
nation, and energy those natural
dynamics of a young country.
But now this nation is middle-aged,
has put on some pounds, been around
the block more than a few times. Ofen
it seems as if all that Yankee ingenuity
has gone to fguring out ways to make
as much cash as soon as possible with
other peoples money.
So, this could be an ISEE-3 moment
for the GPS program, with lessons
that could be learned by all the worlds
GNSS operators.
Any one who has had a bathroom
or kitchen remodel project has come
face to face with the unhappy reality
that many things arent built as good
as they used to be or even available at
any cost. Skill sets and supply lines
have disappeared. Components are
unreliable and fail quickly.
In many ways, legacy GPS space
vehicles have demonstrated the du-
rability of ISEE-3 the latter voyage
begun, by the way, the same year as
the frst GPS satellite was launched.
Te 2nd Space Operations Squadron
recently removed SVN-34 from its pri-
mary orbital slot. Launched in October
1993, the satellite far exceeded its 7.5-
year design life and remains capable of
broadcasting healthy signals.
Te far-reaching changes in stake-
holder leadership described by Dee
Ann Divis in this months Washington
View column gives us the chance to
think outside as well as inside the box.
It provides an opportunity to build not
merely personal resumes but to rebuild
and build out an enterprise that has
in its way transformed modern life as
thoroughly as the Internet or mobile
communications.
GPS is the best non-military deal
this country has gotten since the Na-
tional Defense Highway System. But
just as the crisis in the Highway Trust
Fund has shown, GPS could sufer the
same fate of underinvestment of
ideas as well as money.
Despite our recent diminishing
expectations for the mean mission
duration of products like dishwash-
ers and refrigerators, we need to treat
GNSSs as long-term capital assets with
a leadership perspective as strategic as
the infrastructure and national role
that it represents.
Tat may require us to look beyond
American shores for solutions.
We need to recognize that the
decisions we are making now are not
for ourselves, or at least for ourselves
alone, but for our children and our
grandchildren. So, our mindset should
not be that of quarterly profts and live
for today (sha-na-na) but for a future
world that will have ever-greater needs
for afordable, available, and accurate
positioning, navigation, and time.
GLEN GIBBONS, JR.
Editor
Te Next Big Mac
THINKING ALOUD
Te ISEE-3 Reboot Project is probably the best use of an
abandoned McDonalds hamburger stand that Ive heard of.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
In co-operation with
Made in Germany
www.ifen.com
Winning GNSS Test
Solutions Delivered to You!
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
___________ ________
A
s the Air Force contemplates chang-
ing its GPS III prime contractor in
the face of program delays, sources say it
is also weighing much broader program
changes to help it weather years of con-
tinued lean budgets.
Air Force Space Command (AFSPC)
on June 4 released a request for sources
able to produce up to 22 GPS III space-
craf an efort to assess the industrys
capability to replace both Lockheed
Martin, the current prime contractor,
and Exelis, the subcontractor respon-
sible for the navigation payload. That
payload and con sequently the GPS
III satellites themselves has been sig-
nifcantly delayed, much to the ire of the
Air Force.
The original $1.46-billion contract
let in 2008 anticipated the first GPS
III launch would occur this year. And,
indeed, for the frst few years the Lock-
heed team was running ahead of sched-
ule. However, that launch schedule has
now slipped to late 2015. Exelis (formerly
ITT Space) had a role in building all of
the GPS navigation payloads but could
not produce the GPS III version with the
new civil GPS L1C signal and multiple
legacy signals within the original time-
line.
Numerous U.S. aerospace companies
responded to the AFSPC solicitation, as
well as at least one foreign frm.
Obviously we want a GPS III that
does what its supposed to do, delivered
on time, Lt. Gen. Ellen Pawlikowski
said at a press briefng during this years
National Space Symposium, Defense
News reported. Until recently, Paw-
likowski headed the AFSPCs Space and
Missile Systems Center (SMC) at Los
Angeles Air Force Base home of the GPS
Directorate (GP).
SMC/GP intends to create a competi-
tive two-phase process for producing
up to 22 GPS III satellites afer the frst
eight for which Lockheed has already
been funded. In Phase 1 one or two
Production Readiness Firm Fixed Price
(FFP) contracts will be awarded in Fis-
cal Year 2015 (FY15). Te contractor(s)
would conduct space vehicle (SV) and
navigation payload critical design review
(CDR) with demonstrations and qualif-
cation of the SV subsystem boxes.
Phase 2 will be a limited competi-
tion between Lockheed Martin and the
selected contractor (or contractors) for
up to 22 GPS III production SVs. An
award for those satellites is anticipated
in the FY17/18 timeframe to support a
frst SV available for launch no later than
the frst quarter of FY23.
Aerospace and defense experts have
diverse opinions about the AFs inten-
tions for the new solicitation. Some see
it as merely a shot across Lockheeds bow
to indicate ofcial dissatisfaction with
the delays, some as a serious opportu-
nity for new entrants into the program,
others as cover for a reworked Lockheed
design for the GPS IIIs.
Chance of SV Redesign?
Al t hough t he sources sought
announcement indicates that military
ofcials are looking for a production con-
tractor and not one to tackle a develop-
ment program, experts tell Inside GNSS
that broader changes are being weighed.
Everything is in fux, said one well-
informed source, who compared the
pace and search for solutions to the sort
seen during wartime. Te GPS III solic-
itation had a deadline for responses of
less than two weeks afer publication of
the announcement.
Te expert, like the others who dis-
cussed the contract situation, spoke on
condition anonymity in order to be able
to discuss the program freely.
Another source said the Air Force
might be considering launching at least
some of the GPS satellites without the
nuclear detection payload an idea that
has been foated before.
There is a very strong possibility
that a new GPS III without the (Nuclear
Detonation Detection System [NDS]) on
360 DEGREES
it could, in fact, be built, said the expert.
What makes it possible, this expert said, is
that enough satellites with the NDS pay-
load would be on orbit afer the 12th GPS
III spacecraft is launched to fulfill the
nuclear detection mission through 2040
making it unnecessary to lof more.
Its an avenue thats being explored,
said the expert. Tey can fy some cheap-
er GPS IIIs without NDS for a while, and
if at that point they need to put it back,
they can.
Pulling NDS from some of the GPS
III satellites could allow dual launch
of those spacecraft at a significant
cost savings. Although the Air Force
announced earlier this year that it
would not pursue dual launch, the GPS
Program Ofce is keeping the capabil-
Find out more at
GNSS Hotspots
online!
Portland, Oregon USA
SMART BALL Adidas has
designed every ofcial World
Cup ball since 1970. And thats
not all! The Adidas Innovation
Team in Portland, Oregon,
spent 4 years on a smart soccer
ball with a six-axis MEMS
accelerometer sensor package
that can detect speed, spin,
strike and ight path data and
whip it on over to their special
GPS app on your iPhone. The
app interprets the data for you,
coaches you, and keeps a video
to show your friends. On sale
now for only $299.
Images and Credits
1. Rick Dikeman image of a soccer player, Wikimedia commons
2. NASA astronaut Steve Swanson captured this view of Cape
Canaveral, Florida from the International Space Station on April
14, 2014.
3. Harrisons chronometer H5 (Collection of the Worshipful
Company of Clockmakers)
4. Screen shot of Google maps comparison of different routes
in the Yahoo Labs experiment.
5. Apple image of frequent location screen on iPHone.
GPS III Info Request Draws
Competitors, Redesign Ideas
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
ity in its plan for the spacecraf so that
the option is available later if needed,
according to this source.
Search for Payload
Supplier Alternatives
One of the conditions under the potential
new contract is to preclude Exelis from
being the navigation payload provider.
Other U.S. frms might be able to pro-
vide that capability, but sources suggested
that the Air Force may be contemplating
looking even farther afeld.
I know theyre looking at whether or
not they can buy that package from Galileo
and modify it enough for GPS, said one
source. Teyve got a guy actually spend-
ing time and efort to do that analysis.
Te American arm of Europes Sur-
rey Satellite Technology Ltd. (SSTL),
manufacturer of the Galileo navigation
payloads, is certainly willing to discuss
the opportunity.
We are very interested in the alter-
nate-source GPS III activities and [are]
looking to undertake studies and add
value to whatever team we might be able,
said Douglas Gueller, the chief operating
ofcer for SSTLs U.S. subsidiary. His frm
has manufacturing facilities in Denver, he
said, and is now executing it frst missions
in the United States, which include pay-
loads for the Air Force and JPL.
One U.S. company likely to have the
necessary expertise is Ball Aerospace
which would also be happy to step in.
[We seek] to assist the Air Force
with afordability for GPS by leveraging
360 DEGREES
GNSS Hotspots
Barcelona, Spain and
Torino, Italy
HAPPY ROUTES Yahoo Labs
in Spain and researchers in
Turin came up with a mapping
algorithm to nd the most
emotionally pleasant ways
to get around. They used 3.7
million locations in London from
Geograph and Street View, then
crowdsourced the response to
each location using UrbanGems.
org. They plotted the most
beautiful, quiet or happy routes
and compared them to the
shortest. Guess what? It takes
only 12% more time, on average,
to walk in beauty.
Greenwich, United Kingdom
LONGITUDE PRIZE II
The Longitude Prize of 1714
was won in bits and pieces by
John Harrison for his marine
chronometer. Some 300 years
later, the British Government
decided to encourage solutions
to 21st century problems using
a renewed version of that
drawn-out competition. This
one is worth 10 million (US$17
million). The royal museums at
Greenwich joined in with a new
exhibit: Ships, Clocks & Stars,
running through January 4,
2015, and featuring the original
Longitude Act and Harrisons
ve clocks.
Cape Canaveral and Guiana
Space Center
ROOM FOR MORE In less than
a decade, well have 100+ GNSS
satellites in orbit. The U.S. and
the E.U. are doing their part
with launches this summer.
The seventh GPSII-F will go up
from Cape Canaveral on July 31/
August 1 and two Galileo FOC
satellites will head for the skies
from Kourou around August 22.
GLONASS
BeiDou
Military
Breaking
Policy
Signal Launch
Other Systems Galileo
GPS
Bright Idea
Technology
Commercial
Consumer
Satellite
History
Glitch
Conference
Beijing, China and
Cupertino, California
NATIONAL SECURITY
Chinas state broadcaster CCTV
called the iPhone OS7 a national
security concern on July 11. They
interviewed Ma Ding of Peoples
Public Security University in
Beijing, who said the frequent
location function could
provide Apple with condential
information about the economy
or even state secrets. The
expensive iPhone is popular
among high-level bureaucrats
and business leaders. Apple said
they have never allowed access
to their servers nor created a
backdoor for any agency of
any government and we never
will.
[our] strong navigation payload capabili-
ties, said Roz Brown, Ball media rela-
tions manager, in response to a query on
the announcement.
Boeing has also confrmed its inter-
est in competing for the potential post-
deal. As the prime contractor for the GPS
Block IIF satellites now being launched,
the company is the likely lead contender,
sources agreed.
[We continue] to believe there are
afordable low-risk alternate GPS solu-
tions, and looks forward to supporting
the Air Force in the Sources Sought pro-
cess to best meet the future warfighter
needs, said Paula Shawa, spokesperson
for the companys Communications,
Space & Intelligence Systems division in
a statement.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
P
h
o
t
o
:

J
o
h
a
n

B
r
y
h
n
i
NORWEGIAN EXTREME ARTIST,
ESKIL RONNINGSBAKKEN, DURING
THE WORLDS HIGHEST ONE-HANDED
HANDSTAND FACING A 1,969 FEET
FREE FALL AT PULPIT ROCK, ABOVE
THE LYSEFJORD, NORWAY.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
INSPIRE US
When size, performance and robustness matter
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Northrop Grumman has also sub-
mitted a response to the Air Force, con-
frmed spokesman Lon Rains.
Space Systems/Loral, now known as
SSL, wants to be counted in, although it
has not yet stepped forward ofcially.
SSL did not submit an RFI response
for GPS 3, but that does not indicate a
lack of interest in ofering the USAF a
low cost commercial bus solution, said
Chuck Cynamon, vice president for
business development at SSL Federal in
an emailed statement.
While we do not build precision
navigation and timing (PNT) payloads,
Cynamon added, we have integrated
PNT capabilities onto commercial sat-
ellites in the past. If the USAF elects to
move forward with an alternative acqui-
sition strategy, we would be interested
in opportunities to ofer a commercially-
based solution.
Lockheed Martin and Exelis are
hardly out of the picture, however, and
have recently had some success with the
navigation payload.
Right now Lockheed Martin is
focused on delivering GPS III satellites
that meet all mission requirements at an
afordable cost, said Lockheed Martin
spokesperson Chip Eschenfelder. As
with any complex development pro-
gram, there are first article technical
challenges; in this case, with GPS III, it
has been with the new advanced naviga-
tion payload.
E s c h e n f e l d e r
added, Test data indi-
cates we have resolved
all known technical
issues. Recently, the
last major payload
subsystem to be test-
ed, the Mission Data
Unit (MDU) which
is the heart of the pay-
l oad compl eted
Acceptance Test and
was added to the pay-
loads panel.
The compl et ed
navigation payload
panel will now under-
go final panel-level
testing prior to deliv-
ery as a completed
navigation payload
this fall.
L o c k he e d ha s
closed its Newtown,
Pennsylvania, facilities
where the GPS III sat-
ellite was developed,
with about 350 of
those employees relocating to company
sites in Sunnyvale, California, and near
Denver, Colorado, where components
will be manufactured and the satellites
will be assembled.
GNSS Contributes
to New Seismic
Maps, Early Quake
Warning System
Updated National Seismic Hazard Maps
released by the U.S. Geological Survey
(USGS) on July 17 indicate a higher level
of earthquake risk for the West Coast
and some areas of the Midwest and East
Coast then previously thought..
An accompanying USGS report,
Documentation for the 2014 Update of
the United States National Seismic Haz-
ard Maps, acknowledged the increas-
ing role of geodetic data in assessing
earthquake risk. Tese data are derived
primarily from more than 1,800 high-
accuracy, continuously operating GPS
reference stations.
While all states have some poten-
tial for earthquakes, 42 of the 50 states
have a reasonable chance of experienc-
ing damaging ground shaking from an
earthquake during the next 50 years.
Te hazard is especially high along the
West Coast, intermountain West, and in
several active regions of the central and
360 DEGREES
2014 USGS National Seismic Hazard Map
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
________________
A miniature IMU
Weight: 0,12 lbs (55g)
Volume: 2,0 cu. in. (35cm
3
)
sales@sensonor.com www.sensonor.com
When size, performance and robustness matter
Available now contact us to discuss your application
Certain missions demand unsurpassed precision,
stability and reliability. Having perfect control
and fully understanding the smallest detail is
what it takes to be a world leader.
With this in mind, we developed the Inertial
Measurement Unit STIM300, a small, utra-high
performance, non-GPS aided IMU:
ITAR free
Small size, low weight and low cost
Insensitive to magnetic elds
Low gyro bias instability (0.5/h)
Low gyro noise (0.15/h)
Excellent accelerometer bias instability (0.05mg)
3 inclinometers for accurate leveling
STIM300 is the smallest and highest performing,
commercially available IMU in its category,
worldwide!
TO BE EVEN BETTER!
eastern United States, such as near New Madrid, Missouri, and
near Charleston, South Carolina.
Te latest hazard assessment provides additional fuel for
an efort to establish a West Coast Earthquake Early Warning
(WC-EEW) system. Te USGS, along with partners among uni-
versities and state and local agencies, has proposed to develop
and operate an EEW system called ShakeAlert, for the high-
est-risk areas of the United States. Te system would leverage
current earthquake monitoring capabilities of the Advanced
National Seismic System along with the GPS networks.
Te early warning system exploits physical characteristics of
earthquakes, which generate two main types of waves: primary
or P-waves and secondary (S) waves. P-waves travel at high
speeds outward from the earthquake epicenter, but rarely cause
damage. S-waves travel more slowly but result in more intense
ground shaking that causes damage.
By detecting and analyzing the location and magnitude of
an earthquake refected in the P-wave energy, expected ground-
shaking levels across a region can be estimated and warnings
sent to local populations before more damaging shaking arrives
with or afer the S-wave. Te advanced warning can range from
seconds up to more than a minute, depending on the distance
an afected area is from the earthquakes origin.
Tat may not sound like much of a head-start on preparing
for a quake, but it can provide a crucial edge for eforts to mini-
mize damage from a quake, says Ken Hudnut, a geophysicist at
the USGS Earthquake Science Center in Pasadena, California,
and chairman of the GNSS Working Group for the WC-EEW.
Benefts envisioned for ShakeAlert include such eforts as get-
ting school children to safety under their desks sooner, operating
automatic shut-of valves for utilities and machinery, putting
computer systems into a safer state, or switching other automat-
ed systems to try to prevent loss of life or damage to property.
Hudnut discusses the role of GNSS in earthquake research
and the WC-EEW in greater detail in a feature beginning on
page 32.
Capital investment costs for a West Coast EEW system are
projected to be $38.3 million, with additional annual main-
tenance and operations totaling $16.1 million, according
to a recent USGS Technical Implementation Plan for the
ShakeAlert Production System.
On July 15, the U.S. House Appropriations Committee
voted to include $5 million in an appropriations bill for the
U.S. Department of the Interior and Environmental Protec-
tion Agency, the frst time Congress has specifcally committed
funds to the Early Earthquake Warning System.
Early Warning, Forecast, and Prediction
WC-EEW leaders emphasize that the system is not an earth-
quake prediction system, but a rapid reporting method for seis-
mic activity already under way. It is also distinct from earth-
quake forecasts, such as are represented by the seismic hazard
maps published by USGS.
People seem to understandably mix up these concepts, says
Hudnut, who ofers the following defnitions; forecast (very
long-term, around 30 to 50 years), prediction (short-term, a
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
________________
matter of days), and early warning (very
short-term, measured in seconds). We
can do forecasting and early warning, but
unfortunately we cannot do earthquake
prediction in a useful way, he says.
Early warning refers to the ability to
detect an earthquake while it is in prog-
ress and report to users instantaneously,
Hudnut says. The damaging seismic
waves travel at about three kilometers
per second, so, a system like the WC-
EEW can outpace them by sending the
messages ahead through a telemetry sys-
tem. WC-EEW must make sure the event
detection is solid, then measure the mag-
nitude accurately. Tese tasks in addition
to some minor processing and telemetry
delays can add up to ten seconds.
Hudnut uses the example of a big San
Andreas earthquake in southern California
with a magnitude 7.8, starting southeast of
Palm Springs and rupturing towards Los
Angeles. Tat city could get up to as much
as 50 seconds of advance warning from the
system, Hudnut says. Tat section of the
fault is just over 180 kilometers away; so,
the damaging S-waves would take 60 sec-
onds to arrive and it takes us about 10 sec-
onds to get the size accurately and deliver
the alert message.
In contrast to emerging EEW sys-
tems operating in realtime, earthquake
prediction is a heavily loaded term,
says Hudnut. People have wanted us to
predict earthquakes for years, and what
they seem to mean is they want several
days of warning prior to a Big One, that
is, one that causes signifcant damage at
their particular location.
To successfully issue a real predic-
tion in this sense, would require the
ability to accurately specify magnitude,
location, and time. Doing that in a use-
ful way is essentially impossible to do
today, and will probably remain so into
the very far future, Hudnut says, add-
ing, I am continually enthused by new
technologies like GNSS coming along,
as these allow us to discover new things
about earthquakes; so, I do not con-
sider prediction totally hopeless. But it
is clearly a very, very long way of and
maybe never attainable.
Growing Role for GNSS
GPS-based geodetic data charts the rate
(velocity) and direction of movement of
the strategically sited reference, refect-
ing crustal plate motion and deforma-
tion or strain of the underlying Earth
layers. USGS experts combine the geo-
detic data with geologic data and seismic
sensors that measure slippage and vibra-
tion along fault lines.
Modern GPS data provide more spa-
tially complete observations on ground
deformation, cover a period of 25 years
(from 19872012), and record the most
recent plate-loading rates with submil-
limeter per year precision, which is more
precise than data provided by geologic
studies, the USGS researchers wrote in
their report.
In long-term forecasting, geodesy is
useful because we assess the rate of load-
ing of active fault zones, says Hudnut.
We measure the velocity vector very
precisely over many years for each of
our continuously operating GNSS sta-
tions, the Plate Boundary Observatory
with over 1,000 GNSS stations and all
of the USGSs over 100 additional GNSS
stations along the San Andreas fault sys-
tem and across the Los Angeles and San
Francisco urban areas.
Tese data are combined with other
data, such as geological fault slip rates
and the USGS seismicity catalog, to
make the long-range earthquake hazard
assessments or forecasts. With better
understanding of potential ground shak-
ing levels, such forecasting can support
various risk analyses by considering
factors like population levels, building
exposure, and building construction
practices.
In turn, these analyses can be used for
establishing building codes and estimating
seismic risk for key structures. Tey can
also help in determining insurance rates,
emergency preparedness plans, and private
property owners evaluating their homes in
order to make them more resilient.
Communications
Act Rewrite Could
Adversely Affect
GPS Community
Two powerful lawmakers are weighing
rewriting the rules for the way frequen-
cies are allocated as part of an overhaul
of the nations telecommunications laws.
Te efort, which is likely to see legis-
lation drafed next year, is considering
options such as flexible licensing and
receiver standards that could directly
afect the GPS community.
Its been quite some time since
theres been any type of update or a num-
ber of hearings on reviewing the Com-
munications Act, said Rep. Fred Upton,
R-Michigan, in a taped announcement
in December 2013.
Were prepared in essence to talk
about a launch of a number of hear-
ings that we will have next year. . . . Our
goal, in fact, will be to use these hearings
throughout the course of the next year
to begin to actually launch an update
beginning in 2015.
Upton, who chairs the House Energy
and Commerce Committee, made the
announcement with Rep. Greg Walden,
R-Oregon, the chairman of the Com-
munications and Technology Subcom-
mittee. Te two committees oversee the
Federal Communications Commission
(FCC) and have primary authority over
spectrum issues in the House.
Since their announcement, the two con-
gressmen have held an FCC oversight hear-
ing, another hearing in January on updat-
ing the Communications Act of 1934, and
issued three white papers for comment in
a lead-up to drafing legislation.
One of the points they underscored
in the policy papers is how much tech-
nology has changed since the communi-
360 DEGREES
See additional news stories at
WWW.INSIDEGNSS.COM/NEWS
Galileo IOV Satellite Failure Mystery Still
Unsolved
GPS OCX Program Being Restructured as Budget
Pressures Mount
Harbinger Sues U.S. Government over
LightSquared
FCC Courts GPS Community in Effort to Solve
Spectrum Crunch
GNSS Monitoring Stations Slide into
U.S.-Russia Rift
Homeland Security Researching GPS
Disruptions, Solutions
. . . and more.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_______________
cations law was written and even since
its most recent major update.
Spectrum Allocation
White Paper
Te issue of spectrum allocation, and the
need for more frequencies for commer-
cial use, seems to be near the top of the
work list deserving of a white paper of
its own. Te paper asks for feedback on a
number of questions including whether
the FCC should be allowed to take the
potential revenue from spectrum auc-
tions into account when making allo-
cation decision, something that would
likely put additional pressure on the
highly desirable L-band, where the GPS
frequencies are located.
It also asks about shifting to flex-
ible use licenses which, the paper said,
would permit licensees to use their
spectrum for any service, including
wireless, broadcast, or satellite services,
as opposed to the current system of des-
ignating what uses can be made of spe-
cifc frequencies.
Te GPS Innovation Alliance noted
in its submitted comments that flex-
ible use is described in another white
paper by the FCC Technical Advisory
Committee (TAC) as permitting uses
up to and including high power mobile
network downlinks.
While such a regime provides a great
measure of freedom to the licensee who
acquires flexible use spectrum, the
Alliance wrote, this flexibility comes
at a cost to any adjacent spectrum hold-
er, who will be expected to be able to
accommodate the full range of permit-
ted operations, up to and including very
high powered operations.
If the adjacent band use is not imme-
diately compatible with high powered
use, the TAC White Paper appears to
suggest that adjacent spectrum holders
will be forced to accommodate the use
over time.
Te TAC assumed in its white paper,
without providing technical evidence,
said the Alliance, that with the right
amount of time and some unknown
level of investment in new technology
or alternative product design, any spec-
trum use should be able to accommo-
date high-powered, cellular-like opera-
tions in directly adjacent spectrum. Tis
assumption has not been thoroughly
tested and validated, and based upon
past instances involving signifcant inter-
ference between dissimilar uses in close
spectral proximity, may be unfounded.
The Alliance suggested a zoning
and re-farming approach, which groups
similar types of services together. Tis
method could reduce the need for spec-
trum by reducing the need for the guard
bands, or buffer zones between users,
they suggested.
Receiver Standards &
Reducing NTIA Role
The lawmakers also asked about the
desirability of setting standards that
TERRASTAR
Global Precise Point Positioning using
GPS and GLONASS at better than
6cm horizontal (95%)
From the leading
independent supplier
of Differential GNSS
services comes TERRASTAR.
Available, accurate and reliable
24hr service for all civil and
military applications on land,
near shore, and in the air.
Where you want it, when you
want it, backed by years
of experience and support.
Terrastar GNSS Ltd
Unit A03, Ground Floor
The Nottingham Geospatial Building
University of Nottingham
Triumph Road, Nottingham
NG7 2TU United Kingdom
info@terrastar.net www.terrastar.net
Continued on page 80.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
____________ ____________
W
ashington, D.C., has
a peculiarity of sea-
sons. While most of the
world marks the shifts
between winter and spring, summer and
autumn, the politicos on the streets of the
U.S. capital count the passage of time in
two-year increments.
New operatives and appointees fock
to the centers of power in the early days
of each administration and the open-
ing of each Congress, then migrate to
friendlier climes as congressional elec-
tions loom and the administration winds
down as it is now.
Te GPS community, in particular,
seems to be caught up in this seasons
migration. New personnel are stepping
into key leadership posts in the depart-
ments of defense and transportation as
well as other agencies that
work together to guide the
GPS program.
Te National Coordina-
tion Ofce (NCO) for Space-
Based Positioning, Naviga-
tion, and Timing (PNT),
which supports all of these
efforts, has been operat-
ing for months without a
top executive. Changes are
occurring in the congressio-
nal committees that autho-
rize the GPS budget and
perhaps in those that can
infuence the fortunes of the constella-
tions operations and utility.
Tese personnel shifs are occurring
just as a new efort to rewrite the nations
telecommunications laws is emerging,
potentially threatening GPS frequencies
and applications.
The most notable shifts are within
the Department of Defense (DoD).
General William Shelton, who became
Commander of Air Force Space Com-
mand (AFSPC) in January 2011, will
retire September 1. By all accounts his
leadership of those responsible for the
modernization and support of the GPS
system has been admirable. Although no
job at this level is ever easy, his tenure has
weathered one squall afer another with
years of unusually serious budgetary
challenges, glitches in eforts to enhance
the system, and the rise of new threats
to, and doubts about, the availability of
GPS signals.
When he assumed the helm in 2011,
Shelton stepped immediately into the
middle of what would become the big-
gest threat so far to GPS frequencies.
Tat same month a Virginia frm called
LightSquared, amid great controversy,
won conditional permission from the
Federal Communications Commission
(FCC) to build a coast-to-coast, high-
powered terrestrial broadband network
using frequencies neighboring those used
by GPS. Te frms project ft neatly into
the Obama administrations plan to pro-
vide more spectrum for broadband com-
panies and encourage competition within
the wireless communications industry.
RF power f rom
LightSquareds trans-
mitters and mobile
devi ces, however,
threatened to over-
whelm GPS receivers
across the country.
Shelton took a stand
against LightSquareds
aspirations and made
nat ional news for
ref usi ng to submit
to pressure from the
White House to sofen
congressional testi-
mony about the clear interference prob-
lems that testing had shown. Although
the issue is not fully resolved, the Light-
Squared project is currently sidelined.
Patent, Jamming, and
Budget Problems
Other challenges followed, including an
efort by the British Ministry of Defense
New Leaders at the GPS Helm
CHANGES ACROSS THE BOARD AS NEW THREAT
TO SPECTRUM EMERGES
DEE ANN DIVIS
Dee Ann Divis has covered
GNSS and the aerospace
industry since the early
1990s, writing for Janes
International Defense
Review, the Los Angeles Times, AeroSpace Daily and
other publications. She was the science and technology
editor at United Press International for ve years,
leaving for a year to attend the Massachusetts Institute
of Technology as a Knight Science Journalism Fellow.
WASHINGTON VIEW
U
S
A
F

p
h
o
t
o
s
.
Gen. William Shelton,
AFSPC Commander
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Anti-jam antenna
GNSS + INS
High precision
GNSS
GNSS jamming
trainer
SAASM GPS
NAVWAR technology
Battle Tested. Field Proven.
ENSURE NAVIGATION SUPERIORITY
FOR MISSION SUCCESS
www.novatel.com/defense
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
to patent one of the modernized GPS
signal structures a patent that was
withdrawn afer what experts described
as government-to-government discus-
sions. Other concerns arose about the
aging GPS constellation, repeated dem-
onstrations that the GPS signal could
be spoofed and jammed, and surging
threats from hackers.
Shelton also found himself caught up
in budget battles consuming Washington
as well as signifcant problems with the
navigation payload on the new GPS III
satellites and the new ground system. To
tackle both the technical and fnancial
problems, his team weighed a variety of
alternatives including reconfguring the
GPS constellation to take advantage of
very small satellites and investing in the
ability to launch more than one satellite
at a time. Among the outcomes of that
work is the recent restructuring of the
ground system contract and the release
in June of a request that could lead to a
new prime contractor for the last 22 GPS
III spacecraf.
Shelton seized some opportunities as
well. He is credited with turning on the
navigation message in the L2C and L5
signals, a move expected to spur receiver
development and innovation as the com-
mercial sector gains experience with the
new signals.
New AF Space Commander
Shelton will be replaced by Lt. Gen. John
Hyten, who has served as his AFSPC
vice-commander for the last two years.
Hyten has a Harvard engineering degree
and a masters in business administration
and has served in Washington as direc-
tor for space programs in the Ofce of
the Assistant Secretary of the Air Force
for Acquisition experience that should
serve him well has as sequestration
reemerges and squeezes the GPS budget
tighter still.
Hytens experience as the direc-
tor of space forces during Operations
Enduring Freedom and Iraqi Freedom,
and in senior engineering positions on
anti-satellite weapon system programs,
also could prove very valuable as the
challenges from adversaries mount. He
also has worked with the GPS program
before, noted one source, and has direct
experience with the interagency process
that guides GPS policy.
Hes defnitely one of the GPS people,
said the well-informed source, who asked
not to be identifed due to a lack of autho-
rization to speak publicly. Were glad to
have him over in Space Command.
In yet another change for Space Com-
mand, Lt. Gen. Ellen Pawlikowski, the
head of the Space and Missile Systems
Center at Los Angeles Air Force Base,
has lef to become military deputy with-
in the Ofce of the Assistant Secretary of
the Air Force for Acquisition replacing
Lt. Gen. Charles Davis as the services top
military acquisition ofcial. SMC, which
is part of Space Command, is home to the
Global Positioning Systems Directorate.
Taking over at SMC is Lt. Gen. Sam-
uel Greaves, who was serving as deputy
director in the Missile Defense Agencys
Ofce of the Undersecretary of Defense
for Acquisition, Technology and Logistics.
Other PNT Leadership
Changes
The handoffs within Space Command
are not the only changes in the GPS
management team at the Department of
Defense.
Robert Work has replaced Ash-
ton Carter as the deputy secretary of
defense and takes his place as co-chair
of the National Executive Committee
for Space-Based Positioning, Navigation,
and Timing (PNT ExCom). Work, who
served 27 years in the U.S. Marines, was
undersecretary of the Navy from 2009 to
2013. He has a master of science degree
in space system operations and served
on the DoD transition team for the in-
coming Obama administration. He was
confrmed as deputy secretary in April.
His co-chair on the PNT ExCom
is also relatively new to the job. Vic-
tor Mendez is acting deputy secretary
of transportation. He stepped into the
roll in late December 2013 to replace
the departing John Porcari. Mendez,
who has an MBA and a degree in civil
engineering, was sworn in as the federal
highway administrator in July 2009. He
also served on the White House transi-
tion team.
The PNT ExCom is supported by
the Executive Steering Group, which
meets more ofen than the ExCom, and
deals with interagency matters that can
be handled without elevating them to
the deputy secretary level. Te Steering
Group has representatives from the same
departments as the ExCom plus agencies
within those departments such as the
Air Force and the Federation Aviation
Administration.
Te Pentagons chief information of-
cer (CIO) and the Department of Trans-
portation (DoT) assistant secretary for
research and technology co-chair the
Steering Group. Greg Winfree has held
the latter position for a while, but the
DoD CIO, whose ofce is the nexus for
U
S
A
F

p
h
o
t
o
s
.
Lt. Gen. John Hyten, incoming chief of Air
Force Space Command
Although no job at this level is ever easy, Sheltons tenure at
AFSPC has weathered one squall after another with
years of unusually serious budgetary challenges, glitches in
eforts to enhance the system, and the rise of new threats
to, and doubts about, the availability of GPS signals.
WASHINGTON VIEW
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
innovation standard
For more information, visit aerospace.honeywell.com\innovationstandard
2014 Honeywell International Inc. All Rights Reserved
For 100 years, Honeywell has been synonymous with
aviation innovation that improves performance while
saving time, fuel, money and lives. With products that pay
for themselves. And technologies that create the future.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
____________________________
DoD decisions on GPS policy and pro-
curement, is new. Teri Takai left that
job at the beginning of May with only a
weeks notice to the surprise of many in
the GPS community
I think she did a lot to establish the
infuence of the CIO ofce, said Scott
Burgett, director of automotive OEM
platform engineering for GPS user equip-
ment manufacturer Garmin Internation-
al, who praised her as a very efective
advocate for GPS.
Takai has been replaced, for now, by
Terry Halvorsen, who is the Pentagons
acting CIO. His role, however, may not
be permanent. According to Defense
Daily, the DoD has issued a formal state-
ment from spokeswoman Lt. Col. Valerie
Henderson stating, Mr. Halvorsen will
lead the DoD CIO organization until
a permanent DoD CIO is selected by
[Defense] Secretary [Chuck] Hagel.
Halvorsens situation underscores the
challenges facing anyone in a job in an
acting capacity. When so much depends
on the skills, expertise, perspectives, and
advocacy of individuals, having people
who may not be staying or are per-
ceived as temporary makes planning
and progress much harder.
ExCom Newbies
Te departments of state, commerce and
homeland security all have representa-
tives to both the ExCom and the steering
group who are either new to their jobs or
still in an acting capacity.
Bruce Andrews was just named acting
deputy secretary of commerce in June.
His experience in telecommunications
could be especially useful as the battle
over spectrum heats up. He was general
counsel for the Senate Committee on
Commerce, Science, and Transportation,
the committee that oversees the FCC.
Prior to that he was a telecommunica-
tions attorney and managed government
afairs for Ford Motor Company.
Commerce is represented in the PNT
Executive Steering Group by former
astronaut Kathryn Sullivan, who was
fnally confrmed in March as the under
secretary of commerce for oceans and
atmosphere and as National Oceanic and
Atmospheric Administration (NOAA)
administrator. She is an oceanographer
who, when she served as NOAAs chief
scientist, oversaw a research and tech-
nology portfolio that included fsheries
biology, climate change, satellite instru-
mentation, and marine biodiversity.
The GPS commercial sector may
particularly appreciate the background
of Judith Garber, the State Depart-
ments ExCom representative and the
acting assistant secretary of state for
the Bureau of Oceans and International
Environmental and Scientific Affairs
(OES). She is a career foreign service
ofcer who has held economic and busi-
ness development posts around the world
and was named to her new post at the
end of April.
The State Departments person in
the steering group is Jonathan Margolis,
acting deputy assistant secretary at OES
for the science, space, and health. He
has a Ph.D. in psychology from Harvard
University, focusing on negotiation and
confict resolution, and a masters degree
from the Fletcher School of International
Law and Diplomacy as well as hands-
on experience organizing international
communications. Given that the current
round of international negotiations over
spectrum is coming to a head next year,
his skills could be very valuable.
Just joining the PNT ExCom is Deputy
Secretary of Homeland Security (DHS)
Alejandro Mayorkas. A former U.S. attor-
ney, he had been serving as the director
of DHSs United States Citizenship and
Immigration Services (USCIS), which
operates the largest immigration system in
the world. He was sworn in last December.
Confrmed in March, as under secre-
tary for DHSs National Protection and
Programs Directorate, Suzanne Spauld-
ing brings a wide range of experience in
intelligence and cybersecurity matters.
She served at the Central Intelligence
Agency for six years and on the staf of
both the House and Senate intelligence
committees.
Spauldings role at DHS gives her the
responsibility for protecting critical infra-
structure. GPS already is considered an
essential element in most of the nations
critical infrastructure sectors, and some
top experts have argued GPS should itself
be designated critical infrastructure.
New Leadership at NCO
Te leadership at the National Coordi-
nation Office, the permanent staff for
PNT ExCom, is also in f lux. Former
NCO director Jan Brecht-Clark retired
in December after serving just a year.
Te NCOs deputy director, Col. Harold
Stormy Martin, retired from military
service several months later.
Now that Martin has lef the military
he could potentially return as the NCOs
director the path taken by Anthony
Russo, who was NCO director from
January 2010 through December 2012.
Te DoT selects the NCO director and is
expected to name its choice in early July,
although nothing had been announced
as of press time.
Meanwhile, on Capitol Hill
Key personnel changes are also under
way in Congress with some members
retiring, some moving to diferent roles
because of term limits, and others forced
out by election losses.
Among those retiring will be the
chairmen of the both the House and Sen-
ate armed services committees, which
oversee the GPS program. Overall, these
committees and their chairmen have
been very supportive of GPS, particularly
in their budget authorizations.
House Armed Services Chairman
Howard Buck McKeon, R-California,
When so much depends on the skills, expertise,
perspectives, and advocacy of individuals, having people
who may not be staying or are perceived as
temporary makes planning and progress much harder.
WASHINGTON VIEW
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
GNSS receivers
Everywhere it matters, we deliver
Millions of critical decisions are made every day in the complex GNSS
operational environment. Thales is at the heart of this. Embedded on
many military platforms worldwide, our powerful GNSS 1000S suite
is best-in-class with full SAASM functionality, meets current and future
navigation requirements, and is suitable for line-t and
retrot programs. We give decision makers the
information and control they need to make more
eective responses. Everywhere, together with our
customers, we are making a dierence.
NAVWAR OPERATION
Enable ultra-fast Direct-Y code
acquisition and CPRA interface
EASY UPGRADE
Provide a single GRAM-S standard
receiver for all air and naval missions
CIVIL AIRSPACE INTEGRATION
Supply full product roadmap
including MSO certication
SAASM CAPABILITY
GPS Precise Positioning Services capacity
approved by US GPS Wing Directorate
GALILEO COMPATIBILITY
Update applications with Dual
Mode GPS M-Code and Galileo PRS
t ha l e s gr oup. c om
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
who is retiring from Congress, told
reporters that term limits, which man-
date his giving up his chairmanship,
were the biggest motivator for his
decision. Rep. Mac Tornberry of Texas,
who is described in news reports as being
extremely knowledgeable on defense
matters and who has already made his
interest in the chairmanship clear, is the
most likely replacement.
Sen. Carl Levin, D-Michigan, chair
of the Senate Armed Services Commit-
tee, is also retiring. He has endorsed Sen.
Jack Reed, D-Rhode Island, as his suc-
cessor, according to the web site Daily
Kos, but Reed is also in line to take the
soon-to-be-vacant chairmanship of the
Senate Banking, Housing and Urban
Affairs Committee. Senators Bob Nel-
son, D-Florida and Claire McCaskill,
D-Missouri, would be next in line by
seniority although Nelson has a number
of other options.
Te leadership situation in two other
Senate committees is particularly worth
watching. Sen. John D. Rockefeller,
D-West Virginia, who now heads the
Commerce and Transportation Com-
mittee, is retiring afer fve terms. Mark
Pryor, D-Arkansas, the chairman of the
Communications, Technology and Inter-
net Subcommittee, is in a very close race
this fall. Tese two committees have the
lead on spectrum issues in the Senate and
played a role in the LightSquared debate
a few years ago.
Te chairmanships are particularly
important in this case because the heads
of the corresponding House committees
have launched an effort to update the
Communications Act of 1934 and have
placed issues that could greatly impact
GPS squarely on the table.
House Energy and Commerce
Committee Chairman Fred Upton,
R-Michigan, and Communications and
Technology Subcommittee Chairman
Greg Walden, R-Oregon, announced
their multi-year review in December.
Tey have asked for feedback on issues
including receiver standards and the idea
of altering the role the National Telecom-
munications and Information Adminis-
tration (NTIA).
NTIA watches out for the federal gov-
ernment users of frequencies includ-
ing those who rely on GPS and played
a key role in protecting the GPS spec-
trum during the LightSquared contro-
versy. As things now stand, the FCC and
NTIA have to agree on frequency allo-
cations, an arrangement deemed dupli-
cative by some who would like the FCC
to have most if not all of the decision-
making power. For more on this story,
see news article on page 20.
Upton and Walden appear well posi-
tioned and, given that term limits will
force Upton to relinquish his chairman-
ship by 2017, well motivated to launch
legislation next session. Whether they
succeed or not depends in part on who
chairs the Senate committees. Sen. Bar-
bara Boxer, D-California, is next in line
for the chairmanship, but she already
leads other committees and the Demo-
crats have not imposed term limits on
their members. Democrats Bill Nelson of
Florida and Maria Cantwell of Washing-
ton state would seem to be likely choices
based on seniority.
In any case, its too soon to know,
particularly since most political experts
give the Republicans better than even
odds of taking control of the Senate in
this falls elections. If that happens, then
Republicans will control the chairman-
ships and set the agenda. If they can
come to agreement amongst themselves,
they will be in a much stronger position
to push changes through.
Unfortunately, the GPS community
already has lost some of the members
who acted to protect GPS frequencies
during the LightSquared fracas.
Of the six members that organized
Dear Colleague letters opposing Light-
Squareds request in the spring of 2011,
half are gone or on their way out. Sen.
Ben Nelson, D-Nebraska, and Rep. Steve
Austria, R-Ohio, both declined to seek
re-election in 2012. Rep. Ralph Hall,
R-Texas, the oldest-serving member of
Congress, lost his primary bid to a Tea
Party challenger this spring. A fourth
GPS advocate, Rep. Collin Peterson,
D-Minnesota, is more likely than not to
win, according to Larry Sabato, an expert
on electoral politics at the University of
Virginias Center for Politics, but he is in
a competitive race.
All in all, the GPS community is
facing a substantial new challenge over
spectrum with a team that is largely new
to GPS issues. It also has fewer proven
friends on Capitol Hill to speak on its
behalf. The good news is that the new
contingent of GPS leaders has an array
of particularly useful skills and time to
plan ahead for the next fght. Whether
they will be ready or not remains to be
seen.
Rep.Greg Walden, R-Oregon, at Communications Act update event.
U
.
S
.

H
o
u
s
e

C
o
m
m
i
t
t
e
e

o
n

E
n
e
r
g
y

a
n
d

C
o
m
m
e
r
c
e

p
h
o
t
o
WASHINGTON VIEW
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Learn more: www.locatacorp.com mining.leica-geosystems.com
* Leica Geosystems Literature: http://bit.ly/wM5x8D 1. International Mining Magazine: http://bit.ly/1boyukv
THE WORLDS FIRST
GNSS+LOCATA RECEIVER
THE NEW LEICA JPS GNSS+LOCATA RECEIVER PRECISE POSITIONING WHERE GNSS FAILS
ITS A GAME CHANGER. ITS SHIPPING.
*
The worlds rst next-generation wireless positioning receiver is here.
While everyone else
was standing around
talking about GNSS+
one of Locatas
integration partners
was busy making it!
Delivering the future.
Leica Mining and their Swiss
colleagues are rst to deliver what
is clearly becoming an inevitable
future for wireless positioning
the powerful combination of
satellite+terrestrial PNT
signals giving one seamless
GNSS+ solution. Its a world-rst
achievement for Leica.
Unprecedented perfomance.
Leicas new GPS+Glonass+Locata
receiver treats terrestrial Locata
signals in the Kalman Filter in
the same way as signals from
GNSS satellites. Its elegant and
revolutionary... and provides
spectacular performance to Leicas
open-cut mining eet customers
where GNSS fails
1
.
This is only the beginning.
As our partners work to miniaturize
Locata technology into next-gen
GNSS+ chipsets they will create
more amazing new products for
many markets, including indoor
and mobile. In the process they will
deliver new capabilities and a level
of control that satellite-based
systems will NEVER achieve.
Locata is a game changer. GNSS can never be the same again.
Locata. YOUR OWN GPS
TM
PRODUCT
OF THE YE AR
LEI CA JPS
POWERED BY
LOCATA
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
___________________________
N
ot surprisingly, GNSS posi-
tioning accuracy is largely
dependent on the level of
measurement errors induced by orbital
inaccuracies, atmospheric efects, mul-
tipath, and noise. Tis article discusses
how, specifcally, these errors manifest
as position errors.
Estimating a Position
For the purpose of our discussion here,
we only consider least-squares estima-
tion with no a priori knowledge of
the receivers position or time. To this
end, least-squares assumes the mea-
surements, , are related to the states
(position and clock bias, in the case of
GNSS), , as follows
where h(-) is the measuiement model
assumed, for convenience, to be
non-linear and is the vector of
measurement errors. Tis equation is
linearized to yield
where is the current estimate of
the state vector (point of expansion),
is the error of relative to the
(unknown) true states, H is the Jaco-
bian matrix (also called the design
matrix, observation matrix, or geom-
etry matrix), and is the misclosure
vector, which is the diference between
the true measurements ( ) and the
measurements estimated from the cur-
rent states (i.e., ).
Te well-known solution to equa-
tion (2) is as follows:
where R is the covariance matrix of the
measurement errors. Te initial state
estimates are then updated as follows
Because the model is non-linear,
we can use iteration to converge to the
fnal solution.
Role of GNSS Errors
For the purpose of this article, the
pseudorange measurement equation
(equivalent to equation [1]) is written as
where is the vector of pseudoranges
from all satellites in view, is the vec-
tor of geometric distances between
the receiver and the satellites, b is the
receiver clock error (common across
measurements), and is the aggregate
measurement error from all error
sources. Although we aggregate all of
measurement errors together, indi-
vidual components (e.g., troposphere)
could be separated and easily worked
through the following development.
Let us now consider the specifc case
where the initial state estimate was
perfect such that . Although this
is an unrealistic scenario (if you knew
the true position in advance, you do
not need GNSS!), it serves as a useful
illustration of how measurement errors
afect the fnal solution. Furthermore,
since the least-squares approach will
yield the same position estimate for all
reasonable initial state estimates (in
this case, reasonable would include a
position accurate to at least 1,000 kilo-
meters), this scenario is not limiting.
For the assumed case, the true value
of is zero. It follows that if the value
estimated from equation (3) difers
from zero, this actually represents the
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 below.
GNSS SOLUTIONS
MARK PETOVELLO
is a Professor in
the Department of
Geomatics Engineering
at the University of
Calgary. He has been
actively involved in many
aspects of positioning and navigation
since 1997 including GNSS algorithm
development, inertial navigation, sensor
integration, and software development.
Email: mark.petovello@ucalgary.ca
How do measurement errors propagate
into GNSS position estimates?
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_______________
error in the estimated states. To obtain
a more explicit equation, we frst com-
pute the misclosure vector as follows
In other words, the misclosure vec-
tor contains the measurement errors
only. Finally, substituting this result
into equation (3) gives
Equation (7) shows how measure-
ment errors propagate into the fnal
solution. Although the equation is rela-
tively simple, there is no hard-and-fast
rule for describing how this happens.
Rather, we can only say that two key
things determine the efect of mea-
surement error on the fnal solution:
the relative measurement accuracy
refected in R, and the measurement
geometry as refected in the Jacobian.
Before looking at these aspects in
more detail, note that equation (7)
shows the efect on all state estimates
separately (i.e., as a vector). Tis is
important because some applications
may be more interested in certain
parameters than in others. For exam-
ple, aviation is more sensitive to verti-
cal positioning errors than horizontal
positioning errors. In contrast, timing
applications are not concerned at all
with the position states.
Measurement Accuracy
Intuitively, the more accurate a mea-
surement is assumed to be, the more
weight will be given to that measure-
ment. As R is the covariance matrix of
the measurement errors, this weighting
of the measurements happens auto-
matically within the least-squares
estimation process.
Of course, because the user (or
perhaps sofware programmer) is
responsible for selecting the covariance
model, careful decisions need to be
made in this regard; otherwise results
will be suboptimal.
Measurement Geometry
To further explain the idea of measure-
ment geometry, a single row of the
Jacobian matrix (corresponding to the
i-th single measurement) can be writ-
ten as
where is the unit vector pointing
from the receiver to the i-th satellite.
Te distribution of all satellites relative
to the user refects the measurement
geometry. Tis is ofen quantifed using
dilution of precision (DOP) values.
To illustrate the importance of mea-
surement geometry, consider Figure 1,
which shows two measurement sce-
narios for a two-dimensional position-
ing problem. In both cases, the receiver
(blue) is measuring ranges (not pseu-
doranges) from the transmitters (red).
Each transmitter is assumed to have
an error of one meter, and all measure-
ments are given equal weight (i.e., same
variance).
Te distribution of transmitters
appears to be relatively similar; only
one transmitter is moved (mirrored
across the y-axis). Nevertheless, this
small diference in measurement
geometry results in diferent position
errors.
Similar examples can be developed
for the three-dimensional case, but
this is more complicated to draw and is
omitted here.
Unfortunately, users cannot place
satellites to optimize measurement
geometry. Te best that can be done is
to use mission-planning utilities to col-
lect data during parts of the day where
geometry is best (in the area of the data
collection). Of course, using receiv-
ers that track satellites from multiple
GNSSs will inherently improve the
geometry too.
Estimating Clock Errors
Te examples in the previous section
only considered the case of measured
ranges, meaning the clock error state
does not need to be estimated. Howev-
er, estimating the clock error which
is common across all measurements
can signifcantly afect results.
In particular, although we name the
state the clock error, the estimated
value will include the true clock error
along with anything that appears to be
common across all satellites.
With this in mind, if we repeated
the previous examples using pseu-
doranges (thus requiring the clock
error to be estimated), the fact that all
measurements were assumed to have a
one-meter error means that the least-
squares estimator could not separate
the true clock bias from the common
error. Te result would be that the
clock error estimate would be biased by
one meter (in this case), but the posi-
tion error would actually be zero!
Different Types of Errors
Although equation (7) completely
defnes the propagation of a specifc set
of errors (i.e., at a particular instant of
time) from the measurement domain
to the position (and time) domain, this
equation is usually reserved for system-
atic errors that manifest as biases in the
FIGURE 1 Example of the role of measurement geometry. These two examples assume ranges are
measured to each transmitter and each has an error of 1 meter. Despite the similar geometry, the
resulting position errors are quite different.
x = 0.85 m
y = 0.85 m
x = 0.64 m
y = 1.35 m
x
y
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
F
or at least two decades, GPS experts,
geodesists, and public agencies have
been working together to develop
high-accuracy, large-scale continu-
ously operating GPS reference stations that
provide them the capability to monitor and
model crustal deformation, tectonic plate
movement, and the effects of geohazards
such as earthquakes and volcanic eruptions.
Now, GNSS-augmented advance warning
systems are going into place that can give us
a crucial margin of safety in the event of an
earthquake.
And none too soon.
The latest Updated National Seismic Haz-
ard Maps recently released by the U.S. Geo-
logical Survey (USGS) indicate a higher level
of earthquake risk for the West Coast and
some areas of the Midwest and East Coast
then previously thought. (See the related
news article in this issue on page 18.) In the
next 30 years, the USGS says, California has
a 99.7 percent chance of a magnitude 6.7 or
larger earthquake, and the Pacic Northwest
has a 10 percent chance of a magnitude 8 to
9 megathrust earthquake on the Cascadia
subduction zone.
The Federal Emergency Management
Agency (FEMA) has estimated the average
annualized loss from earthquakes nationwide
to be $5.3 billion. According to FEMA, 77 per-
cent of that gure ($4.1 billion) comes from
California, Washington, and Oregon, with 66
percent ($3.5 billion) from California alone.
So, an ongoing effort by the USGS and
partner agencies and institutions to estab-
lish a West Coast Earthquake Early Warning
(WC-EEW) system as the prototype for an
eventual nationwide ShakeAlert system
seems especially timely.
The early warning system exploits physi-
cal characteristics of earthquakes, which
generate two main types of waves: rapidly
moving primary or P-waves and the slower
secondary (S) and surface waves that cause
more intense and damaging ground shaking.
(See accompanying gure.)
By detecting and analyzing the location
and magnitude of an earthquake reected in
the P-wave energy, expected ground-shak-
ing levels across a region can be estimated
and warnings sent to local populations be-
fore more damaging shaking arrives with or
after the S-wave. The advanced warning can
range from seconds up to more than a min-
ute, depending on the distance an affected
area is from the earthquakes origin.
Ken Hudnut, a geophysicist at the USGS
Earthquake Science Center in Pasadena,
California, and chair of the GNSS Working
Group for the WC-EEW, has a long history in
working in the area of geohazards. Dr. Hudnut
received an A.B. degree in Earth sciences from
Dartmouth College and a Ph.D. in geology
from Columbia University. Before joining the
USGS in 1992, he was a post-doctoral fellow at
the California Institute of Technology Seismo-
logical Laboratory and currently is a visiting
associate in geophysics on the faculty of the
California Institute of Technology.
We called on Dr. Hudnut to discuss the state
of the art in seismic science and the role of
GNSS in that research and in the design and op-
eration of earthquake early warning systems.
The USGS has a long history of
developing instrumentation for the
study of earthquakes and other types
of Earth movement. What does GNSS
positioning bring to the task that
seismic sensors do not provide and,
more specically, how do GPS/GNSS
data benet EEW systems?
HUDNUT: GNSS positioning is especially good
at rapidly giving us the change in a stations
position. Seismic sensors measure vibrations
very well, but GNSS is better at measuring
permanent displacement.
In a big earthquake, a station might move
by several meters in several seconds, and
not just in a simple straight line. The shaking
may include erratic oscillatory displace-
ments that are several times larger than
the permanent displacement. Even though
GNSS was never intended to measure such
large, sudden, and jerky movements, we nd
that it works very well and provides a great
augmentation to the seismic sensors that
are currently in use for earthquake early
warning.
How is GNSS data different from that
obtained from these seismic sensors
and how is it merged in an EEW system?
HUDNUT: GNSS data add to system robust-
ness because they are an independent
measurement. The seismic sensors are
basically a mass on a spring, whereas GNSS is
measuring position variation using changes
in ranges to a constellation of satellites, so
its a totally different kind of observation.
Having both types of data makes the system
stronger because we can immediately rule
out glitches coming from one sensor type or
the other. The diversity of observations
gives us more strength.
As for merging the data, there is an
GNSS & GEOHAZARDS:
EARLY WARNING FOR LIFE ON THE FAULTLINE
TL S
NovAtels Thought Leadership Series
While those of us living on the Pacic Ring of Fire and other
earthquake-prone regions wait for the Big One, seismic
researchers and public agencies have not been idle.
NovAtels Company Values
Innovation and Integration are
cornerstones of our business,
we believe that excellence is
the standard and we always
encourage new ideas.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
P
h
o
t
o

b
y

J
o
h
n

L
i
v
z
e
y
abundance of literature and we are test-
ing everything from uncoupled to loosely
coupled and tightly coupled, and using a
variety of methods. There are trade-offs in
terms of simplicity, speed, and smoothing
that were evaluating on an ongoing basis. We
are creating a hair-triggered system that is
also very robust even during large dynamic
displacements, which is not a slam dunk.
Weve learned a lot by studying the methods
of combining being done for strap-down
avionics, that is, navigation and positioning
systems, and looking at both commercial off-
the-shelf solutions and open-source options.
What GPS/GNSS signals, satellite
observables, and signal components
(e.g., code vs carrier phase) are used in
the EEW system?
HUDNUT: Right now, we are mostly reliant
on GPS alone, but we have upgraded to GNSS
receivers at our stations over the past several
years. Of course were doing phase-differ-
ential, dual-frequency processing to get the
few-centimeter accuracy in real-time; so, we
do widelaning and narrowlaning, but code is
relatively unimportant to us we rely heavily
on the carrier phase. Were using precise
point positioning with ambiguity resolution,
which is possible for GPS these days and
in the future may be possible for GLONASS
as well. Our limited telemetry bandwidth
doesnt allow us to bring back all of the
GLONASS and other GNSS data just yet.
What practical benets are provided by
an impending seismic movement alert
on the order of tens of seconds?
HUDNUT: Applications envisioned are getting
school children to safety under their desks
that much sooner, and operating automatic
shut-off valves, putting computer systems
into a safer state, or switching other auto-
mated systems to try to prevent loss of life
or damage to property. If you were having
surgery performed at that time, wouldnt you
want the surgeon to remove the scalpel to
safety right before the shaking started?
We want to make it possible for people to
invent their own applications, and we expect
this to happen here as it has in Japan, Mexi-
co, and other countries that have already had
EEW for many years and even decades. In
Japan, EEW protects the Shinkansen (bullet
train) system. In California, BART is testing
use of EEW and gures it could help prevent
or lessen the severity of future derailments.
In recent years, a number of
demonstration campaigns have been
conducted, involving public agencies
and citizen participants in sending
and receiving test notications of an
earthquake. What have been some the
most important lessons learned from
those campaigns?
HUDNUT: ShakeOut is our annual public drill
to encourage Drop, Cover, and Hold On
by everybody. We started this in 2008 in
California and it has grown worldwide. We
use that as an earthquake hazard awareness
opportunity for publicity for EEW. In general,
ShakeOut encourages a personal action that
could be done even quicker if one had an
operational public EEW.
With the ShakeAlert EEW system, what
we have been doing for the past couple of
years is a slow roll-out through selected
beta-users. We dont want to roll this out
to the public before its ready because of the
cry wolf gotcha. Most county-level and
large cities emergency operation centers,
plus Caltrans and BART for example, have
the ShakeAlert UserDisplay installed so that
they could potentially relay an alert through
dispatch communications systems.
Ken Hudnut
U.S. Geological Survey
NOVATELS SPONSORSHIP Our customers have ideas.
Lots of them. Turning those ideas into a competi-
tive advantage is what we do. NovAtels integrated
global positioning solutions deliver success time
and time again on land, sea, and in the air. We
help many of the worlds leading companies stay
in the lead by consistently delivering OEM global
satellite positioning products that are recognized
for their technical innovation, unsurpassed quality
and industry-leading customer support.
If you were
having surgery
performed at the
time of an
earthquake,
wouldnt you want
the surgeon
to remove
the scalpel
to safety before
the shaking
started?
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
www.intergeo.de
Host: DVW e.V.
Conference organiser: DVW GmbH
Trade fair organiser: HINTE GmbH
7 9 October 2014
Berlin, Exhibition Grounds
along with
3
rd
National INSPIRE Conference 2014
imaGIne-2 Conference
Sponsors:
short- or long-term. Such errors would include biases result-
ing from unmodeled atmospheric efects, satellite orbital
errors, and so forth.
Measurement blunders would also be considered system-
atic errors. In fact, equation (7) is used when assessing the
reliability and integrity of a positioning system in the pres-
ence of blunders.
Random errors such as multipath and noise, however, are
usually treated a bit diferently. Specifcally, these errors are
usually well characterized by their standard deviation only
(i.e., no bias), meaning their efect can be completely refected
in the measurement covariance matrix.
If this is the case, the efect of these errors on the solution
is directly obtained from the covariance matrix of the esti-
mated parameters, which is computed as
Tis is a by-product of the law of propagation of variances.
As before the result is afected by the measurement geometry
and the measurement accuracy.
Summary
Tis article looked at how measurement errors propagate into
positioning errors. Te primary factors afecting this propa-
gation are measurement geometry and the measurement
accuracy. Tis explains the motivation for receivers that
minimize measurement errors (especially multipath) and
that track as many satellites as possible.
GNSS SOLUTIONS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
____________________________________
____________
GPS Networking has the experience and technical expertise to provide
customized GPS Signal Distribution components to solve any GPS application
challenge. With over 20 years of experience, our record speaks for itself. We
are unsurpassed in the GPS industry for quality and customer service.
Real World GPS Signal Solutions
RELIABLE: Experience Matters
ACCURATE: Trust is built on it
AVAILABLE: Timing and Synchronization
CUSTOM: Built for your needs
800.463.3063 www.gpsnetworking.com
P|ease visit us at ION GNSS+ Booth 526 Connect with us on Linked In
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
____________________________
I
n the past 20 years GPS has simul-
taneously revolutionized both our
modern infrastructure (by provid-
ing real-time navigation, mapping,
and timing support) and our geodetic/
surveying capabilities (by providing
millimeter/centimeter-level position-
ing). At this point, most of the GNSS
innovations we expect to see in the
next decade will come from calculating
positions more accurately and faster,
while expanding from GPS to use of all
available GNSS signals.
Twenty years ago, in an article in
ESA Journal (see Additional Resources
section near the end of this article)
Manual Martin-Neira presented a
new application for GNSS. Instead of
processing the direct GNSS signals for
positioning, timing, and atmospheric
studies, Martin-Neira suggested
employing refected GNSS signals
as the measurement. Te frst GNSS
refection experiments were focused on
altimetry, ocean winds, and soil mois-
ture; later researchers evaluated GNSS
refectometry for sensing snow/ice and
measuring vegetation growth.
Each of these refection studies used
GNSS instruments specially designed to
measure refected signals. In contrast,
geodesists and surveyors use GNSS
instruments that we know are designed
to suppress refected signals (more
commonly referred to as multipath).
While these refections are known to
afect the accuracy of positions derived
from these instruments, there is still no
standardized approach that models (and
eliminates) the efect of refections.
Increasingly sophisticated
uses of GNSS observables
have led to a new era in
remote sensing. A team of
researchers describe the
results of the applications of
interferometric reectometry
to measure snow depth,
vegetation water content,
and soil moisture.
Environmental Sensing
A Revolution in GNSS Applications
KRISTINE M. LARSON
DEPARTMENT OF AEROSPACE ENGINEERING
SCIENCES, UNIVERSITY OF COLORADO
ERIC E. SMALL
GEOLOGICAL SCIENCES, UNIV. COLORADO
JOHN J. BRAUN
COSMIC, UNIVERSITY CORPORATION
FOR ATMOSPHERIC RESEARCH
VALERY U. ZAVOROTNY
EARTH SYSTEM RESEARCH LAB/NOAA
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
In principle, this would suggest that
GNSS refection research is irrelevant
for the tens of thousands of geodetic-
quality GNSS receivers currently in
operation around the world. Certainly
GNSS refections were never consid-
ered as a potential source of soil mois-
ture, snow depth, and vegetation data
when the EarthScope Plate Boundary
Observatory (PBO) <http://pbo.earth-
scope.org> was built in the western
United States between 2005 and 2008.
Unexpectedly, we have shown
that these environmental data can be
retrieved from PBO data without any
instrumentation beyond the existing
GNSS data stream. We can do this
because the multipath data turn the
GNSS site into a quasi-interferometer.
Te distance between the antenna and
the surface refecting material derived
from the interferometric efect will tell
us whether the top of the surface has
moved. Tis means we can use the data
to measure snow depth by comparing
it to data when there is no snow.
If the refected signal travels
through vegetation, the interferometer
will show two efects: the primary
refection is caused by the top of the
soil layer and secondarily, the ampli-
tude of the refected power will be
smaller because it interacted with water
in the vegetation. Changes in soil mois-
ture cause the smallest changes to the
interferometric efect. We can think
of these as being caused by the signals
being refected by the surface soil layers
having various wetness levels.
Tese new measurements of soil
moisture, snow, and vegetation measure-
ments (called the PBO H2O network)
are needed both for climate studies and
satellite validation. Water managers
use the data to predict, and hopefully
mitigate, hazards such as foods and
droughts. Tese new GNSS environmen-
tal data fll a niche between existing sat-
ellite sensors (that have very large foot-
prints) and other in situ sensors (which
tend to have very small footprints).
Tis article describes how we have
created an operational GNSS environ-
mental sensing network. We will frst
describe the network itself, followed by
an overview of how refections manifest
themselves in GNSS observations, and
ending with examples of environmental
signals we have measured using this
network in the western United States.
The Plate Boundary Observatory
and Reections
Consisting of about 1,100 stations, PBO
was built by UNAVCO <http://www.
unavco.org> under a contract with
the U.S. National Science Foundation
with the scientifc goal of studying
Most PBO sites are operated with banks of batteries that are powered by solar panels,
as shown at PBO site P422. Inset: PBO site P101 in Randolph, Utah, which is used to
measure snow depth in the winter and vegetation/soil moisture in the spring, summer,
and fall. UNAVCO photos
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
___________
_______
______
the motion of tectonic plates and the
deformation of the North American
continent. Te locations of PBO sites
(Figure 1) were chosen to address spe-
cifc geophysical problems; thus, half of
the GNSS sites are near fault zones in
California. Te east-west trending sites
in Nevada and Utah are measuring
motion in the Basin and Range prov-
ince. Clusters of sites are also located
near volcanoes (e.g. Yellowstone,
Mammoth, Mount St. Helens, and the
Aleutian arc).
Afer site locations were selected,
the PBO project made special eforts
to attach the GPS antenna to bedrock.
Teir reason for doing so was to ensure
that the position (and velocity) infor-
mation measured at each site would
represent motion related to faults and
volcanoes. For this reason, almost no
PBO sites are located on buildings.
Figure 2 presents a schematic of a
typical PBO site. A dual-frequency,
choke-ring antenna is protected in an
acrylic radome with a nearby equip-
ment box housing the receiver and
the telemetry hardware. Te antennas
drill-braced monument is anchored
to a depth of nearly three meters.
Te standard PBO site operates a
dual-frequency carrier phase receiver
collecting GNSS signal data with a
15-second sampling interval; most
also support 1-second sampling. With
the exception of some sites in Alaska,
the data are telemetered to the central
UNAVCO facility in Boulder, Colo-
rado, afer midnight UTC each day;
fles of carrier phase and pseudorange
data (called RINEX fles) are produced
by UNAVCO and posted online for
public access soon afer. Geophysicists
are able to download the RINEX fles
for reprocessing, or they can download
the frequently updated position time
series in a standard terrestrial reference
frame.
Velocity products that are used
to study faults, earthquakes, and
volcanoes, are also produced for the
geophysical community. Tese PBO
positioning products are based on very
detailed models of the GNSS space-
craf, propagation delays, and Earth
motions.
Although geophysicists and
geodesists are well aware of the
negative efects of refected signals,
there is still no standard model to
remove refection/multipath from
these position/velocity products. Tis
is partly because each GNSS site has
unique refection characteristics.
Furthermore, many eforts to model
multipath rely on stacking carrier
phase residuals from least squares
analyses. In principle these residuals
could be used for environmental
sensing; however, they can and will
be infuenced by mismodeled carrier
phase data. Consequently, parameters
in the least squares analysis could
thus absorb or mask what was a real
environmental change.
On the other hand, if one thinks
about how best to measure multipath
refections rather than trying to model
multipath corrections for carrier phase
data, one might recast the problem to
use signal power data. Tese are the
analogous data to what is being used
by the GNSS refectometry commu-
nity, which typically uses two receiv-
ers/antennas to separately measure the
direct and refected signal. Te GNSS
units used by geodesists and survey-
ors produces a single data stream
and measurements that represent the
interference of the direct and refected
signal. In the latter case, the antenna
is not tuned to measure the refected
signal as it is with traditional GNSS
refectometry.
So, a key question arises: Are the
signal power data collected by geo-
ENVIRONMENTAL SENSING
FIGURE 1 Circles represent locations of Plate Boundary Observatory (PBO) GNSS sites. Blue and cyan
colored circles represent locations for PBO H2O product release versions 1 and 2.
FIGURE 2 Schematic illustration of a typical PBO site. The radome protects the choke-ring antenna.
Most sites are powered by solar panels. Except for some sites in Alaska, the data are telemetered at
least daily. Credit: UNAVCO.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Booth H
Precise
Positioning
+ Attitude
for Unmanned Vehicles
Leverage Trimbles leadership in precise GNSS/inertial positioning, navigation,
guidance, and aerial mapping for all your unmanned applications.
Unmanned. Unlimited.
2014 Trimble Navigation Limited. All rights reserved. TPC090 (08/14)
www.trimble.com/unmanned
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
detic/surveying GNSS units of sufcient quality to become
inadvertent environmental refectometers?
GNSS receivers generate carrier-to-noise density ratio
data, which are stored in a RINEX fle as signal-to-noise
ratio (SNR) observables. Figure 3 shows representative SNR
data set collected by a geodetic-quality GNSS receiver. Te
direct signal has a simple polynomial shape, with lower
SNR magnitudes at the rising and setting sections of the
satellite track. Tese lower values primarily result from the
antenna gain pattern. Superimposed on the direct signal are
the refected signals, which for horizontal planar refectors
manifest themselves as oscillations. Note particularly that
little evidence of refected signals appears above elevation
angles of around 25 degrees. Tis is again due to the antenna
gain pattern.
Te transmitted GPS signal is right hand circularly
polarized (RHCP). Te refection will have both RHCP and
LHCP (lef hand circularly polarized) components. As seen
in Figure 4, the refection coefcients are diferent for RHCP
and LHCP, and depend on both the refection surface and the
satellite elevation angle.
Te frequency of the refected SNR signal is dominated by
geometry, i.e., the extra path length traveled by the refected
signal, as seen in Figure 5. For a planar horizontal refector,
the frequency of the interference of the direct and refected
signal observed in SNR data is constant as a function of sine
of the elevation angle. It is straightforward to extract this
dominant frequency using a periodogram or estimate of the
spectral density of the signal, a quantity that we call the efec-
tive refector height.
If the efective refector height changes, this means that
the surface layer around the antenna changed. For example,
an efective refector height would change from 2.0 to 1.8
meters if it snowed 0.2 meters. To convert these efective
refector heights into an absolute measure of snow depth, we
compare efective refector heights estimated during the win-
ter months with efective refector heights determined when
no snow is on the ground.
Te amplitude of the refection observed in the SNR data
depends on the dielectric constant of the surface material
and, thus, very wet snow produces a diferent amplitude than
very dry snow. Likewise, vegetation with high water content
has much smaller SNR amplitudes than vegetation with very
low water component. Tis is the principle used to defne the
vegetation statistic reported by PBO H
2
O.
In order to defne the snow depth, soil moisture, and veg-
etation water content measurements more rigorously we have
developed forward models that contain information about
the transmitted GPS signal, the gain pattern for the antenna
used by PBO, and refection coefcients for natural surfaces.
Tese models have guided us in developing retrieval algo-
rithms, which have been automated for PBO H
2
O and pub-
lished in the refereed literature. As part of this efort we have
FIGURE 3 SNR data from the L2C signal collected at PBO site P041 for
a single satellite track are shown in black. The smooth blue curve
represents the direct signal. Elevation angles are shown in gray. The
oscillations at low elevation angles are indicative of reected signals/
multipath effects.
FIGURE 5 Schematic of multipath geometry for a horizontal planar surface.
The direct L2 signal (shown in blue) is reected at a planar surface
and travels an additional distance (shown in red). Elevation angle is
depicted by e. The GNSS unit measures the interference between the
direct and reected signals (examples of this interference are shown in
the inset).
FIGURE 4 Reection coefcients for a variety of natural surfaces at GPS
frequencies and RHCP (dashed) and LHCP (solid) signals
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
also conducted validation experiments where we measured
soil moisture, snow depth, and vegetation water content in
situ. Tese experiments have been invaluable in allowing us
to improve our algorithms.
Results from PBO H
2
O
Te PBO H
2
O initiative grew out of experiments conducted
near Boulder, Colorado between 2007-2009. Afer several
years of developing models and retrieval algorithms, the PBO
H
2
O network began operating in October 2012 with a data
portal providing online access to users <http://xenon.colora-
do.edu/portal>. Figure 1 provides the location of the approxi-
mately 350 current sites along with about 200 new sites for
which we plan to begin distributing data in the fall of 2014.
Data are downloaded from the central UNAVCO archive
every evening, and new solutions for soil moisture, snow
depth, and vegetation water content are posted each morn-
ing. To aid in quality control for our products, we also down-
load other environmental datasets, such as hourly samples of
modeled precipitation and temperature data from the North
American Land Data Assimilation System and snow cover
data from NASAs satellite-based Moderate Resolution Imag-
ing Spectroradiometer (MODIS) project. Tese are useful for
identifying outliers in our vegetation and soil moisture prod-
ucts. Photographs, Google maps, digital elevation maps, and
climatology information are also provided for each site.
Te following sections describe a few examples from each
environmental dataset.
Snow. Our frst snow depth measurements were made in
2009 at a fat mesa site south of Boulder. Although the snow
depth retrievals were successful, we needed to demonstrate
that the technique would work in more challenging environ-
ments. Figure 6 shows the next snow site we tested. We chose
a Niwot Ridge, Colorado, site because of its topographic vari-
ability (due to its location in a saddle at an elevation of about
3,500 meters), extreme cold, and very high winds. Power
and Internet access was available from an existing scientifc
installation.
Five years later, the GPS snow depth time series from
this site shows that the refection method is robust, with
very few data outages. Comparisons with in situ data
(the pole in the photograph is measured roughly every
two weeks) show that the method is also very accurate.
Although the monument is three meters tall, as seen in
the inset photograph in Figure 6, the antenna was almost
buried in spring 2011. Te latter was a banner snow year
throughout the western United States, and a handful of
PBO antennas were buried at snow peak.
Figure 7 shows snow levels measured at a PBO H
2
O site
near Island Park, Idaho. Unlike the station position time
series generated for this site by geophysicists, which shows
almost no variability, the snow changes at the site are quite
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_________
_____________
_______________
www.NavtechGPS.com
+1-703-256-8900 - 800-628-0885
www.NavtechGPS.com
+1-703-256-8900 - 800-628-0885
Your ONE Source for GNSS
Products and Solutions

546: Inertial Systems, Kalman Filtering, and GPS/INS
Integration. Taught by Dr. Alan Pue and Mr. Michael Vaujin
(San Diego, Dec. 8-11, 2014)
346: GPS/GNSS Operations for Engineers and Technical

Professionals. Taught by Dr. Chris Hegarty
(San Diego, Dec. 8-11, 2014 )
541: Using Advanced GPS/GNSS Signals and Systems.

Taught by Dr. John Betz
(San Diego, Dec. 8-11, 2014)
336: GPS Fundamentals, Enhancements and Intro to

Diferential GPS. Taught by Dr. Chris Hegarty
(Noordwijk, Nov. 17-19, 2014)
Questions? Contact cmcdonald@navtechgps.com
Contact us to learn how the AsteRx3 can be inte-
grated into your project to mitigate intentional
or unintentional GNSS signal interference.
With Advanced Interference Mitigation
Technology from Septentrio (AIM+)
dynamic. Te frst snowfall generally occurs at the same time
each fall, but the peak snow amount is highly variable, as is
the timing of snowmelt. Te latter measurement is particu-
larly important for predicting potential fooding. A video
combining a series of photos of the station with correspond-
ing weekly snow level plots may be viewed online at <http://
FIGURE 7 Five years of snow depth time series for the GPS site at P360 in
southern Idaho.
FIGURE 6 Top: ve years of snow depth time series for the GPS site at
Niwot Ridge, Colorado; bottom: Niwot Ridge GPS installation photo-
graphed in fall and early spring.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_______________
____
The AsteRx3 OEM Receiver
World-Class
GNSS Courses
San Diego, California
(December) and Noordwijk,
The Netherlands (November)
AIM+ technology
Multi-frequency GPS/GLONASS/Galileo
L1/L2/L5/E5a/E5b/E5 AltBOC code/carrier tracking
COMPASS ready
Patented Galileo AltBOC tracking
APME, RAIM and LOCK+ tracking
RTK+
Versatile OEM Receivers for Demanding Applications
Reshaping
the limits of
positioning.
xenon.colorado.edu/spotlight/index.php?product=spotlight&
station=p360>.
Vegetation. PBO H
2
O vegetation measurements (NMRI,
Normalized Microwave Refection Index) are based on
changes in the refection amplitude, where values of zero
represent the vegetation with the lowest water content. Fig-
ure 8 shows the vegetation data for a GNSS site in eastern
Wyoming designated P042. Tis fgure compares the NMRI
vegetation water content estimates derived from GPS satel-
lite data with the sites normalized diference vegetation
index (NDVI). Te latter are optical measurements typi-
cally generated at 16-day intervals using MODIS sen-
sors that measure greenness, with each pixel representing a
250-square-meter footprint.
Greenness correlates strongly with photosynthesis pro-
duction, and thus NDVI is commonly used to study vegeta-
tion growth. To provide some context, Figure 8 also shows
modeled precipitation data (which is not directly measured
at PBO sites). A close correlation appears between the GPS
NMRI data and NDVI (correlation coefcient of 0.86). Par-
ticularly note the absence of greenness in 2012 and low GPS
FIGURE 8 Top: GPS site at Wheatland, Wyoming; middle: GPS vegetation
measurements (blue) compared with Normalized Difference Vegetation
Index (green); bottom: cumulative precipitation from the North Ameri-
can Land Data Assimilation System (NLDAS). The GPS vegetation index
is also called the Normalized Microwave Reection Index (NMRI).
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
________________________________________
_________
_____________________________________
values during the 2012 drought. Tat year had low snowfall,
a very hot spring, and very little rain. Te P042 data record
also shows a double growth peak in 2008. Tis phenomenon
occurs when there are large gaps between
rainfalls.
Similar sensitivity to vegetation
growth is shown for California site P532
(Figure 9). Here both the GPS and NDVI
records show strong sensitivity to the
efects of drought, including 2014. Note
that, although the rains in late February
2014 brought cumulative precipitation
levels up to near the fve-year average,
vegetation water content as measured
by NMRI has not recovered. Droughts
efects in 2007 and 2009 are also clearly
visible. A further note: the GPS vegeta-
tion data have a shorter season length
than the NDVI data, with NDVI having
a consistently longer growth season than
the GPS measurement of vegetation water
content.
Because GNSS refections are sensitive
to vegetation water content and NDVI is
correlated to chlorophyll production, the
combination of these measurements pro-
vides better constraints to phenologists studying the infu-
ence of climatic variations on periodic plant life cycles.
Soil Moisture. Soil moisture is the most challenging water
cycle parameter to measure with GNSS receivers and faces
some limitations. First, the refection technique cannot mea-
sure soil moisture if there is snow on top of the soil. For PBO
H
2
O sites in the Rocky Mountains, we must remove data
afected by snow. Second, soil covered by vegetation with very
high water content (such as alfalfa) requires a more complex
model of the refections than we currently use.
Even with these restrictions, we have found many PBO
sites that generate accurate soil moisture records. Figure 10
shows such a record from a GNSS site near San Jose, Cali-
fornia. Note that there is strong correlation between soil
moisture changes and precipitation events, and then there is
a dry down. Tis is consistent with the behavior of a shal-
low (0-5 centimeter) soil moisture instrument. Te sensing
depth of the GNSS method is determined by its transmission
frequency (L-band).
Although measurements of soil moisture are needed at
depth as well as the surface, these GNSS data are particu-
larly useful for satellite validation (ESAs SMOS mission and
NASAs upcoming SMAP launch) because these sensors also
operate at L-band.
Expanding PBO H2O to International GNSS Networks
Although the initial emphasis of our project was to use data
from the PBO network, we must stress that no technical
reason exists which prevents GNSS instruments operated by
surveyors and transportation agencies from being used for
environmental sensing. Both geophysicists and surveyors use
dual-frequency carrier phase GNSS receivers and, if prop-
erly confgured, such receivers can generate SNR data that are
FIGURE 9 GPS vegetation growth index (NMRI) compared with NDVI and
accumulated NLDAS precipitation at PBO site P532 located 50 miles
northwest of Santa Barbara, California.
FIGURE 11 Top: GNSS site operated by the Mesa County Surveying Network located in Snowmass,
Colorado; time series of snow depth for COA1 is shown for the 2014 water year; Bottom: the GNSS
station operated near Ashland (ASHL) by the Minnesota Department of Transportation is located
on the side of a road. The GNSS satellite tracks from the west of the monument can be used to
measure snow depth.
FIGURE 10 Daily measurements of volumetric soil moisture measured with
GNSS (blue) and daily precipitation from NLDAS.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
FUTURE
PROOF
E
A
S
Y
T
O
U
S
E
PO
W
E
R
F
U
L
AFFORD
A
B
L
E
Upgradable to all
constellations and signals
More power than
you ever imagined
Ease-of-use and affordable functionality earned our GPS/GNSS simulators a place on
the industrys production test benches. Now our fully congurable, eld-upgradeable
platform moves into R&D laboratories to test the most challenging applications.
Multi-constellation integration
MEMs/INS simulation and testing
Interference detection and mitigation
Automotive applications
Assisted-GNSS for consumer devices
High dynamic platforms for military
and defense
Precision agriculture/surveying:
RTK/Differential measurements
Put us to the test today
Contact us with your most demanding
GPS/GNSS testing requirements, see
for yourself how easy it is to add the
powerful, application-specic features
you need to the Spectracom line of
GPS/GNSS Simulators.
North America +1.585.321.5800
sales@spectracomcorp.com
France +33 (0) 1 6453 3980
UK +44 (0) 1 256 303630
Brazil +55 12 3882 4765
tinyurl.com/Spectracom-GNSS
THE GSG SIMULATOR SERIES BY SPECTRACOM
suitable for refectometry applications.
Confguration examples include
requesting that the receiver track both
legacy (L2 P-code) and new civilian
(L2C) signals. Te latter is preferred for
refection research because the code is
public and the extracted signal power
is higher. Second, some receivers pro-
duce SNR data rounded to the closest
integer by default, which the station
operator can easily change so that it
generates a more precise SNR data
stream.
To demonstrate that surveyor-oper-
ated GNSS sites can be used as snow
sensors, we have recently partnered
with two surveying organizations to
expand PBO H
2
O. Figure 11 shows two
examples from these eforts. In Colo-
rado we have accessed data from the
Mesa County Real-Time Virtual Refer-
ence Network <http://emap.mesacou-
nty.us/GPS_Survey/GPS_Survey.htm>;
the Minnesota data are distributed by
the State Department of Transporta-
tion. Because the Minnesota sites tend
to be located near highways, we used
Google Earth images to window the
data we used to measure show depth.
For the Colorado sites, we used both
photographs and Google Images. In
both cases accuracy of the snow depth
estimates is equivalent to that recov-
ered from the PBO sites.
Can we measure soil moisture,
snow depth, and vegetation at all GNSS
sites? Unfortunately, the short answer
is no. Many GNSS sites have been
installed on buildings and/or near
parking lots, where refections would
be of little interest for the purposes
described in this article. Te locations
of these sites also produce degraded
positioning accuracy, but the degrada-
tion is ofen acceptable to the primary
users of the data geodesists, survey-
ors, and others.
Te second limitation to using
GNSS networks for environmental
sensing has to do with data availability.
While many organizations provide a
RINEX fle to national archives such as
CORS, ofen these RINEX fles do not
include the SNR data. Furthermore,
some archives degrade the RINEX fles
by eliminating observables and deci-
mating the remaining data. Tis makes
it difcult and in some cases impos-
sible to extract useful environmen-
tal data from these. Although these
issues constrain the use of data from
some existing GNSS sites, we hope
that results from PBO H
2
O encourages
future installations in locations that
can measure positions and environ-
mental changes simultaneously.
Final Remarks
Geodesists, geophysicists, and survey-
ors have all established large GNSS
networks. Nearly all of them have open
data policies and encourage broad
usage of their data. Te vast majority of
GNSS data users focus on positioning,
although the timing and atmospheric
communities also value data from
GNSS networks. Here we have shown
how to further extend the value of
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
______________
_________________________
__________________
ground GNSS networks by describing
how to routinely measure soil moisture,
snow depth, and vegetation growth.
Tese data are valuable both to scien-
tists and water managers and a cost-
efective use of existing infrastructure.
Acknowledgments
Tis work has been a collaboration
that includes many colleagues, stu-
dents and post-docs: Felipe Nievinski,
Ethan Gutmann, Andria Bilich, Karen
Boniface, James McCreight, Cheney
Shreve, Clara Chew, Sarah Evans, Evan
Pugh, John Pratt, Penina Axelrad, and
Praveen Vikram. Te PBO H2O portal
is supported by NSF EAR-1144221 and
NASA NNX12AK21G. PBO is oper-
ated by UNAVCO for EarthScope, and
supported by NSF (EAR-0350028 and
EAR-0732947).
Manufacturers
Te standard PBO sites use NetRS
GNSS receivers from Trimble, Sunny-
vale, California, USA.
Additional Resources
GNSS-Reectometry
[1] Cardellach, E., and F. Fabra, A. Rius, S. Pet-
tinato, and S. DAddio, Characterization of
dry-snow sub-structure using GNSS reected
signals, Remote Sensing of Environment, Vol.
124, 122-134, 2012
[2] Egido A., and M. Caparrini, R. Rufni, S.
Paloscia, E. Santi, L. Guerriero, N. Pierdicca, and
N. Floury, Global Navigation Satellite Systems
Reectometry as a Remote Sensing Tool for
Agriculture, Remote Sensing, Vol. 4(8), 2356-
2372, doi:10.3390/rs4082356, 2012.
[3] Garrison, J. L., and S.J. Katzberg, The
Application of Reected GPS Signals to Ocean
Remote Sensing, Remote Sensing of Environ-
ment, Vol. 73(2), 175-187, doi:10.1016/s0034-
4257(00)00092-4, 2000
[4] Garrison J. L., and A. Komjathy, V.U. Zavorot-
ny, and S.J. Katzberg, Wind speed measurement
using forward scattered GPS signals, IEEE Trans-
actions on Geoscience and Remote Sensing, Vol.
40(1): 5065, 2002
[5] Gleason S, S. Hodgart S. Yiping, C. Gom-
menginger, S. Mackin, M. Adjrad, and M.
Unwin Detection and Processing of bistati-
cally reected GPS signals from low Earth
orbit for the purpose of ocean remote sensing,
IEEE Transactions on Geoscience and Remote
Sensing, Vol. 43(6),1229-1241. doi:10.1109/
TGRS.2005.845643, 2005
[6] Katzberg S.J., O. Torres, M.S. Grant, and
D. Masters, Utilizing calibrated GPS reected
signals to estimate soil reectivity and dielec-
tric constant: Results from SMEX02. Remote
Sensing of Environment, Vol. 100(1), 17-28.
doi:10.1016/j.rse.2005.09.015, 2005
[7] Martin-Neira M., A Passive Reectometry
and Interferometry System (PARIS)-Application
to Ocean Altimetry, ESA Journal, Vol. 17(4), 331-
355, 1993.
[8] Ruf, C., and A. Lyons, M. Unwin, J. Dickinson,
R. Rose, D. Rose, and M. Vincent, CYGNSS:
Enabling the Future of Hurricane Predic-
tion, IEEE Geoscience and Remote Sensing
Magazine, Vol. 1(2), 52-67, doi: 10.1109/
MGRS.2013.2260911, 2013
[9] Semmling, A. M., and T. Schmidt, J. Wickert,
S. Schon, F. Fabra, E. Cardellach, and A. Rius, On
the retrieval of the specular reection in GNSS
observations for ocean altimetry, Radio Sci-
ence, Vol 47, doi:10.1029/2012RS005007, 2012
[10] Yang, D., and Y. Zhou and Y. Wang, Remote
Sensing with Reected Signals: GNSS-R Data
processing Software and Test analysis, Inside
GNSS, September/October 2009, pp. 4145
GNSS-Interferometric Reectometry
[1] Chew, C. C., and E. E. Small, K. M. Larson, and
V. Zavorotny, Effects of Near-Surface Soil Mois-
ture on GPS SNR Data: Development of a Retriev-
al Algorithm for Volumetric Soil Moisture,
IEEE Transactions on Geoscience and Remote
Sensing, Vol. 52(1), 537-543, doi:10.1109/
TGRS.2013.2242332, 2014
[2] Larson, K. M., E. E. Small, E. Gutmann, A.
Bilich, J. Braun, and V. Zavorotny, Use of GPS
receivers as a soil moisture network for water
cycle studies, Geophysical Research Letters,
Vol. 35, L24405, doi:10.1029/2008GL036013,
2008
[3] Larson, K. M., E. Gutmann, V. Zavorotny,
J. Braun, M. Williams, and F. Nievinski, Can
We Measure Snow Depth with GPS Receivers?
Geophysical Research Letters, Vol. 36, L17502,
doi:10.1029/2009GL039430, 2009
[4] Larson, K. M., and F.G. Nievinski, GPS Snow
Sensing: Results from the EarthScope Plate
Boundary Observatory, GPS Solutions, Vol 17(1),
41-52, doi 10.1007/s10291-012-0259-7, 2013
[5] Nievinski, F. G., and K.M. Larson, Forward
modeling of GPS multipath for near-surface
reectometry and positioning applications, GPS
Solutions, Vol. 18(2), 309-322, doi:10.1007/
s10291-013-0331-y, 2014
[6] Ozeki, M., and K. Heki, GPS snow depth
meter with geometry-free linear combinations
of carrier phases, Journal of Geodesy, Vol.
86(3), 209219, doi:10.1007/s00190-011-
0511-x, 2012
[7] Small, E. E., and K. M. Larson and J. J. Braun,
Sensing Vegetation Growth with GPS Reec-
tions, Geophysical Research Letters, Vol. 37,
L12401, doi:10.1029/2010GL042951, 2010
[8] Zavorotny, V., and K. M. Larson, J. J. Braun,
E. E. Small, E. Gutmann, and A. Bilich, A
physical model for GPS multipath caused by
ground reections: toward bare soil moisture
retrievals, IEEE Journal of Selected Topics in
Applied Earth Observations and Remote Sens-
ing (JSTARS), Vol. 3(1), pp. 100-110, 10.1109/
JSTARS.2009.2033608, 2010
Authors
Kristine M. Larson received the
B.A. degree in engineering sci-
ences from Harvard University
and the Ph.D. degree in geo-
physics from the Scripps Insti-
tution of Oceanography, University of California
at San Diego. She was a member of the technical
staff at the Jet Propulsion Lab from 19881990.
Since 1990, she has been a professor in the
Department of Aerospace Engineering Sciences,
University of Colorado, Boulder. Her research
interests are focused on developing new appli-
cations and techniques for GPS.
Eric E. Small received a B.A.
degree in geological sciences
from Williams College and the
Ph.D. degree in earth sciences
from the University of California
at Santa Cruz. He is a professor in the Depart-
ment of Geological Sciences, University of Colo-
rado, Boulder. His research is focused on land
surface hydrology.
John J. Braun received the B.A.
degree in physics and mathe-
matics from the University of
Colorado, Boulder, and the
Ph.D. degree from the Depart-
ment of Aerospace Engineering Sciences, Uni-
versity of Colorado. He is a project scientist
within the COSMIC program at the University
Corporation for Atmospheric Research, Boulder.
His research interests include developing new
techniques and using GNSS observations to
study the Earth and its environment, particularly
the water cycle.
Valery U. Zavorotny received the
M.Sc. degree in radio physics
from Gorky State University,
Gorky, Russia, and the Ph.D.
degree in physics and mathe-
matics from the Institute of Atmospheric Phys-
ics, USSR Academy of Sciences, Moscow. From
1971 to 1990, he was a research scientist with the
Institute of Atmospheric Physics of the USSR
Academy of Sciences, Moscow. From 19912000,
he was a CIRES Research Associate in the Envi-
ronmental Technology Laboratory of the National
Oceanic and Atmospheric Administration
(NOAA), Boulder, CO, and became a NOAA/ETL
physicist in 2000. His research interests include
theory of wave propagation through random
media, wave scattering from rough surfaces, and
ocean and land remote sensing applications.
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
__________________
T
he small satellite Technologie-
Erprobungs-Trger 1 (TET-1) is
the frst spacecraf developed for
the German Aerospace Center (DLR)
On-Orbit-Verifcation (OOV) program,
which provides f light opportunities
dedicated to testing and qualifcation of
new technologies in space. Te satellite
was lifed into a low-Earth orbit (LEO)
on July 22, 2012, from the launch site in
Baikonur, Kazakhstan.
TET-1 carries various technology
demonstration payloads, among them
the Navigation and Occultation eXperi-
ment (NOX). Tis payload consists of a
geodetic-grade GPS receiver, which is
connected via an antenna selector to two
GPS L1/L2 patch antennas.
One of the antennas is mounted on
the satellites zenith panel and receives
signals primarily used for precise orbit
determination (POD) experiments. Te
second antenna is pointed towards the
anti-fight direction of the satellite for
collecting measurements of low eleva-
tion satellites for ionospheric and tro-
pospheric occultations. The antenna
switch allows to select either the POD
or the occultation antenna for signal
reception.
Tis article describes the NOX pay-
load on board the TET satellite in detail
and analyzes the receivers tracking per-
formance and the accuracy of its naviga-
tion solution. It also presents the initial
tracking results of the occultation anten-
na, which demonstrate that GPS signals
can be tracked through the ionosphere
below the satellites local horizon at a
minimum, even down to the upper part
of the atmosphere with commercial-
of-the-shelf (COTS) equipment.
Spacecraft Design
& Operation
Te spacecraf bus for TET is based to a
large extent on the Bi-Spectral Infra-Red
Detection (BIRD) satellite bus. Te satel-
A test and evaluation
program demonstrates
that a commercial GPS
receiver can operate as a
spaceborne research tool.
The Navigation and Occultation eXperiment
GPS Receiver Performance
On Board a LEO Satellite
ANDR HAUSCHILD, MARKUS MARKGRAF, OLIVER MONTENBRUCK
GERMAN SPACE OPERATIONS CENTER (GSOC) GERMAN AEROSPACE CENTER (DLR), GERMANY
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
lite has a height of 880 millimeters and
a depth of 670 millimeters. In launch
confguration, TET-1s width is 580 mil-
limeters, which increases to 1,540 mil-
limeters with the solar panels deployed.
Its total mass is approximately 120 kilo-
grams, which includes 50 kilograms of
payloads.
TET-1 is equipped with star sensors,
sun sensors, gyroscopes, and magnetic
feld sensors for attitude determination.
A set of four reaction wheels and mag-
netic coils provides three-axis stabiliza-
tion. Te satellite bus system is equipped
with a single-frequency receiver with its
own dedicated patch antennas. Note that
this system, which provides position-
ing and timing for the satellite bus, is
completely independent from the NOX
payload. Te satellite carries 11 payloads
in total, including the NOX, which has
been designed to demonstrate the suit-
ability of COTS technology for space
applications.
Figure 1 shows a schematic illus-
tration depicting the front view of the
TET-1 satellite without its multi-layer
insulation. Te fgure shows the various
components of NOX, which will be dis-
cussed in further detail in the following
section, and illustrates the orientation of
the body-fxed coordinate system of the
satellite.
Te TET-1 satellite is operated in dif-
ferent attitude modes depending on the
payload operation or mission require-
ments. Te two attitude modes relevant
for the operation of the NOX system are
the Earth-pointing mode (EPM) and the
Sun-pointing mode (SPM). In the EPM,
the satellites body-fxed +x-axis points
into the direction of fight and the +z-
axis is oriented towards the center of the
Earth. Te SPM is used to recharge the
satellite batteries. For this purpose, the
satellites solar panels, which are mount-
ed on the -z-panel are pointed towards
the Sun to maximize their power output.
TET-1 was launched into a sun-
synchronous LEO orbit at a height of
approximately 500 kilometers with an
inclination of 97.5 degrees on July 22,
2012. Afer testing all satellite subsystems
and payloads during the commissioning
phase afer launch, the OOV mission was
conducted until October 2013. Te satel-
lite still continues operation and is now
part of the Firebird mission, with the task
of fre detection from orbit.
Overview of the NOX
Te Navigation and Occultation eXperi-
ment on TET-1 has been designed to
demonstrate the suitability of commer-
cial-ofthe-shelf technology for space
applications. Figure 2 provides a sche-
matic of the NOX hardware layout. Te
experiment consists of a dual-frequency
GPS receiver connected to an RF relay,
which allows operators to select one of
two L1/L2 passive patch antennas for
signal tracking. A low noise amplifier
On July 22, 2012, the
rst small German
satellite in the On-
Orbit-Verication
program was carried
into orbit from the
Cosmodrome in
Baikonur, Kazakhstan,
by a Russian Soyuz
launch vehicle. TET-1
is a technology testbed
with 11 experiments on
board that have been
operated in space for a
year. DLR photo. Left:
Artists impression
of the TET-1 small
satellite. DLR/Astro-
und Feinwerktechnik
Adlershof GmbH
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
(LNA) with a gain of 26 decibels at the L1 and L2 frequencies is
used to ensure an adequate signal strength at the receiver input.
Te receiver is standard of-the-shelf hardware. Standard
receiver frmware, however, cannot be used for spaceborne
applications, because height- and velocity-constraints pro-
hibit the operation on board a satellite. Te NOX receiver is
therefore equipped with a special frmware with these NATO
limits removed. In addition to these modifcations, the Dop-
pler search window has been increased 45 kilohertz to facilitate
signal acquisition under high dynamics.
With its 48 channels the receiver can track L1 C/A code
and L1/L2 P(Y) code signals of 16 satellites simultaneously. It
provides pseudorange, carrier-phase, Doppler, and carrier-to-
noise density ratio (C/N
0
) observations for L1 C/A and L2 P(Y)
signals and as well as pseudoranges and C/N
0
for L1 P(Y) at
data rates of up to 10 Hz.
Te receiver underwent extensive pre-fight testing to ensure
its suitability for use in space. Te papers by J. Leyssens et alia
and M. Garcia-Fernandez, listed in the Additional Resources
section near the end of this article, describe this testing in fur-
ther detail. Te receiver board is mounted together with an
interface (IF) board on the -y-panel of the payload compart-
ment of the satellite as shown in Figure 1. Te interface board
serves as the power and commanding interface between NOX
and the TET satellite bus. It also contains a protection circuit
against single-event latch-up efects and a switch line, which
allows the operation of the RF relay for antenna selection.
The antennas of NOX are two identical patch antennas,
which are mounted on diferent sides of the satellite bus struc-
ture without dedicated ground planes or choke rings. Te anten-
na used for precise orbit determination is mounted on the -z-
panel of the satellite. Te POD antenna points towards zenith,
when TET-1 is operated in Earth-pointing mode. Tis attitude
is therefore preferred for the operation of the navigation system,
because it maximizes the visibility of GPS satellites.
Restrictions of the battery capacity, however, require TET-1
satellite to be operated in Sun-pointing mode for recharging
when it is not in the Earths shadow. As a result, the POD anten-
nas boresight vector does not always point towards zenith, but
deviates signifcantly from the local zenith vector during parts
of the orbit when the batteries are charged. As a result, the
antennas feld of view will be obstructed by the Earth, which
limits the number of satellites available for tracking. We will
show, however, that the NOX payload can still provide robust
navigation solution most of the time.
Te second antenna for radio occultation is mounted on the
-x-panel. It points towards the Earths horizon in anti-fight
directions when the satellite is in Earth-pointing attitude mode.
Tis antenna orientation facilitates the tracking of GPS signals
through the Earths atmosphere for GPS radio occultation (RO)
measurements, which we will discuss in more detail later.
As only one antenna can be used at a time, the LEO orbit
and clock ofset determination during occultation experiments
is performed using measurements of the occultation antenna.
Tis is a simplifed concept compared to most modern RO mis-
sions, which use a dedicated navigation antenna in parallel to
one or more occultation antennas.
We should mention that the RO experiment of NOX is not
intended to routinely provide data for weather prediction or
climate research. Its purpose is merely to demonstrate the capa-
bilities and limitations of current COTS hardware, along the
lines of a similar experiment conducted on board the Micro-
Lab1 in 1995 and described in the article by R. Ware et alia.
Receiver Performance
In the following subsections, we analyze the in-fight perfor-
mance of the GPS receiver and the antenna system with respect
to activation behavior and signal tracking characteristics.
Where available, the on-orbit results are compared to pre-fight
tests with a signal simulator.
Receiver Start-Up Behavior. Unlike the GPS receiver on the
satellite bus, which is operated continuously during the entire
mission, the NOX receiver is only activated during dedicated
experiment time slots typically once a week. Te experiments
ON BOARD A LEO SATELLITE
FIGURE 1 Drawing of the TET-1 satellite without multilayer insulation
showing key elements of NOX and the orientation of the satellites
body-fxed coordinate system. The white plate above the NOX
housing is a heat radiator, which is part the of the satellites thermal
control system (image courtesy Kayser-Threde/Astro- und Feinwerk-
technik).
NOX POD antenna (not
visible, on -z-panel)
NOX housing with GPS
receiver and IF board
NOX occultation antenna
(on-x-panel)
+x
+y
+z
FIGURE 2 Schematic of the NOX payload on-board TET [7]
Switch line, Reset line
Tx/Rx (RS422) +16VDC ~ +40VDC
H
o
u
s
i
n
g
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
have a varying duration between 12 and 24 hours. Te receiver
is only supplied with power during these intervals and other-
wise is completely switched of.
Upon activation at the beginning of an experiments time
slot, the receiver starts to search for satellite signals. Te receiv-
ers non-volatile memory still contains the last valid position
solution, broadcast almanacs and navigation data obtained
before the last deactivation. In the case of the NOX receiver,
this information is used to compute a list of visible satellites,
which are prioritized in the signal search. If this warm start
acquisition fails because no satellite has been acquired afer
45 seconds, a sequential search over the full GPS constellation
is performed instead until a position fx could be successfully
computed.
Te receivers time to frst fx (TTFF) is an important per-
formance measure. Te histogram in Figure 3 shows the TTFF
statistics for 20 receiver activations using the POD antenna.
Tese data indicate that the receiver has achieved a frst posi-
tion fx afer less than two minutes in more than half of the
cases. Note that this includes the time necessary for receiver
boot and self-test.
The shortest and longest TTFF encountered in the 20
receiver activations are 85 seconds and 189 seconds, respec-
tively. Te mean and standard deviation of the TTFF is 2.03
0.50 minutes, which show good agreement with values obtained
from hardware-in-the-loop tests using a signal simulator. Te
TTFF for the simulated scenario yielded a mean and standard
deviation of the TTFF of 2.56 0.50 minutes, as described in
the article by J. Leyssens et alia. Te short TTFF clearly shows
the benefts of a smart signal-search concept and optimal use
of the 48 channels during acquisition.
FIGURE 3 Time to frst fx of the internal receiver navigation solution
after receiver activation. The diagram is based on 20 receiver starts
using the POD antenna.
Time to First Fix [min]
100%
80%
60%
40%
20%
0%
C
u
m
u
l
a
t
i
v
e
F
r
e
q
u
e
n
c
y
<1.0 <1.5 <2.0 <2.5 <3.0 <3.5
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
____________
Choose one for your high precision applications Choose one for your high precision applications
Tel: +86 21 6405 6796 | Email: sales@comnavtech.com | www.comnavtech.com
ComNav Technology Ltd.
K508
GPS L1/L2/L5
GLONASS L1/L2
BeiDou B1/B2/B3(support)
K500
GPS L1
GLONASS L1
BeiDou B1
K501G
GPS L1/L2
GLONASS L1/L2
K501
GPS L1/L2
BeiDou B1/B2
/B3(support)
It is interesting to note that the receiver has never acquired
a position fx within the frst 45 seconds of activation, meaning
that the frst fx has always been achieved afer a signal search
over the full constellation. Using almanac and navigation solu-
tion information from previous fxes during a warm start is
obviously not helpful due to the long interruptions between
consecutive activations of the NOX payload.
Tracking Performance. The number of simultaneously
tracked satellites is a key parameter to determine the avail-
ability of a navigation solution. Figure 4 shows the statistics
for the number of satellites tracked on L1 C/A-code and L2
P(Y)-code signals over the entire period of time when NOX was
activated using the POD antenna. Statistics for L1 P(Y)-code
are not displayed, because they are virtually identical with L2
P(Y) results.
Te fgure shows clearly that the receiver tracks between
8 and 12 satellites most of the time, which provides sufcient
redundancy for a robust computation of a navigation solution.
Tese results are consistent with results from signal-simulator
tests. Te peak for P(Y)-code is shifed slightly towards a lower
number of simultaneously tracked satellites compared to C/A-
code, which refects a lower sensitivity and deferred acquisition
of the semi-codeless tracking of P(Y)-code.
Te receiver tracks four or more satellites for 99.99 percent
of the time on C/A-code and 99.73 percent of the time on P(Y)-
code. We can thus conclude that the receiver has a high avail-
ability of single- and dual-frequency navigation solutions.
In order to assess the measurement quality of the NOX
receiver and antenna system, we analyzed the carrier-to-noise-
density ratio (C/N
0
). Figure 5 depicts a polar plot of the carri-
er-to-noise-density ratio (C/N
0
) for C/A-code measurements
based on 24 hours of data recorded on August 30, 2012, using
the POD antenna. Te coordinate axes in Figure 5 are aligned
with the local frame of the antenna, which exhibits a diferent
orientation than the satellites body-fxed coordinate axes.
Te measured C/N
0
ranges from 30 dB-Hz near the hori-
zon to approximately 50 dB-Hz at higher elevation angles. Te
C/N
0
pattern is not rotationally symmetric, but exhibits a clear
azimuthal dependency. Tis efect is especially pronounced on
the lef side of the diagram and results most likely from the
mounting position of the antenna close to the edge of the panel.
Without a choke-ring or a dedicated antenna ground plane,
the non-uniform satellite structure afects the gain pattern of
the antenna.
Accuracy of Navigation Solution. Te NOX receiver computes
a navigation solution based on dual-frequency pseudorange
observations and outputs results at intervals of 30 seconds.
Te receivers internal position flter has been turned of in the
NOX experiment; thus, the reported positions correspond to
independent epoch-by-epoch navigation solutions.
Te errors of the receiver navigation solution are assessed
by a comparison with a precise reference trajectory from a
reduced-dynamic orbit determination based on carrier-phase
measurements. In the next section we will provide more details
on how the reference trajectory has been obtained.
For our analysis, we selected navigation solution results
from a period of almost 24 hours on August 30, 2012. Figure 6
shows the errors in radial, tangential (in the direction of fight),
and normal directions with respect to the orbital coordinate
frame with the corresponding statistics listed in Table 1.
It becomes obvious that the errors of the radial component
exhibit the largest scatter, which is an expected result, because
the vertical component is always most afected by the largest
dilution-of-precision in a single-point solution.
FIGURE 5 Polar plot of carrier-to-noise-density ratio for C/A code
measured with the POD antenna.
-y (Az=180)
Carrier-to-Noise Density Ratio [dB-Hz]
-
x

(
A
z
=
2
7
0
)
+
x

(
A
z
=
9
0
)
30 32 34 36 38 40 42 44 46 48 50
+y (Az=0)
Antenna Frame
FIGURE 4 Simultaneously tracked satellites on the L1 C/A and the L2
P(Y) signal using the POD antenna.
Number of Satellites
25%
20%
15%
10%
5%
0%
F
r
e
q
u
e
n
c
y
2 4 6 8 10 12 14 1 3 5 7 9 11 13 15
C/A L1
P(Y) L2
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
For the majority of the epochs, the radial errors fall between
5.0 meters, whereas the tangential and normal components
are typically less than 2.5 meters. At one epoch, though, the
radial error reaches 13.81 meters.
When interpreting these results, it is important to note
that the reference solution refers to the satellites center of mass
whereas the receivers navigation solution refers to the antenna
phase center. Te ofset between these two points is on the order
of 0.66 meter and is projected diferently on the radial-, tan-
gential-, and normal-coordinates, depending on the satellite
attitude, and afects the error statistics.
Nevertheless, we can conclude that the receiver typically
provides navigation solutions with meter-level accuracy. Te
satellite has alternately been operated in Earth-pointing mode
during eclipse and otherwise in Sun-pointing mode. Despite
the possible obstruction of the antennas feld of view in Sun-
pointing attitude mode, the receiver has provided continuous
navigation solutions for the entire time due to its high number
of tracking channels and fast acquisition of satellites.
POD Performance
In this section, we will present the results of precise orbit deter-
mination of NOX. Here we will briefly introduce the orbit
determination process. As no reference solution is available for
the satellites orbit, overlap comparisons serve as a metric to
assess accuracy of the POD results. A phase-center pattern for
the NOX POD antenna computed from carrier-phase residuals
is also presented.
Reduced Dynamic Orbit Determination. Precise orbit solutions
have been computed using measurements from the satellites
POD antenna at an update rate of 30 seconds. As a frst a priori
trajectory, a single point solution based only on pseudorange
measurements is computed. Tis coarse trajectory is smoothed
using a least-squares filter to fit the satellite positions to a
dynamic orbit model. Tis smoothed orbit is then used as an a
priori orbit for a reduced dynamic orbit determination, where
pseudorange and carrier-phase measurements are processed in
a least-squares flter with a dynamical orbit model.
Te estimation parameter vector comprises the satellite
position and velocity state at the reference epoch, a scaling
factor for the accelerations due to solar radiation pressure
and atmospheric drag, as well as the ionosphere-free carrier-
phase foat ambiguities. Further, empirical accelerations in
radial, along-track, and cross-track direction are estimated to
compensate for defciencies in the deter-
ministic model. A more detailed descrip-
tion of the POD procedure and the orbit
model can be found in the article by O.
Montenbruck et alia cited in Additional
Resources.
As a more precise reference solution
or independent measurements from satel-
lite laser ranging are not available, a direct
assessment of the errors in the precise orbits
is not possible. Terefore, orbit overlap comparisons are used
here to yield at least an indication of the orbit quality. For this
purpose, we computed 19 orbit solutions based on a data arc of
fve hours on August 30, 2012. Te frst hour of each orbit solu-
tion with the central hour of a previous orbit solutions, starting
two hours earlier.
Te mean values of the pseudorange residuals are consis-
tently between 70and 75centimeters for all 19 POD runs. Te
carrier-phase residuals are two orders of magnitude smaller
and vary between 7.5millimeters and 8.5millimeters. Figure 7
presents the results for the 3-D RMS overlap errors. Te maxi-
mum and minimum errors are 35millimeters and 5millime-
ters, respectively, with an average of 18millimeters.
FIGURE 6 Accuracy of NOX navigation solution compared to a pre-
cise reference orbit for August 30, 2012.
Time [hh:mm]
15
10
5
0
-5
-10
P
o
s
i
t
i
o
n

E
r
r
o
r

[
m
]
00:00 04:00 08:00 12:00 16:00 20:00 24:00
normal
tangential
radial
mean [m] STD [m] rms [m] max [m] min [m]
radial 0.27 1.45 1.48 13.81 -6.16
along-tr. 0.12 0.67 0.68 4.15 -5.08
cross-tr. 0.52 0.56 0.77 3.00 -2.53
3-D 1.54 0.93 1.80 14.11 0.05
TABLE 1. Statistics of errors of the PolaRx2 navigation solution for a 24h data set on August
30, 2012. Listed in the table are the mean and standard-deviation, the rms error, maximum
and minimum errors
FIGURE 7 Statistics for POD RMS overlap errors in radial, along-track
and cross-track direction based on 19 orbit solutions with a data arc
5 h on August 30, 2012. The frst hour of an orbit has been compared
to the central hour of an orbit solution, which starts two hours
earlier.
Orbit Number [-]
40
35
30
25
20
15
10
5
0
P
M
S
P
o
s
i
t
i
o
n

E
r
r
o
r

[
m
]
0 2 4 6 8 10 12 14 16 1 3 5 7 9 11 13 15 17
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Experience shows that the inclusion of empirical accelera-
tions in the estimation leads to a reduced stifness of the solu-
tion, which allows the estimated trajectory to closely follow
the observations. As a result, the overlap comparisons tend to
be too optimistic and the true orbit errors can be expected to
be larger. Based on experience of previous missions with dual-
frequency GPS receivers, we would expect the achievable 3-D
RMS accuracy to be on the order of decimeters or better.
Antenna Phase Pattern. Inspection of the carrier-phase
residuals from the reduced dynamic orbit determination has
revealed clear systematic efects with azimuth- and elevation-
dependent variations. Tese systematic residuals are due to
antenna phase pattern variations of the receiving antenna
caused by the efect of the satellites structure on the antenna
in the absence of a choke ring or ground plane.
An antenna phase-center variation pattern can be derived
based on carrier-phase observations processed over a longer
time interval. For this purpose, we grouped the residuals of
the POD into azimuth and elevation bins depending on the
direction of the received signal. Te phase center variation cor-
rection is then computed as the average of the residual in this
bin. Te resulting phase pattern correction is then used again
for a POD and refned with corrections based on the residuals
of further iterations.
Figure 8 depicts the results for the POD antenna based on
the data of all NOX activations between August and December,
2012. Te maximum amplitude of phase center variations is
25 millimeters. Te phase pattern exhibits an irregular shape
with rapid changes between maximum and minimum varia-
tions. Te standard deviation of the carrier-phase residuals in
the POD can be reduced from 12 millimeters to 8 millimeters
using this correction pattern.
Radio Occultations
During a radio occultation (RO), the signal of a GNSS satellite
is tracked by a receiver on the opposite side of the Earth close to
the horizon. Because the signal is received through the Earths
atmosphere, it is afected by delays and bending depending on
the refractivity of the atmospheric layer.
Te refractivity can be approximated as a function of total-
electron content in the ionosphere as well as temperature, pres-
sure, and humidity in the troposphere. We can compute the
bending angle and the corresponding ray height of the signal
from carrier-phase measurements, which
allows us to retrieve the refractivity index
and solve for atmospheric and ionospheric
parameters.
Due to the change in geometry between
the two satellites, the signal is received
through diferent layers of the atmosphere
during an occultation event. If high rate
carrier-phase measurements are available,
bending angle profiles for different alti-
tudes can be recorded. Te derived atmo-
spheric parameters serve as input data for
weather prediction and climate research,
which are the main motivations for radio
occultations.
Figure 9 presents a schematic of a radio
occultation. Te GNSS and the LEO satel-
lite travel with velocities of v
s
and v
r
, respec-
tively. Te direct straight-line connection
FIGURE 9 Schematic of occultation measurements between a GPS and a LEO satellite, which
travel with velocities v
s
and v
r
, respectively. The signal is bent by an angle due to the iono-
sphere and troposphere at the impact height a.
SLTA
FIGURE 8 Polar plot of carrier-phase pattern of the NOX POD antenna
iteratively computed from carrier-phase residuals from the reduced
dynamic orbit determination.
-y (Az=180)
Phase Center Variation [mm]
-
x

(
A
z
=
2
7
0
)
+
x

(
A
z
=
9
0
)
-25 -20 -15 -10 -5 0 5 10 15 20 25
+y (Az=0)
Antenna Frame
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
between the LEO satellite and the GNSS satellite determines
the straight-line tangent altitude (SLTA), which is the distance
of this line to the surface of the Earth.
Due to the bending, the actual signal path does not follow
this straight-line connection, but is curved around the Earth
at an impact height a. Tis curvature is characterized by the
bending angle at the corresponding impact height. Due to the
signal bending, the receiver can still track the GNSS satellite,
even though it may be below the Earths horizon as seen from
the LEO satellite. However, note that the bending depicted in
the plot is highly exaggerated and does in reality not exceed
one degree.
Te fundamental relation to be solved for in RO processing
is the dependency of the bending angle on the impact height.
For this purpose, the carrier-phase measurements are cor-
rected for geometrical range between LEO and GNSS satellite,
receiver and GNSS satellite clock ofsets, and relativistic efects,
which only leaves the delays due to ionosphere and troposphere.
Tis residual is referred to as the excess phase delay. Te LEO
orbit and the clock ofset corrections are obtained from a POD
using the occultation antenna. Te bending angle can then be
retrieved from the change in excess phase, referred to as the
excess Doppler, and the LEO and GNSS satellite velocity.
Figure 10 shows results for an occultation event of GPS satel-
lite SVN 47 (PRN 22) on September 2, 2013, between approxi-
mately 17:42 and 17:48 UTC. Te tangent point coordinates of
the occultation are 28.18S and 82.81E, which correspond to
a location in the Indian ocean, halfway between Madagascar
and Western Australia. GPS measurements have been taken
with a data rate of fve hertz.
Te top plot in Figure 10 shows the signal-to-noise density
ratio (C/N
0
) for L1-C/A code and L2-P(Y) code together with
the straight-line tangent altitude. Te plot starts at an SLTA
of approximately 500 kilometers, where the signal of the GPS
satellite is tracked through the upper ionosphere. Te C/N
0
is
practically constant at 46.6 dB-Hz and 33.3 dB-Hz for C/A and
L2-P(Y), respectively, for almost the entire period of time. Only
during the last 30 seconds of tracking at an SLTA of 50 kilome-
ters and less, the C/N
0
starts to drop and exhibit larger varia-
tions due to the signal attenuation in the lower atmosphere.
Also note that the L2-P(Y) tracking is disrupted earlier than
L1 C/A.
The bottom of Figure 10 plot shows the corresponding
excess carrier-phase for L1 and L2 together with the slant ion-
ospheric delay computed from dual-frequency carrier-phase
measurements. Te slant ionospheric delay for the L1 frequency
I
L1
has been computed from
In this equation, and are the L1 and L2 frequency,
respectively, and and are the carrier-phase measure-
ment in units of length. This geometry-free carrier-phase
combination removes all frequency-independent terms such
as geometry, clock ofsets, and tropospheric delay. Te delay
has the positive sign convention of the pseudorange delay,
even though it has been computed from carrier-phase mea-
surements. Note that the ambiguities and frequency-dependent
signal delays do not cancel out in Equation 1. Terefore, the
absolute value of the ionospheric delay cannot be recovered,
but only the temporal variation.
As expected, the ionospheric delay is smallest at high alti-
tudes, where the electron content is low. As the signal path
proceeds into the lower ionosphere, the delay increases and
reaches a peak at a straight-line altitude of approximately 300
kilometers. For lower tangent altitudes, the delay decreases
again. Te maximum amplitude of the ionospheric delay varia-
tion over the data arc is about fve meters, or ~30 total electron
content (TEC) units. Since the measurements were taken on the
dark side of the Earth, the ionospheric electron content is low.
Te excess phases for L1 and L2 depicted in the bottom plot
of Figure 10 show a maximum amplitude of ~240 meters for a
straight line tangent altitude of -20 kilometers, where the signal
is tracked through the lower part of the atmosphere. At higher
altitudes, only small variations of the excess phases due to the
ionospheric delays are present. As a result, the bending angle
FIGURE 10 Measurements of an occultation event for GPS satellite SVN 47 (PRN 22) on September 2, 2013. The top plot depicts the carrier-to-
noise density ratio for L1 and L2 measurements as well as the straight-line tangent altitude (SLTA). The bottom plot shows the excess carrier-
phase measurements for L1 and L2, as well as the ionospheric delay for the L1 frequency.
50
40
30
20
10
0
400
300
200
100
0
4.0
3.0
2.0
1.0
0.0
-1.0
C
/
N
0

[
d
B
-
H
z
]
0.20
0.16
0.12
0.08
0.02
E
x
c
e
s
s

p
h
a
s
e

[
k
m
]
S
L
T
A

[
k
m
]
L
1

I
o
n
o

[
m
]
42m 43m 44m 45m 46m 47m
2013/09/02-17:42
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
is small at high altitudes and increases rapidly as soon as the
signal path enters the troposphere.
If the excess phases of Figure 10 are converted into excess
Dopplers, the bending angle of the signal can be computed as a
function of the impact height. Figure 11 depicts the correspond-
ing results for the L1 and L2 frequency. Te plot shows that the
bending angle for the L1 and L2 frequency difer due to the
frequency-dependent ionospheric delays. In order to remove
this efect, an ionosphere-free (or neutral) bending angle (LC)
has been computed as a linear combination of the L1 and L2
bending angles at the same impact height.
The minimum impact height for this occultation is
approximately 7 kilometers. Te bending angle at this height
is approximately 0.6 degree. For higher altitudes, the bending
angle decreases until it reaches a level of about 0.001 degree for
impact heights larger than 50 kilometers. Te bending angles
exhibits only very low noise for impact heights lower than 40
kilometers.
At higher altitudes, the noise increases signifcantly. Te
reason for the diferent noise levels becomes clear from the plot
of the excess phases in Figure 10. For high altitudes, the signal
delays are small, and measurement noise and model imper-
fections dominate. Apparently, bending angles less than 0.001
degree cannot be observed, because they are below the noise
foor in the current processing. When the signal crosses the
troposphere, the delays grow quickly with decreasing impact
height and become distinguishable from the noise.
Summary and Conclusions
Tis article as presented initial fight results of the Naviga-
tion and Occultation eXperiment on-board the small satellite
TET-1. Te experiment has demonstrated that commercial-of
the-shelf hardware can be used in space-borne applications,
with only minor changes to the receivers frmware.
With the height and velocity constraints removed and an
increased Doppler search window, the receiver has reliably
acquired and tracked sufcient satellites for a continuous navi-
gation solution, typically within less than three minutes. Te
3-D RMS errors of the navigation solutions are on the order of
a few meters only. No latch-ups or other receiver failures have
been observed during the entire mission.
Plots of the C/N
0
variation and the phase pattern variations
in the antenna diagram indicate an efect of the satellites struc-
ture on the antenna characteristics. If a ground plane or choke
ring is used to mount the antenna, these efects can be expected
to be less pronounced.
In the absence of a more precise reference solution or inde-
pendent measurements, for example from satellite laser rang-
ing, the precise orbit determination accuracy cannot be directly
assessed. Orbit overlap comparisons have shown errors of a few
centimeters, between the frst hour and the central hour a two
fve-hour orbit arcs. Te use of an empirically derived correction
pattern for phase center variation could reduce the carrier-phase
residuals of the POD, typically from 12 millimeters to 8 mil-
limeters.
Radio occultation experiments have shown that dual-
frequency carrier-phase signals can be tracked through the
Earths troposphere with a data of fve hertz. Te data enables
researchers to monitor the ionospheric delay and derive slant
TEC variations of the upper atmosphere. A bending angle pro-
fle for L1 and L2 carrier-phase measurements has been derived
in the troposphere down to an impact height of about seven
kilometers.
Te NOX experiment proves that a low-cost GPS system,
which fulflls the requirements for precise orbit determination,
can be realized with COTS hardware. Tis approach may be
appealing for research groups seeking to gain inexpensive
access to relevant data and even help to identify possibilities
for cost reduction in future satellite missions.
Analysis of NOX RO observations has demonstrated that
GPS signals could be tracked through the ionosphere and tro-
posphere below the satellites horizon. We must note, however,
that the performance of this setup not sufcient to produce
RO data ready to be used for use in weather forecast or climate
research. Several special modifcations such as open-loop
tracking, an autonomous occultation prediction and channel
allocation algorithm in the receiver, and a higher sampling rate
as well as a high-sensitivity antenna system would be needed
to make this system competitive to modern RO payloads.
Acknowledgments
Te authors would like to acknowledge the contributions of
FIGURE 11 Bending angle profle for an occultation event for GPS
satellite SVN 47 (PRN 22) on September 2, 2013. Depicted are the L1
and L2 bending angles and a ionosphere-free combination (LC) of
bending angles.
Bending angle [deg]
80
70
60
50
40
30
20
10
0
I
m
p
a
c
t

h
e
i
g
h
t

[
k
m
]
1e-5 1e-4 1e-3 1e-2 1e-1 1e0 1e1
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
their former colleagues Ccile Renaudie
and Miquel Garcia-Fernandez, who have
worked on the design and manufactur-
ing of the NOX experiment. Te support
of colleagues at Kayser-Trede GmbH
during the design, implementation, and
operation of NOX payload is greatly
appreciated. Te team at Septentrio is
acknowledged for technical support and
discussions.
Te authors would also like to thank
the TET operations team at the Ger-
man Space Operations Center, especially
Andreas Sprl, Andreas Pohl, and Jens
Richter, for their help in planning and
implementing the NOX experiments.
DLR Space Administration is acknowl-
edged for the free fight opportunity of
the NOX experiment on-board TET-1.
This article is based primarily on
a paper presented at the ION GNSS+
2013 conference in Nashville, Tennes-
see, USA.
Manufacturers
The NOX payload incorporates a
PolaRx2 GPS receiver from Septentrio
nv, Leuven, Belgium; two S67-1575-14
L1/L2 passive patch antennas from Sen-
sor Systems, Inc., Chatsworth, Califor-
nia, USA; and a Spectrum Microwave
LNA 310-025105-011 low noise amplifer
from API Technologies Corporation,
Orlando, Florida, USA. An STR4760 sig-
nal simulator from Spirent Communi-
cations, plc, Paignton, United Kingdom,
has been used during pre-fight testing
and qualification of the receiver. The
TET-1 satellite bus navigation system
uses a Phoenix-S receiver by from DLR.
[1] Eckert, S., and S. Ritzman, J. Eckler, and W.
Brwald, On-Orbit Verication with a Technology
Test Carrier TET, in Proceedings of 6th IAA Sym-
posium on Small Satellites for Earth Observation,
Berlin, Germany, April, 2326, 2007
[2] Fckersperger, S. and G. Staton, and M. Turk,
Future Small Satellite EO Missions Based on TET,
in Proceedings of the Small Satellites Systems
and Services Symposium 2012, Portoro, Slove-
nia, June 48, 2012
[3] Garcia-Fernandez, M., and O. Montenbruck,
M. Markgraf, and J. Leyssens, Affordable Dual-
frequency GPS in Space, in Proceedings of the
6th International ESA Conference on Guidance,
Navigation and Control Systems, Loutraki, Greece,
October 1720, 2005
[4] Gleason, S., and D. Gebre-Egzabher, GNSS:
Applications and Methods, Artech House, Nor-
wood, Massachusetts, USA, 2004
[5] Hajj, G.A., and E. R. Kursinski, L. J. Romans, W.
I. Bertiger, and S. S. Leroy, A technical descrip-
tion of atmospheric sounding by GPS occultation,
Journal of Atmospheric and Solar-Terrestrial
Physics, 64:451469, 2002. doi: 1364-6826/02/
[6] Kursinski, E. R., and G. A. Hajj, J. T. Schoeld,
R. P. Lineld, and K. R. Hardy, Observing Earths
atmosphere with radio occultation measurements
using the Global Positioning System, Journal of
Geophysical Research, 102(D19):23,42923,465,
1997. doi: 0148-0227/97/97 JD-01569
[7] Lemke, N. M. K., and C. Kaiser, S. Fcker-
sperger, G. Staton, and T. Stuffler, TET-Based
Small Satellite Family, in Proceedings of the 63rd
International Astronautical Congress, Naples,
Italy, October 15, 2012
[8] Leyssens, J., and and M. Markgraf, Evaluation
of a Commercial-Off-The-Shelf Dual-Frequency
GPS Receiver for Use on LEO Satellites, in Pro-
ceedings of the ION GNSS, Long Beach, California,
USA, September 1316, 2005
[9] Markgraf, M., and C. Renaudie, and O. Mon-
tenbruck, The NOX Payload-Flight Validation of a
Low-Cost Dual-Frequency GPS Receiver for Micro-
and Nanosatellite Applications, in Proceedings
of the IAA Symposium on Small Satellite Systems
and Services (4S), Rhodes, Greece, May 2630,
200.
[10] Markgraf, M., and P. Swatschina, The Navi-
gation and Occulation eXperiment (NOX) onboard
TET-1, presented at 2nd TET Customer Day, Kay-
ser-Threde, Munich, Germany, July 5, 2010
[11] Melbourne, W.G., and E. S. Davis, C. B. Dun-
can, G. A. Hajj, K. R. Hardy, E. R. Kursinski, T. K.
Meehan, L. E. Young, and T. P. Yunck, The Appli-
cation of Spaceborne GPS to Atmospheric Limb
Sounding and Global Change Monitoring, JPL
Publication 94-18, 1994
[12] Montenbruck, O.,l and T. van Helleputte,
R. Kroes, and E. Gill, Reduced dynamic orbit
determination using GPS code and carrier
measurements, Aerospace Science and Tech-
nology, 9(3):261271, 2005. DOI 10.1016/j.
ast.2005.01.003
[13] Ware, R., and M. Exner, D. Feng, M. Gorbunov,
K. Hardy, B. Herman, Y. Kuo, T. Meehan, W. Mel-
bourne, C. Rocken, W. Schreiner, S. Sokolovskiy,
F. Solheim, X. Zou, R. Anthes, S. Businger, and K.
Trenberth, GPS Sounding of the Atmosphere from
Low Earth Orbit: Preliminary Results, Bulletin of
the American Meteorological Society, 77, 1940.
DOI 10.1175/1520-0477(1996)077<0019:GSOTAF
>2.0.CO;2
[14] Yoon, Z., and T. Terzibaschian, C. Raschke, and
O. Maibaum, Robust and Fault Tolerant AOCS of
the TET Satellite, in Proceedings of the 7th IAA
Symposium on Small Satellites for Earth Observa-
tion, Berlin, Germany, May 47, 2009
Authors
Andr Hauschild is a mem-
ber of the scientic staff of
the GPS Technology and
Navigation Group at DLRs
German Space Operations
Center (GSOC). His eld of
work focuses on real-time
precise clock estimation for GNSS satellites as well
as multi-GNSS processing using modernized GPS
and new satellite navigation systems. He is also
involved in projects with space-borne GNSS
receivers for scientic applications like precise
orbit determination and radio occultation. Andr
graduated in aerospace engineering from Tech-
nische Universitt Braunschweig, Germany, and
received his Dr.-Ing. from the Technische Univer-
sitt Mnchen, Germany.
Markus Markgraf is a senior
research engineer in the
GNSS Technology and
Navigation Group at DLR/
GSOC. He started working
at DLR in 2000 after grad-
uating as a Dipl.-Ing. (FH)
for electrical engineering and communication
technology. His current research activities com-
prise GNSS receiver technology for satellites and
sounding rockets, scientic applications of GNSS,
and mission support and analysis. He was the key
engineer for the Navigation and Occultation
Experiment (NOX) on TET-1 during the design and
implementation phase.
Oliver Montenbruck is head
of the GNSS Technology
and Navigation Group at
DLRs German Space Oper-
ations Center (GSOC),
Oberpfaffenhofen. His cur-
rent research activities
comprise spaceborne GNSS receiver technology,
autonomous navigation systems, spacecraft for-
mation ying, and precise orbit determination as
well as new constellations and multi-GNSS pro-
cessing. Dr. Montenbruck presently chairs the
GNSS Working Group of the International GPS Ser-
vice and coordinates the performance of the MGEX
Multi-GNSS Experiment. He has authored numer-
ous technical papers and various textbooks.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
I
n recent years, numerous, relatively
inexpensive hardware platforms for
conducting scientifc research using
the sofware defned radio (SDR) para-
digm have become commercially avail-
able. The Manufacturers section near
the end of this article lists examples of
several of these. In turn, this has spurred
universities and research groups around
the world to adopt this technology for
advanced GNSS signals-based research
and development.
Popular research topics exploit-
ing GNSS SDR receivers include first
look GNSS signal capture and analy-
sis, interference/spoofng detection and
mitigation, GNSS signal authentication
by means of nominally present satellite
signal distortions (i.e. signal fngerprint-
ing), signal quality and deformation
monitoring, GNSS bi-static radar and
synthetic aperture radar (SAR)-based
imaging, multi-platform combined GNSS
signal processing, advanced GNSS mul-
tipath mitigation, multi-element phased
array processing, ultra-tight integration
of GNSS with multiple sensors, vector
tracking loops and other holistic and
open loop signal tracking approaches,
ionospheric research using multi-fre-
quency GNSS observables, and general
multi-constellation/multi-frequency
GNSS receiver development, prototyping,
testing and algorithm validation.
Te general approach to carrying out
such research involves one or more data
collection campaigns followed by mul-
tiple cycles of algorithm development,
sampled data processing, and analysis.
Realtime processing capability is gener-
ally not required at this stage of devel-
opment. However, to achieve maximum
productivity researchers fnd it highly
desirable to have fexibility in algorithm
development by way of high-level pro-
gramming languages and robust user-
friendly development environments
with extensive built-in math library sup-
port and data visualization capabilities.
Arguably, within the satellite navigation
community, MATLAB has become the
de facto standard in this regard.
Many of t he afore-ment ioned
research topics can involve sampled sig-
nal data collection at wide bandwidths,
high dynamic range, and multiple coher-
ently sampled streams. For example,
consider a wideband GNSS data collec-
tion campaign for investigating phased
array based interference mitigation tech-
niques using a seven-element, controlled
reception pattern antenna (CRPA). In
this case, bandwidth, dynamic range,
and multiple channels are all in play.
Assuming typical front-end hard-
ware specifications for such an appli-
cation of 60 megasamples per second,
14-bit samples (extended to two bytes
for data transfer) and eight channels
(one channel being a separate reference
antenna), the data capture rate equals
960 Mbytes/second. Even with lesser
requirements, it is not uncommon to
return from a collection campaign with
Use of the software dened
radio paradigm for GNSS
receiver design and associated
research are proliferating
rapidly as computer processing
power increases and costs
decline. However, diverse
approaches to the software
employed for the high-level
development environments
of these designs limit the
cross-platform utility and full
exploitation of their computing
platforms. The article describes
the latest version of a
universal GNSS SDR processing
toolbox that is distributed
as a plug-in for high-level
algorithm development.
SANJEEV GUNAWARDENA
RECEIVER TOOLBOX
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
www.trimbledimensions.com trimble_dimensions@trimble.com
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
______________________________
multiple hundreds of gigabytes (if not
terabytes) of data. Thus, a solution is
needed to process such large datasets in
a reasonable amount of time.
Todays general-purpose desktop and
laptop computers provide tremendous
numerical computation capability at
low power and cost. Tis is made pos-
sible with multiple processor cores
each clocking at multiple gigahertz and
supporting wide single-instruction-mul-
tiple-data (SIMD) instructions. In addi-
tion, todays afordable consumer-grade
solid-state drives feature sustained read
speeds on the order of 500 megabytes/
second. Hence, these machines are good
candidates for crunching through large
amounts of SDR data. For further dis-
cussion about computational workloads,
see the sidebar, Cost/Beneft Justifca-
tion for a GNSS SDR Toolbox.
Unfortunately, the layers of sofware
abstraction built into high-level devel-
opment environments to facilitate user-
friendly coding is one of the main rea-
sons why, in general, these tools cannot
take full advantage of the computation
capabilities of the platforms they run
on. Thankfully, all such tools support
extensions to allow users to integrate
their own custom libraries written in
low-level code. In MATLAB, this exten-
sion framework is known as MATLAB
executable (MEX).
The goal of the work reported in
this article is to develop a truly uni-
versal GNSS SDR processing toolbox
for education and research that could
be distributed in the form of a plug-in
for high-level algorithm development
platforms specifcally MATLAB. Te
following high-level features were envi-
sioned for the toolbox:
- suppoits the ability to peifoim a wide
range of cutting-edge GNSS signals-
based research topics as described
previously
- suppoits most SDR data fle foimats
and front-end frequency plans
- suppoits all cuiient and emeiging
GNSS signal structures and other
signals of opportunity
- has a usei inteiface and confguiation
methodology that is easy to learn and
apply to the various research topics
described here
- piovides as many open-souice func-
tional examples as possible, thus
shortening the learning curve for both
beginners as well as advanced users.
The work described in this article
achieves, to a large extent, all of these
objectives and, more importantly, builds
the framework for the baseband signal-
processing layer of a truly universal
GNSS SDR architecture. The toolbox
has been used successfully to process
the following open GNSS signals using
live data: GPS L1 C/A, GPS L2C, GPS
L5, GLONASS FDMA signals on L1
and L2, Galileo E1 CBOC signals using
BOC(1,1), BOC(6,1) and CBOC(6,1,1/11)
processing; Galileo E5a and E5b, BeiDou
B1, satellite-based augmentation systems
(WAAS and EGNOS), and WAAS sig-
nals on L5.
Te sofware is currently distributed
as a MATLAB toolbox and can be down-
loaded free of charge for education and
research use.
One important note: this toolbox
is not a complete GNSS receiver in the
sense that it does not output position,
navigation, and time (PNT) solutions.
However, the processed-signal outputs
(available at a one-kilohertz rate) contain
all the information needed for subse-
quent processing of PNT solutions.
Supporting Multiple GNSS
SDR File Formats
Most SDR data collection systems store
their IF-sampled or baseband-sampled
data in binary format. For uninterrupt-
ed collections over prolonged intervals,
data are sometimes written to multiple
small fles because such a strategy allows
fles to be managed more efectively than
one fle written to a large-capacity vol-
ume. For systems that collect SDR data
continuously for the purpose of record-
ing rare anomalous signal events, this
multi-file collection strategy allows
older fles to be deleted to make space
for new ones, thus extending the avail-
ability of past history to the size of the
storage array in contrast to the capacity
of a memory-based bufer.
In some systems, the GNSS samples
may be interlaced with binary data
from other sensors such as IMUs, laser
scanners or cameras to achieve inher-
ent time synchronization between
these sensors. In this case, additional
metadata information is needed to
extract GNSS samples from the fle and,
Cost/Benet Justication for a GNSS SDR Toolbox
For GNSS SDR, the most numerically intensive computations involve correlation
of hundreds of millions of signed integer samples for each second of processing.
However, these samples are typically less than one byte. Trough some straight-
forward pre-processing steps to reduce dynamic range, the result of correlation
over a one-millisecond interval can usually be made to ft within 16-bit signed
integers with negligible loss of performance.
Hence, these structurally regular fxed-point computations can be parallelized
by factors of 8 or 16 using 128-bit and 256-bit wide Streaming SIMD Extensions
(SSE) or Advanced Vector Extensions (AVX), respectively. (AVX has been sup-
ported in all x86 processors shipping since 2011.) Further parallelization over the
available number of logical processors (up to 8 in most consumer PCs) can yield
up to 128 theoretical performance improvement compared to un-optimized
code.
Such optimizations require the correlation algorithm to be partitioned so that
subsets of the computations can be performed independently in each processor.
Tis type of fne-grained architecting of an algorithm to exploit the feature set
of a particular generation of processors to the maximum extent possible is best
done by human programmers as opposed to optimizing compilers.
Because sample correlation is such a critical component of any GNSS SDR
and the algorithm essentially does not change signifcantly with sampling rate
or GNSS signal structure, the cost of low-level optimization can be justifed by
considering the subsequent time savings that can be gained. Tis is especially true
if the correlation engine can be architected such that it supports a wide range of
applications and use cases.
RECEIVER TOOLBOX
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
optionally, decode the sensor data.
Currently no standard exists within
the PNT community that allows GNSS
SDRs to work seamlessly with fles writ-
ten by any SDR data-collection system.
Tis means that the user is forced to set
data decoding parameters in an ad hoc
manner. When fles from a diferent sys-
tem are sourced, these parameters and
the decoder must be changed manu-
ally a process that is prone to human
error.
Part of the effort described in this
article aims to address this issue so that
fles from any data collection system can
be seamlessly integrated into any GNSS
SDR processing platform. Te proposed
solution is to pair a metadata fle with
each binary data fle. Te metadata fle
includes all the information needed to
integrate the SDR fle into the processor
and decode its contents.
As the format for the metadata fle,
eXtensible Markup Language (XML)
provides a desirable option. All oper-
ating systems and application develop-
ment suites support XML, which is a
low-overhead human-readable format,
thus providing a straightforward process
to integrate it into any data collection
system. Figure 1 shows an example of
an SDR metadata fle written in XML. It
contains all of the necessary information
to decode the multi-stream samples cor-
rectly as well as other information per-
taining to the data collection campaign.
The SDR toolbox uses this meta-
data mechanism to open and decode
SDR data fles from many data collec-
tion systems. When opening a specifed
SDR fle, the reader automatically parses
the XML fle and imports the metadata
into the MATLAB workspace as a struc-
ture. For multiple fles, the user specifes
the name of the frst fle along with the
maximum number of one-millisecond
blocks to be processed. Tis information
is used to automatically fnd and splice
the necessary fles to fulfll the request.
Supporting a Wide Range
of Research Applications
In broad terms, GNSS baseband signal
processing can be divided into three
stages. The following sections sum-
marize the features required in each of
these stages to support a wide range of
research applications.
Pre-Correlation Processing. As is well
known, correlation losses become neg-
ligible for sample quantizations beyond
two bits. However, this does not hold
true in the presence of interference. In
this case, we can use additional dynamic
range to perform interference reduction
processing prior to correlation. Typical
pre-correlation processing includes sam-
ple covariance computation (for interfer-
ence detection and location) and digital
fltering and excision techniques applied
in the time and/or frequency domains.
Te various types of pre-correlation
processing that a researcher may want to
apply to a GNSS processing application
could be supported by including 1) an
optimized sample statistics processor, 2)
a sample masking processor for blank-
ing interference-dominated samples
from being correlated, 3) a configu-
rable time-domain filter implementa-
tion (such as Direct-Form II), and 4) a
fast Fourier transform (FFT) engine for
implementing frequency-domain inter-
ference detection and excision tech-
niques.
These processing blocks could be
integrated into the sample streams using
a sofware plug-in interface. Since imple-
mentations already exist in MATLAB,
developing a fully featured pre-correla-
tion processor was considered a lower
priority compared to the correlation
engine. However, Version 3 of the tool-
box does include a sample statistics and
noise processor as described in below.
Correlation Processing. Three fun-
damental techniques exist for sample
correlation: time-domain correlation,
parallel frequency correlation, and par-
allel code correlation. The latter two
methods provide a large number of cor-
relation outputs corresponding to Dop-
pler frequency offsets or code phases,
respectively.
The limited resolution of parallel
correlation algorithms and the inability
to steer the local replicas that produce
them with adequate precision (par-
ticularly with respect to code phase)
preclude their use in precision signal
tracking applications. Te parallel code
correlation algorithm is most efcient
when researchers need a large swath
of code correlation space observability
such as during signal acquisition. Other
uses include correlation space monitor-
ing (also known as delay-Doppler map
monitoring) for applications such as
spoofer detection.
In any case, a low update rate on the
order of one to several seconds is typi-
cally sufcient for monitoring applica-
FIGURE 1 Proposed GNSS SDR metadata XML schema
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
tions. As MATLAB already contains optimized FFT implemen-
tations to write parallel correlation algorithms, no attempt was
made in this version of the toolbox to accelerate FFT-based
parallel correlators.
Many of the GNSS SDR research applications described here
require several more time-domain correlators than the typical
two to fve needed for traditional signal tracking. Supporting
a given processing algorithm for all current and future GNSS
signals can become cumbersome due to their various signal
structures.
The ability to instantiate any number of correlators per
channel (where each channel can be setup for any GNSS signal
structure), and have all these correlators and channels managed
with little-to-no user intervention is one of the most desired
features of a universal GNSS SDR. Tis is because it allows
the researcher to focus on higher-level algorithm development
without having to be concerned with correlator implementation
details. Bringing this idea to fruition was one of the major goals
and contributions of this efort. Te following section describes
the architecture of these universal GNSS correlators.
Post-Correlation Processing. Following sample correlation,
the data rate is reduced to an easily handled value of one kilo-
hertz. Most of the specialized GNSS signal processing algo-
rithm development occurs in this post-correlation domain.
Tis is also where the strengths of high-level algorithm devel-
opment tools such as MATLAB shine in terms of a researcher
being able to modify scripts and visualize the efects quickly
and easily.
An SDR toolbox must feature an interface to and from this
domain that is both efcient and intuitive in terms of confgur-
ing and controlling the various types of channels and correla-
tors as required by the researcher.
Functional Architecture
Tis article serves as an introduction to Version 3 of the GNSS
SDR toolbox. Tis versions functional architecture is signif-
cantly diferent to that of the previous version (v2) that was
described in the paper by S. Gunawardena (2013) listed in Addi-
tional Resources.
Figure 2 shows the high-level functional block diagram of
the GNSS SDR toolbox for MATLAB. Sampled data streams
are read from source SDR data fles, followed by bufering and
decoding into one or more data streams. Te streams are fed
into two main signal-processing blocks: a stream statistics and
noise processor, and a multi-channel ChipShape correlation
engine.
To maintain a regular channel architecture that is not spe-
cifc to any GNSS signal structure, the toolbox uses memory
codes exclusively for all pseudorandom noise and masking
sequences. Tese codes are fetched from fles and saved in a
cache that is accessible to both processing blocks. Tis code
cache is fully confgurable by the user such that unused codes
can be swapped out for new ones at runtime.
Te stream statistics and noise processor computes sample
means, variances, and histograms for every one-millisecond
block of samples. Sample statistics provide a valuable low-laten-
cy situational awareness indication of in-band interference.
Researchers can use the raw one-millisecond outputs of this
processor to prototype a range of interference detection/moni-
toring algorithms. Te toolbox includes commands to disable
these computations if not used.
In GNSS receivers, a channel control state machine is typi-
cally used to handle the transition from acquisition to steady-
state tracking (and subsequent reacquisition to tracking fol-
lowing loss-of-lock events). A low-latency signal-to-noise ratio
(SNR) estimate is used as one of the inputs to this controller.
Hence, the SNR calculation requires an estimate of noise power,
in general for each sample stream.
Some receivers employ a spare channel to compute this noise
estimate by correlating with a PRN sequence that is known to
be absent in the data. Te toolbox implements these noise cor-
relators within the stream processor block. To reduce computa-
tion load, noise correlators implement only the real component,
and the numerically controlled oscillator (NCO) phase register
sizes are also smaller than those used for tracking channels.
As with the statistics processes, each noise correlator can be
turned of to improve runtimes. Because these noise correla-
tors can be set to correlate with any of the confgured memory
codes (including for example, a dedicated random-noise code
of any length), the likelihood of signifcant cross-correlation
with in-band signals can be minimized.
Version 2 provided instantiation of any number of correlator
FIGURE 2 GNSS SDR Toolbox Version 3 high-level functional block
diagram
RECEIVER TOOLBOX
The ability to instantiate any number
of correlators per channel (where each
channel can be setup for any GNSS signal
structure), and have all these correlators
and channels managed with little-to-
no user intervention is one of the most
desired features of a universal GNSS
SDR.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
points per channel, where each could be
connected to one or more universal code
generators with independently variable
relative code phase delay. Even though
this architecture facilitates a wide range
of applications, this earlier version
repeated the underlying sample-level
multiply-accumulate operations when
points were placed with less than one-
chip separation from each other.
Version 3 eliminates these repeated
operations by natively performing Chip-
Shape correlation on an array of points.
Ranges of points from this ChipShape
output vector can be combined (at the
user level) to form any desired points
of the traditional triangular correla-
tion function. For binary ofset carrier
(BOC) signals, the need for a subcarrier
replica in the correlation process is also
eliminated because the user can apply
any subcarrier function as part of the
ChipShape-to-triangular conversion
step described previously. Further, by
applying chip masking patterns that
are specifc to the spreading code (com-
bined with long coherent integration),
transients of the underlying signal can
be observed at high fdelity.
This technique has applications in
advanced multipath mitigation, signal
quality monitoring and authentication.
The paper by S. Gunawardena et alia
(2012) listed in Additional Resources
provides an overview of ChipShape
processing, the concept of chip mask-
ing, and its applications in signal quality
monitoring.
Figure 3 shows the functional archi-
tecture of a Version 3 channel. As with
previous versions, the user can instanti-
ate any number of these channels in the
toolbox. The main user-configurable
channel parameters are shown in red.
Te Stream Index parameter selects
the input data stream to be processed by
a channel. Carrier wipeof is performed
on this selected stream using the replica
generated by the carrier NCO con-
trolled by phase-rate commands updat-
ed each millisecond. Te carrier-wiped
stream is then sent to independent banks
of correlators that perform ChipShape
correlation for a one-millisecond block of
samples. Te result is a ChipShape vector
for each bank that is transferred to the
user space (i.e., MATLAB workspace).
Each ChipShape bank is confgured
independently by means of three param-
eters: the number of correlation points
per chip N
C
, the whole number of chips
spanning to the early side N
E
(relative
to code NCO integer phase), and the
whole number of chips spanning to the
late side, N
L
. Hence, the size of the Chip-
Shape vector is given by N
C
(N
E
+ N
L
+ 1),
and the spacing between points is given
by 1/N
C
.
As shown in Figure 3, ChipShape
processing essentially splits traditional
correlation into partial accumula-
tions, where the fractional state of the
code NCO determines the array index
applicable to the partial accumulation
being processed. Splitting the correla-
tion operation in this way maximizes
opportunities for these accumulations
to be combined in user space to form
numerous correlation and/or code dis-
criminator functions depending on the
application. Another welcome beneft is
that this method reduces the dynamic
range required to prevent overfow of
these accumulators by a factor of 1/N
C
compared to a traditional correlator.
Not shown in Figure 3 are the three
levels of enable/disable logic featured
in the toolbox to improve runtimes:
1) enable/disable entire channels that
were instantiated (also disables chan-
nel NCOs), 2) enable/disable banks that
were instantiated within a channel, and
3) enable/disable individual points with-
in a given bank.
If a ChipShape correlator is imple-
mented as described thus far, the output
vector would simply be the diferential
of a traditional triangular correlation
function. Although useful, it does not
provide full insight into chip transition
edges and their precise zero crossings.
Te rising, falling, and stationary parts
of a GNSS signals underlying code
sequence can be recovered by correlat-
ing with a local replica that corresponds
only to these events (e.g., to recover the
rising-edge, keep all -1 to +1 chip transi-
tions in the code sequence and set others
to zero).
As shown in Figure 3, this function-
ality is implemented by multiplying the
carrier-wiped sample stream with an
optional masking sequence. (In actual-
ity the mask bit is used to disable accu-
mulation for that sample.) Each bank is
configured independently to point to
any code and/or mask sequence stored
in the Code Cache shown in Figure 2.
Applying the
ChipShape Correlator
The native ChipShape correlator
architecture of Version 3 signifcantly
expands possibilities for advanced GNSS
FIGURE 3 Functional architecture of a Version 3 ChipShape correlator channel
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
signals-based research beyond what was
possible with Version 2. Among these
are the following examples.
Built-In Acquisition and Rapid-Reac-
quisition. A dedicated channel can be
instantiated for acquisition and/or rapid
reacquisition. Tis channels ChipShape
vector would span a signifcantly larger
range of integer chip ofsets with coarse
inter-point spacing (e.g. N
C
=2 for BPSK,
N
C
=4 for BOC(1,1), and so on). A given
PRN can be acquired by pointing to the
corresponding memory code and pro-
gressively searching for sets of code phase
ofsets and Doppler frequencies over time.
For reacquisition, the channels car-
rier and code NCOs are set with best
estimates of Doppler frequency and
codephase, respectively. In this case,
ChipShape points corresponding to
unnecessary span may be disabled to
reduce runtimes. An estimate from the
noise processor can be used as the basis
for setting up the acquisition detection
threshold.
Spoofer Monitoring. A civilian GPS
spoofng scenario described in the paper
by T. E. Humphreys et alia (Additional
Resources) attempts to pull a receiver
tracking channel away from the genu-
ine signals correlation peak by coercing
it to lock on to a stronger peak produced
by the spoofer. Even if Doppler offset
and codephase are perfect matches to
the genuine signal, the superposition
of the two would cause signifcant dis-
tortion of the ChipShape function due
to the spoofers RF transmitter transfer
function (which itself is a function of its
characteristic analog RF components
including modulators, amplifers, flters
and antenna).
A monitor/detector could be imple-
mented using the GNSS SDR toolbox
based on a high-resolution ChipShape
output computed by an additional bank
in each channel. To reduce runtimes
(which corresponds to reducing power
in a practical application), this bank can
be activated at periodic intervals or at
the onset of in-band noise power fuc-
tuations (as monitored by sample vari-
ance and/or noise correlators), which
is a cheaper frst indicator of possible
in-band interference.
Chip EdgeBased Code Tracking for
Advanced Multipath Mitigation. As evident
from the ChipShape functions shown
in the examples section, zero-crossing
rising-edge and falling-edge transitions
are the highest-frequency components
attainable from any received GNSS sig-
nal through correlation processing. Tis
is true regardless of signal structure.
Hence, code tracking techniques that
are primarily based on these transitions
stand to produce the best pseudorange
accuracy and multipath mitigation per-
formance possible for any receiver of
that bandwidth. Researchers can use the
highly confgurable ChipShape outputs
produced by this toolbox as an enabler
for researching novel edge-based code
tracking techniques for precision GNSS
applications.
GNSS Signal Authentication. Variations
present in signal transmission payloads
of satellites are known to cause subtle
signal deformations that are detectable
using appropriate processing techniques.
Not surprisingly, ChipShape functions
are the cornerstone of these techniques.
For authentication applications, the
deformation caused only by the satellite
payload (as a function of nadir angle)
must be isolated from nuisance compo-
nents that include multipath, receiver
antenna/front-end transfer function
(including any variations due to tem-
perature, vibration, and aging), and
ionospheric efects.
Signal-Processing Applications
Tis section provides two GNSS signal-
processing examples that illustrate the
configuration and capabilities of the
toolbox.
Tracking and Eye Diagram Extraction
for BPSK(1) Signals: GPS L1 C/A. For this
example, a Version 3 channel was con-
fgured with fve banks as follows:
- Bank 1: N
C
=120, N
E
=N
L
=1, Code:
GPS C/A PRN, Mask: None
- Bank 2: N
C
=120, N
E
=N
L
=1, Code:
GPS C/A PRN, Mask: GPS C/A
PP PRN
- Bank 3: N
C
=120, N
E
=N
L
=1, Code:
PN PRN
- Bank 4: N
C
=120, N
E
=N
L
=1, Code:
NP PRN
- Bank 3: N
C
=120, N
E
=N
L
=1, Code:
NN PRN
where PRN is the C/A code PRN number
used and PP, PN, NP, and NN
correspond to masking codes in which
adjacent positive (P) and negative (N)
chip events are isolated from the origi-
nal C/A code. (Te distribution includes
a utility that generates these and other
masking code files from a given PRN
code fle.)
The GPS L1 C/A signal was pre-
acquired using the FFT-based Quick
Acquisition utility included in the
distribution. The latest distribution
includes a fully open-source, single
channeltracking script that features a
built-in tracking state controller. Tis
state machine was confgured to pull-in
from acquisition, perform bit synchro-
nization, and settle with the following
steady-state tracking loop parameters:
20 milliseconds pre-detection integra-
tion time, 18-hertz, third-order phase
locked loop (PLL) bandwidth, one-hertz
carrier-aided first-order delay locked
loop (DLL) bandwidth, and coherent
early-minus-late code phase discrimi-
nator with early-late correlator spacing
of 0.0167 chips.
Te fnal state activates banks 2 thru
3. Ten, using the navigation databit sign
derived from Bank 1 (i.e., sign[Prompt-
Q]) to keep rising, falling, positive, and
negative components of the underlying
signal together, the one-millisecond
ChipShape outputs are coherently com-
bined for approximately 100 seconds.
Accompanying fgures show the result-
ing normalized ChipShape outputs.
Figure 4 shows the GPS C/A code eye
diagram from a GPS front-end module
with approximately four megahertz
bandwidth. Te efect of narrow front-
end bandwidth compared to the results
depicted in the following two fgures is
clearly evident.
Figure 5 and Figure 6 show eye dia-
grams for GPS Block IIF-6 (SVN67
PRN06) at 78-degree elevation processed
from data obtained with the TRIGR
GNSS data collection system developed
by the Ohio University Avionics Engi-
neering Center. Te fnal-stage IF flters
for these two data streams included a
RECEIVER TOOLBOX
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
transversal surface acoustic wave (SAW)
flter with 24-megahertz/3-decibel band-
width and a lumped element elliptic
response flter comprised of a series of
coaxial bandpass flters with 3-decibel
bandwidth of 18 megahertz.
The bandwidth is sufficiently high
in both eye diagrams to observe the 10
cycles of ripple that occurs within a C/A
chip. As described in the article by S.
Gunawardena et alia (2012b), these oscil-
lations have been determined to be cross-
talk from the P(Y) code modulation.
Careful observation of time intervals
just prior to a chip transition in Figure
6 reveals a slight buildup of power.
Tis efect, not observable in Figure 5,
is primarily due to the finite impulse
response-type characteristic of trans-
versal SAW flters as will be reported in
detail in a forthcoming presentation by
S. Gunawardena et alia (2014) at the ION
GNSS+ in September.
Tracking and E1C/E1B Subcarrier
Extraction for CBOC(6,1,1/11) Signals:
Galileo E1. For this example, a Version 3
channel was confgured with four banks
as follows:
- Bank 1: N
C
=120, N
E
=N
L
=1, Code:
GAL E1C PRN, Mask: None
- Bank 2: N
C
=120, N
E
=N
L
=1, Code:
GAL E1B PRN," Mask: None"
- Bank 3: N
C
=120, N
E
=N
L
=1, Code:
GAL E1C PRN, Mask: GAL E1C
FF PRN
- Bank 4: N
C
=120, N
E
=N
L
=1, Code:
GAL E1C PRN," Mask: GAL E1B
FF PRN
where FF corresponds to masking
sequences where adjacent chips with the
same sign are isolated from the original
spreading code.
Banks 1 and 2 aie used foi pilot sig-
nal tracking and data symbol extraction,
respectively. The ChipShape outputs
from these banks are correlated with
the ideal CBOC subcaiiiei functions to
produce traditional early, prompt, and
late coiielation points. Banks 2, 3, and
4 are initially deactivated. Afer steady-
state tracking is reached, the ChipShape
outputs fiom Banks 3 and 4 aie cohei-
ently integrated for approximately 100
seconds by performing symbol wipeof
using the known overlay symbols and the
FIGURE 4 GPS L1 C/A signal eye diagram processed from a front-end with ~4-MHz pre-corre-
lation bandwidth
ChipShape, GPS PRN28, Bw=4MHz
Relative CodePhase [chips]
N
o
r
m
a
l
i
z
e
d

M
a
g
n
i
t
u
d
e
1.5
1
0.5
0
-0.5
-1
-1.5
PP Q
PN Q
NP Q
NN Q
PP I
PN I
NP I
NN I
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
FIGURE 5 GPS L1 C/A eye diagram processed from a front-end with 18-MHz pre-correlation
bandwidth
ChipShape, GPS BLK IIF-6 (PRN6) EI=78 deg. Bw=18MHz
N
o
r
m
a
l
i
z
e
d

M
a
g
n
i
t
u
d
e
1.5
1
0.5
0
-0.5
-1
-1.5
PP Q
PN Q
NP Q
NN Q
PP I
PN I
NP I
NN I
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
FIGURE 6 GPS L1 C/A eye eiagram processed from a front-end with 24 MHz pre-correlation
bandwidth
ChipShape, GPS BLK IIF-6 (PRN6) EI=78 deg. Bw=24MHz
N
o
r
m
a
l
i
z
e
d

M
a
g
n
i
t
u
d
e
1.5
1
0.5
0
-0.5
-1
-1.5
PP Q
PN Q
NP Q
NN Q
PP I
PN I
NP I
NN I
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
data symbols derived from the prompt
correlator of Bank 2, respectively.
Similar to the GPS C/A code tracking
example, the channel state machine was
confgured to obtain the same steady-
state tracking parameters. However,
instead of the bit synchronization state
used for GPS C/A code, the Galileo E1
tracking demo uses the included overlay
code synchronizer.
Figure 7 shows the one-millisecond
prompt correlator outputs (pilot and
data) from the acquisition pull-in state
to just afer activation of banks 24.
Figure 8 and Figure 9 show the Gali-
leo FM3 E1 CBOC(6,1,1/11) pilot and
data component subcarriers as observed
from front-end bandwidths of 18 and 24
megahertz, respectively. As to be expect-
ed, the multi-level subcarrier functions
experience more distortion with the
18-megahertz front-end compared to
24 megahertz. Also notice that for tra-
ditional early-minus-late discriminator-
based code tracking, zero crossings do
not occur at zero codephase due to band-
limiting.
Conclusion
Tis article introduced the GNSS SDR
Toolbox for MATLAB (Version 3). Tis
sofware performs GNSS SDR baseband
signal processing using an optimized
multi-threaded approach. The main
motivation behind the development of
this tool was to accelerate ofine pro-
cessing times for large GNSS SDR datas-
ets. Te toolbox improves runtimes by at
least a factor of 30 compared to equiva-
lent MATLAB-only scripts.
The main feature of Version 3 is a
multi-channel universal GNSS Chip-
Shape correlation engine that can be
used as the foundation for advanced
GNSS receiver development, algorithm
design, and prototyping. It can also be
used as an educational tool for demon-
strating advanced GNSS signal process-
ing techniques.
Te Version 3 distribution contains
numerous open-source scripts that
demonstrate the setup and use of all
major features. The toolbox is avail-
able free of charge for educational and
non-commercial research use. The
software and additional resources are
FIGURE 7 From example of tracking and E1C/E1B subcarrier extraction for CBOC: one-millisec-
ond Galileo E1C (pilot) and E1B (data) prompt correlator outputs over time
Galileo FM3 1 ms Prompt Correlator Outputs
Time [seconds]
C
o
r
r
e
l
a
t
o
r

M
a
g
n
i
t
u
d
e
1.5
1
0.5
0
-0.5
-1
-1.5
0
1
2 3 4 5 6 7 8
x10
4
FIGURE 8 Galileo FM3 CBOC(6,1,1/11) E1C and E1B subcarrier functions processed from a front-
end with 18 MHz pre-correlation bandwidth
Galileo FM3 CBOC(6,1,1/11) SubCarrier. EI: 85 deg. BW: 18 MHz
N
o
r
m
a
l
i
z
e
d

M
a
g
n
i
t
u
d
e
2
1.5
1
0.5
0
-0.5
-1
-1.5
-2
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
FIGURE 9 Galileo FM3 CBOC(6,1,1/11) E1C and E1B subcarrier functions processed from a front-
end with 24 MHz pre-correlation bandwidth
Galileo FM3 CBOC(6,1,1/11) SubCarrier. EI: 85 deg. BW: 24MHz
N
o
r
m
a
l
i
z
e
d

M
a
g
n
i
t
u
d
e
2
1.5
1
0.5
0
-0.5
-1
-1.5
-2
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
RECEIVER TOOLBOX
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
available through the authors blog:
<ChameleonChips.com>. Minimum
sofware requirements needed to run the
toolbox include Microsof Windows (32
or 64-bit) and MATLAB version 2007B
or above.
Acknowledgment
Tis article was adapted in part from a
presentation given by the author at the
ION GNSS+ 2013 conference on Sep-
tember 19, 2013. Te views expressed in
this article are solely those of the author
and not those of any other person, insti-
tution, organization, or entity.
Manufacturers
Te MATLAB toolbox described in this
article was developed using Microsof
Visual Studio from Microsof Corpora-
tion, Redmond Washington USA. Te
sofware was profled using Intel Paral-
lel Studio from Intel Corporation, Santa
Clara California USA. Te sofware runs
on, and plots for this article were gener-
ated using MATLAB from Te Math-
works, Inc. Natick, Massachusetts,
USA. The operating system used was
Windows 7 64-bit from Microsof Cor-
poration, Redmond Washington, USA.
Narrowband SDR data for the example
presented in the section titled Signal-
Processing Applications were collected
using a SiGe GN3S Sampler V3 from
Sparkfun Electronics, Boulder, Colo-
rado USA. Wideband GPS L1/Galileo E1
data were collected using a TRIGR GNSS
data collection system from the Ohio
University Avionics Engineering Cen-
ter, Athens, Ohio USA. Te fnal-stage
IF flters for the two data streams incor-
porated a SAWTEK 854672 transversal
SAW flter from TriQuint Semiconduc-
tor Inc., Hillsboro, Oregon, USA, and a
lumped element elliptic response flter
comprised a series of six SBP-70+ coax-
ial bandpass flters from Mini-Circuits,
Brooklyn, New York, USA.
Examples of low-cost data collection
hardware platforms that support GNSS
bands include the SiGe GN3S Sampler
(see Sparkfun Electronics publication,
Additional Resources), the Univer-
sal Sofware Radio Peripheral (USRP)
(Ettus Research in Additional Resourc-
es), and products based on fully inte-
grated feld programmable RFICs such
as the Loctronix ASR-2300 (Loctronix
Corporation, Additional Resources) and
bladeRF (Nuand, Additional Resources).
[1] Ettus Research, Universal Software Radio
Peripheral (USRP), <https://www.ettus.com/
product/details/UN210-KIT> (accessed July
2014)
[2] Galileo Open Service Signal in Space Inter-
face Control Document (OS SIS ICD), issue 1.1,
<http://ec.europa.eu/enterprise/policies/sat-
nav/galileo/files/galileo-os-sis-icd-issue1-
revision1_en.pdf> (accessed August 2013)
[3] Gunawardena, S. (2007), Development of
a Transform-Domain Instrumentation Global
Positioning System Receiver for Signal Quality
and Anomalous Event Monitoring. Electronic
Dissertation, Ohio University, 2007 <https://
etd.ohiolink.edu> (accessed August 2013)
[4] Gunawardena, S. (2013), A Universal GNSS
Software Receiver MATLAB Toolbox for Education
and Research, Proceedings of the 26th Interna-
tional Technical Meeting of The Satellite Divi-
sion of the Institute of Navigation (ION GNSS+
2013), pp. 1560-1576, Nashville, Tennessee,
USA, September 2013
[5] Gunawardena, S. (2011), and F. van Graas,
Multi-Channel Wideband GPS Anomalous Event
Monitor, Proceedings of the 24th International
Technical Meeting of The Satellite Division of
the Institute of Navigation (ION GNSS 2011), pp.
19571968, Portland, Oregon, USA, September
2011
[6]Gunawardena, S. (2012a), F. van Graas, High
Fidelity Chip Shape Analysis of GNSS Signals
using a Wideband Software Receiver, Proceed-
ings of the 25th International Technical Meeting
of The Satellite Division of the Institute of Navi-
gation (ION GNSS 2012), pp. 874-883, Nashville,
Tennessee, USA, September 2012
[7] Gunawardena, S. (2012b), and F. van Graas,
Analysis of GPS Pseudorange Natural Biases
using a Software Receiver, Proceedings of the
25th International Technical Meeting of the
Satellite Division of the Institute of Navigation
(ION GNSS 2012), Nashville, Tennessee, USA,
September 2012
[8] Gunawardena, S. (2014), and F. van Graas,
Analysis of GPS-SPS Inter-PRN Pseudorange
Biases due to Receiver Front-End Components,
Proceedings of the 27th International Technical
Meeting of The Satellite Division of the Institute
of Navigation (ION GNSS+ 2014), Tampa, Florida,
USA, September 2014
[9] Humphreys, T. E., and B. M. Ledvina, M. L.
Psiaki, B. W. OHanlon, and P. M. Kintner, Jr.,
Assessing the Spoong Threat: Development
of a Portable GPS Civilian Spoofer, Proceedings
of the 21st International Technical Meeting of
the Satellite Division of The Institute of Naviga-
tion (ION GNSS 2008), Savannah, Georgia, USA,
September 2008, pp. 2314-2325
[10] Loctronix Corporation, ASR-2300 MIMO
SDR, <http://www.loctronix.com/en/products_
asr_2300.html> (accessed July 2014)
[11] MathWorks Inc., MATLAB: the language of
technical computing<http://www.mathworks.
com/products/matlab> (accessed August 2013)
[12] Mathworks Inc., Introducing MEX-
Files,<http://www.mathworks.com/help/mat-
lab/matlab_external/introducing-mex-files.
html> (accessed July 2014)
[13] Nuand, bladeRF Software Defined Radio,
<http://nuand.com> (accessed June 2014)
[14] Ouvry, L., and C. Boulanger and J. R.
Lequepeys, Quantization effects on a DS-CDMA
signal, Spread Spectrum Techniques and Appli-
cations, 1998, Proceedings of the 1998 IEEE 5th
International Symposium, vol.1, pp. 234,238
vol.1, 2-4 September 24, 1998
[15] Sparkfun Electronics, SiGe GN3S Sam-
pler v3, <https://www.sparkfun.com/prod-
ucts/10981> (accessed August 2013)
Author
Sanjeev Gunawardena is a
GNSS research and
development engineer
with more than 15 years
of professional experi-
ence in the field. He
received his Ph.D. in
electrical engineering from Ohio University. His
research interests include RF design, digital sys-
tems design, high performance computing, soft-
ware radio, and all aspects of GNSS receivers and
associated signal processing.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_______________________
________
________________
______________________
__________________________
___
_________________________
__________________________
_________
____________________
______
________________
_____
_________
______________
I
n GNSS receivers, the acquisition
process is the frst stage of the sig-
nal-processing module. It consists in
assessing the presence of GNSS signals
and providing a rough estimation of the
incoming signal parameters: the Doppler
frequency and the code delay.
To detect the presence of the signal,
the received signal is correlated with a
succession of locally generated replicas
until the acquisition detector crosses a
predefined threshold. One commonly
used criterion of acquisition perfor-
mance is the probability of detection
when the parameters of the local replica
are (close to being) correct. Tis prob-
ability should be as high as possible but
under unfavorable conditions, such
as adverse environments, detection
becomes a challenge.
Initially, GNSS signals were only
defned on one component (such as GPS
L1 C/A) but the new generation of sig-
nals has two components (such as GPS
L1C, GPS L5, Galileo E1 OS, Galileo E5
a/b, and so forth): a data component
that carries the navigation message and
a pilot component, which does not carry
any useful information.
Designers of the modern civil signals
introduced the pilot component in order
to avoid the data bit transition prob-
lem during the tracking process. From
the point of view of signal acquisition,
however, the presence of a systemati-
cally known secondary code on the pilot
component still implies bit sign transi-
tion. Te presence of the pilot signal also
means that the total signal power is split
between components, thus impacting
the way to process such a signal to gather
all the signal power.
Te objective of this article is to study
the typical sources of performance deg-
radations of the GNSS acquisition pro-
cess that are generally overlooked in the
literature and to assess their efects on
the acquisition of new GNSS civil sig-
nals. We will focus on degradations due
to (1) the uncertainties brought by the
choice of the acquisition grid, (2) the
presence of bit sign transition, and (3)
the non-compensation of the code Dop-
pler. Further to the pure acquisition per-
formance, we also analyze the acquisi-
tion of the secondary code for new GNSS
signals and the frequency refinement
because these factors are necessary con-
ditions with which to initiate standard
tracking.
Tis study takes place in the context
of the development of a GNSS sofware
receiver that aims at acquiring any GNSS
civil signals at 27 dB-Hz and higher with
iStockphoto.com/ albln
The low power and spread
spectrum nature of GNSS
signals make their detection
and acquisition a key, but
challenging aspect of receiver
processing designs. A team
of researchers investigated
the performance of four new
GNSS signals and the legacy
GPS L1 C/A code, comparing
their probability of detection
at a specic level of received
signal strength. Factors of
particular interest included
the bit sign transition,
acquisition bin size, and
uncompensated code Doppler.
MYRIAM FOUCRAS, BERTRAND EKAMBI,
FAYAZ BACARD
ABBIA GNSS TECHNOLOGIES
OLIVIER JULIEN, CHRISTOPHE MACABIAU
COLE NATIONALE DE LAVIATION CIVILE
(ENAC)
WORKING PAPERS
Assessing the
Performance
of GNSS Signal
Acquisition
New Signals and
GPS L1 C/A Code
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
a strong probability of detection set to 95
percent. As a consequence, all presented
results refer to this test case.
In this article, we will frst introduce
the required acquisition parameters to
achieve the 27 dB-Hz/95 percent objec-
tive without considering any aforemen-
tioned source of degradation. Ten, we
discuss each point of degradation inde-
pendently and analyze its efect on the
probability of detection.
GNSS Signals
In this article, we consider the civil GPS
and Galileo signals in the L1/E1 and L5/
E5 bands.
Te main points of design of a GNSS
signal are:
- the caiiiei fiequency f
L
- the spieading codes c
1
characterized
by their length N
c1
, its chipping rate
f
c1
, or equivalently its chip duration
T
c1
= 1/f
c1
- the spieading code chip modulation
- the navigation message d on the data
component and the secondary code
c
2
on the pilot component (and some-
times also on the data component).
Table 1 summarizes the main signal
features for the considered GNSS sig-
nals.
The down-converted and filtered
composite GNSS signal entering the
correlation block of the receiver can be
generically represented as follows:
where
- x stands for D for the data compo-
nent and P for the pilot component
- A
x
is the signal amplitude on the
component and depends upon the
total signal power C
- p
x
is the subcarrier modulating the
spreading codes
- is the receiver PRN code delay
- f
1F
is the received intermediate fre-
quency of the receiver
- f
d
is the incoming Doppler frequency
-
0,x
is the initial phase on each com-
ponent depending on the initial
phase of the incoming signal
- n is the incoming noise, which is
assumed to be a white noise with
centered Gaussian distribution and
a constant two-sided power spectral
density equal to N
0
/2 dBW-Hz.
Note that in this expression, the role
of the RF front-end equivalent flter is
purposely ignored for simplification
reasons.
To complete the generic expression
of the received GNSS signal (1), Table 2
provides the value for each parameter.
As can be seen, the GNSS L5 signals are
in quadrature; however, the phase rela-
tionship between the two components of
GPS L1C is not yet specifed. (For details,
see the article by J. W. Betz et alia listed
in Additional Resources section near the
end of this article.)
For the purposes of this article, we
designated L1C to be an in-phase signal
as is the case for Galileo E1 OS. GPS L1C
presents a power diference in both com-
ponents 75 percent of the power in the
pilot component and 25 percent of power
f
L
(MHz) Modulation
Spreading code Data Secondary code
Length
T
c1
(ms)
N
c
1
(chips)
Rate wrt
MHz
Symbol
duration
T
d
(ms)
Code length
T
c
2
(ms)
N
c
2
(bits)
Bit duration
(ms)
GPS L1 C/A 1575.42 BPSK
1
1023
f
0
20 None None
GPS L1C
Data 1575.42 BOC(1,1)
10
10230
f
0
10 None None
Pilot 1575.42 TMBOC(6,1,1/11)
10
10230
f
0
None
18 000
1800
10
GPS L5
Data 1176.45 BPSK(10)
1
10230
10 f
0
10
10
10
1
Pilot 1176.45 BPSK(10)
1
10230
10 f
0
None
20
20
1
Galileo E1
OS
Data 1575.42 CBOC(6,1,1/11,+)
4
4092
f
0
4 None None
Pilot 1575.42 CBOC(6,1,1/11,-)
4
4092
f
0
None
100
25
4
Galileo E5a
Data 1176.45 BPSK(10)
1
10230
10 f
0
1
20
20
1
1
10230
10 f
0
None
100
100
1
Galileo E5b
Data 1207.14 BPSK(10)
1
10230
10 f
0
1
4
4
1
1
10230
10 f
0
None
100
10
1
TABLE 1. Key feaatures of GNSS signals
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
in the data component whereas the
total signal power is split in half 50/50
for the other GNSS composite signals.
GNSS Acquisition Performance
in Ideal Case
This section presents the acquisition
process in the case when none of the
sources of error mentioned in the intro-
duction are considered which can be
found in many assessment articles in the
literature. As explained in the introduc-
tion, the chosen test case is to acquire
any GNSS civil signals at 27 dB-Hz (total
signal carrier-to-noise-density ratio or
C/N
0
) with a probability of detection of
95 percent.
Correlation Operation. Considering the
correlation operation for one component
of the GNSS signal and assuming that:
- theie is no data bit sign tiansition
during the correlation process
- the coiielation opeiation lasts foi T
I
seconds
- the paiameteis of the piocessed sig-
nal and the local replica are constant
during the correlation operation such
that the code delay error
and the
Doppler frequency error
f
are con-
stant and the carrier phase error at
the beginning of the correlation pro-
cess is
0
.
Te in-phase and quadrature-phase cor-
relator outputs can be modelled as:
where
- n
x,I
and n
x,Q
are the noises at the cor-
relator output (independent) that fol-
low a centered Gaussian distribution
with variances
- R
x
is the autocorrelation function on
the x component of the signal.
Note that the aforementioned corre-
lator outputs model neglects the cross-
correlation between the data and pilot
component because the spreading codes
were chosen to be as orthogonal as pos-
sible. Note also that the local spread-
ing code is assumed to have the same
modulation as the spreading code of the
received signal.
Acquisition detector. A receiver can
acquire composite GNSS signals by
using correlator outputs based on one
of the two components (in general, the
pilot component) or both data and pilot
components. In either case, the acqui-
sition detector is defned as the sum of
the squared correlator outputs (2 when
only one component is used, 4 when two
components are used). Te acquisition
detector for one component is thus
where K represents the number of non-
coherent summations. In this case KT
I
is
referred to as dwell time, and the param-
eters of the local replica (local PRN code
delay and Doppler ) are constant for
the K correlations.
The acquisition detector based on
the use of two components can be easily
derived accordingly.
Probability of detection. The basic
principle of acquisition is to sequentially
compute the acquisition detector for all
possible values of local code delay and
local Doppler until the detector crosses
a predefned threshold T
h
. Te set of the
tested couples ( , ) is defined as the
WORKING PAPERS
Data component Pilot component
A
D
c
2,D
p
D
0,D
A
p
P
p
0,P
GPS L1 C/A 1 1
0
0 None None
GPS L1C
25%
1 p
BOC(1)
(t)
0
None
0
GPS L5
50%
NH
10
1
0
50%
1
0
Galileo E1 OS
50%
1
0
50%
0
Galileo E5a
50%
1 1
0
50%
1
0
Galileo E5b
50%
1 1
0
50%
1
0
with
where p
BOC(y)
(t) = sign(sin(2 y f
0
t))
TABLE 2. GNSS signals features
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
acquisition matrix, and its size depends
upon the uncertainty on the incoming
signal code delay and Doppler frequency
and on the sampling of these uncertain-
ties.
Te tested values of the acquisition
matrix are referred to as acquisition
bins, and the distance between two
consecutive tested values is referred to
as bin size. Te detection performance
of such a detector is generally computed
based on a hypothesis test for each vis-
ited acquisition matrix bin: hypothesis
H
0
assumes that the desired signal is not
present and is tested against hypothesis
H
1
that assumes that it is present.
Under hypothesis H
0
the correlator
outputs only consist of independent
Gaussian noises. In this case, the (nor-
malized) detector follows a centered
2
distribution with 2K or 4K degrees of
freedom for the one-component and
two-component cases, respectively.
For a desired probability of false alarm
P
fa
, we can thus defne the appropriate
threshold T
h
.
The alternat ive hypot hesis H
1
assumes that the signal is present, mean-
ing that the parameters of the local rep-
lica are almost aligned with the ones of
the received signal. In this case, the (nor-
malized) detector follows a non-central
2
distribution with 2K or 4K degrees of
freedom for one-component and two-
component cases, respectively.
Te non-central parameter of the
2
distribution depends upon the receiver
signal C/N
0
, the correlation duration T
I
,
and the uncertainty of the parameters
( , ) due to the acquisition bin size.
We can then compute the probability
of detection P
d
by comparing the detec-
tor distribution to the threshold T
h
. As
a synthesis, the key acquisition param-
eters are presented in Table 3.
Minimum Dwell Time to Reach a Desired
Probability of Detection. To fnd results
related to our test case, we selected a
desired probability of false alarm P
fa
=
1e
3
as described in the RTCA, Inc. arti-
cle referenced in Additional Resources.
To reach this objective, determining the
dwell time KT
I
is important. As T
I
is
generally taken equal to the spreading
code period during the acquisition pro-
cess, K is the critical parameter to play
with.
Assuming that the acquisition bin
size is infinitely small (thus meaning
that
=
f
= 0), Table 4 indicates the
value of K to reach the proposed objec-
tive. Tis table shows that the composite
GNSS signals having a data/pilot power
share of 50/50 percent require a dwell
time twice as short when both compo-
nents are used compared to when only
one component is used.
In the table, note that for GPS L1C,
with a data/pilot power share of 25/75
percent, using only the pilot component
or both components produces equivalent
results. Finally, the well-known prefer-
ability of having a long coherent inte-
gration time to improve the acquisition
detection performance explains why, for
example, the GPS L1C and Galileo E1
OS require a lower dwell time than GPS
L1 C/A or GPS L5. (See the discussion
in F. Bastide et alia cited in Additional
Resources.)
Effect of Acquisition Bin Size
on Acquisition Detection
Performance
Clearly, it is irrelevant to assume that the
acquisition bin size is infnitely small.
Indeed, a trade-off should be chosen
between the acquisition bin size and
the acquisition duration: a large bin size
leads to degradation of the acquisition
performance (the error between the test-
ed values and the true values can be sig-
nifcant), while a narrow bin size means
a signifcant number of bins potentially
have to be visited, thus increasing the
mean-time-to-acquire the signal.
In general, the acquisition grid is
defned as a function of the maximum
acceptable degradation on the detec-
tor. Following the example used in the
RTCA/DO-235B, we chose
- a Dopplei bin size of 1/2TI, corre-
sponding to an equivalent degrada-
tion of the received signal C/N
0
of
0.9 dB, which corresponds to a maxi-
mum Doppler frequency error |
f
|
1/4T
I
- a bin size in the code delay domain
sufficient to generate a maximum
equivalent degradat ion of t he
received signal C/N
0
of 2.5 decibels.
Te code delay bin size thus depends
on the autocorrelation function shape
(and in fact on the RF front-end flter
as well). For example, it corresponds
to a bin size of one-half chip for an
unfltered GPS L1 C/A or GPS L5 sig-
nal.
Figure 1 shows the probability of
detection as a function of the Doppler
Pilot component acquisition Total signal acquisition
Acquisition detector T = T
P
T = T
D
+ T
P
Treshold
Probability of detection
Non-centrality parameter
where F
2(ddl)
is the approximately equal cumulative distribution function of a
2
distribution
with ddl degrees of freedom.
TABLE 3. Acquisition as a detection problem
GPS L1 C/A
GPS L1C GPS L5 Galileo E1 OS Galileo E5a and E5b
Pilot Both Pilot Both Pilot Both Pilot Both
K 126 6 5 433 217 40 20 433 217
Dwell time
K
T1
(ms)
126 60 50 433 217 160 80 433 217
TABLE 4. Required dwell time to acquire signal with a C/N
0
or 27 dB-Hz for a desired probability of detection of 95%
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
uncertainty created by the bin size for
the selected test case and the number
of non-coherent summations given by
Table 4. In the worst case (limit of the
cell), the probability of detection falls
from 0.95 down to 0.8. A more rel-
evant fgure is the average probability
of detection over the bin, assuming that
the actual Doppler error is a random
variable uniformly distributed over the
entire bin. The average probability of
detection is also plotted in Figure 1 and
equals 0.91.
Figure 2 shows the same thing for
the code delay uncertainty within one
acquisition bin. In the worst location (on
the edge of a bin in the acquisition grid),
it goes down from 0.95 to 0.43 while the
average probability of detection over the
bin is 0.74.
If the worst cases in the frequency
and time domains are combined, the
total loss on the equivalent received C/
N
0
is 3.4 decibels and results in a prob-
ability of detection down to 0.25 instead
of 0.95. Te average probability of detec-
tion over the bin is around 0.67, thus
showing a theoretical degradation of
performance of 28 percent.
Bit Sign Transitions
and Receiver Performance
The presence of bit sign transitions
affects receiver performance in signal
acquisition detection. The following
discussion addresses this phenomenon
and associated factors.
Bit Transition Problem. Te correlator
output models provided in Equation (3)
assumed that the data and/or the sec-
ondary code bits are constant during
the correlation interval. During the
acquisition process, however, we have
no reason to assume that the integra-
tion interval is aligned with the data bit.
Although ofen neglected in the litera-
ture, it thus seems necessary to develop
the correlation output model consider-
ing bit sign transitions. The authors
have performed such a study, including
the theoretical aspects for single- and
dual-component signals, and a paper
M. Foucras et alia (2014a) in Additional
Resources describing the results will
be submitted for publication. Te fol-
lowing is a short summary with corre-
sponding results.
The presence of a bit sign transi-
tion during the correlation operation
degrades the useful part of the correlator
output without modifying the power of
the noise. Tis results in a degradation
of the acquisition detector amplitude,
the nature of which will depend upon
the location of the bit sign transition
in the integration interval, the number
of non-coherent summations, and the
f
as described
in the paper by C. ODriscoll. In particu-
lar, the expression of the non-centrality
parameter in case of a bit sign transition
during the integration interval is given
in M. Foucras et alia (2014a). As might
be expected, the worst case is for a bit
sign transition occurring in the middle
of the correlation interval.
For all GNSS signals discussed in
this article except GPS L1 C/A, a bit sign
transition can occur at each spreading
code period. Tis means that the corre-
lation duration should be limited to the
code duration, and that even then, a bit
sign transition can potentially degrade
all correlator outputs. In the article by
M. Foucras et alia (2014b), the authors
have identifed for each GNSS signal the
resulting average probability of detec-
tion for the number of bit sign transi-
tions, taking into account the probabil-
ity of occurrence.
In contrast, the acquisition perfor-
mance of the GPS L1 C/A signal, when
considering bit sign transition, depends
on the correlation duration. Indeed,
because the data bit duration is 20 times
longer than the spreading code period,
we can use correlation durations of 1, 2,
4, 5, 10 or 20 milliseconds. Each case will
have a diferent probability of undergo-
ing a sign transition during the corre-
lation. Consequently, for an equivalent
dwell time say, 20 milliseconds the
effect on the acquisition performance
depends on the choice of T
I
as explained
in M. Foucras et alia (2014b).
As shown in Figure 3, when the T
I
is too short, the effect of the bit sign
transition is slight, but it does not allow
optimal detection. On the contrary, for
long T
I
, the efect of the bit sign transi-
tion is signifcant. Based on Figure 3, it
WORKING PAPERS
FIGURE 2 Probability of detection versus the code delay uncertainty for BPSK-
modulated signals
Max. degradation of 2.5 dB code delay uncertainty : 0.43164)
1
0.8
0.6
0.4
0.2
0
P
r
o
b
a
b
i
l
i
t
y

o
f

d
e
t
e
c
t
i
o
n
0.2 0.1 0 0.1 0.2
Code delay error
f
(chip)
P
d
Mean probability : 0.73585
P
d
= 0.95
FIGURE 1 Probability of detection versus the Doppler frequency uncertainty
Max. degradation of 0.9 dB Doppler frequency uncertainty : 0.80026)
1
0.8
0.6
0.4
0.2
0
P
r
o
b
a
b
i
l
i
t
y

o
f

d
e
t
e
c
t
i
o
n
200 100 0 100 200
f
(Hz)
P
d
Mean probability : 0.90907
P
d
= 0.95
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
appears that a correlation duration of 4
to 10 milliseconds is optimal to have the
lowest dwell time to reach a probability
of detection of 95 percent.
Resulting Probability of Detection.
Table 4 provided the required dwell
time to reach a probability of detection
of 95 percent for a signal with a C/N
0
of
27 dB-Hz without considering bit sign
transition or uncertainty due to the
acquisition bin size. For the same dwell
time and C/N
0
, Table 5 shows the aver-
age probability of
detection based on
Monte Carlo simu-
lations assumi ng
that the distribution
of the location of bit
sign transition is
uniform within the
correlation inter-
val (assumed equal
to one spreadi ng
code). As discussed
previously, for GPS
L1 C/A we chose
the coherent inte-
gration time to be
the spreading code
period (one milli-
second). Note that
in this latter case
considering the bit
sign transition, Fig-
ure 3 showed that
this value for T
I
is
not the optimal one.
Table 5 shows
that all the compos-
ite signals are highly
afected, mostly due
to the fact that the correlation duration
has to be chosen equal to the data bit/
secondary code bit duration. As a con-
sequence, it seems necessary for these
signals to use techniques that are insen-
sitive to data bit sign transitions, such as
the techniques described in the article
by M. Foucras et alia (2012). Tese tech-
niques are generally more demanding
in terms of resources. However, GPS L1
C/A is almost not afected thanks to its
structure based on a data bit duration
20 times longer than the spreading code
duration.
Uncompensated Code Doppler
and Receiver Performance
We now turn to the question of the efect
of an uncompensated code Doppler on
acquisition detection performance.
Code Doppler problem
Te Doppler frequency, mainly caused
by the satellite motion and the receiver
local oscillator, afects the processed sig-
nal by modifying
- the cential caiiiei fiequency - a
change estimated by the acquisition
process
- the code fiequency (chipping iate)
resulting in a code Doppler f
cd
which
depends on the incoming Doppler
frequency f
d
, the carrier frequency
f
L
and the chipping rate frequency f
c1
according to
The modification of the code fre-
quency leads to a change in the spread-
ing code period as can be seen in Figure
4 where three periods of a four-chip
spreading code are represented:
- A positive code Dopplei fiequency
causes the spreading code duration
to shrink (T
cd
< T
c1
).
- A negative Dopplei shif causes the
spreading code duration to expand
(T
cd
> T
c1
).
The problem of the presence of an
uncompensated code Doppler resulting
in a difference between the code fre-
quency of the received and the local sig-
nals for GPS L1 C/A has been addressed
by several authors. E. D. Kaplan and C.
Hegarty. (See Additional Resources).
Foucras et alia (2014c) showed that the
degradations due to uncompensated
code Doppler are even more signifcant
for the new generation of GNSS signals
FIGURE 3 Average probability of detection at 27 dB-Hz for GPS L1 C/A
Average probability of detection (on t
0
)
1
0.8
0.6
0.4
0.2
0
P
r
o
b
a
b
i
l
i
t
y

o
f

d
e
t
e
c
t
i
o
n
20 40 60 80 100
Total integration time KT
I
(ms)
T
I
= 1 ms
T
I
= 2 ms
T
I
= 4 ms
T
I
= 5 ms
T
I
= 10ms
T
I
= 20ms
GPS L1 GPS L1C GPS L5 Galileo E1 OS Galileo E5a and E5b
C/A Pilot Both Pilot Both Pilot Both Pilot Both
0.94 0.71 0.67 0.56 0.56 0.62 0.62 0.56 0.56
TABLE 5. Probability of detection when considering bit sign transitions for a C/N
0
of 27 dB-Hz
GPS L1 C/A
(dwell Time =
126 ms)
GPS L1C
(dwell Time =
50 ms)
GPS L5
(dwell Time =
217 ms)
Galileo E1 OS
(dwell Time =
80 ms)
Galileo E5a
(dwell Time =
217 ms)
Galileo E5b
(dwell Time =
217 ms)
Incoming
Doppler
frequency
1 kHz 0.081 0.033 1.887 0.052 1.887 1.839
5 kHz 0.409 0.162 9.435 0.260 9.435 9.195
10 kHz 0.818 0.325 18.870 0.520 18.870 18.390
TABLE 6. Offset between the local and received spreading code after the dwell time (in chips)
FIGURE 4 Code Doppler effect on the spreading code period
1st period 2nd period 3rd period
1st period 2nd period 3rd period
Local code
Received code
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
(higher code frequency, lower L-band central frequency, BOC
modulation).
Table 6 presents the ofsets between the local and received
spreading codes afer the dwell time for the signals being con-
sidered in this article. For GPS L1 C/A, GPS L1C, and Galileo
E1 OS, the ofset is lower than one chip even for high incoming
Doppler frequency. Still, the ofset can sometimes be greater
than one code delay bin size, which can be problematic. For L5
signals, the ofset exceeds one chip for an incoming Doppler
of several hundreds of hertz with the considered dwell time.
For high Doppler frequencies this means that the ofset is too
high to provide correct acquisition performance, as it will be
shown later.
To illustrate this point, Figure 5, Figure 6, Figure 7 and Figure
8 represent the deformation of the squared correlation func-
tions between the incoming signal spreading code and the
local replica spreading code for GPS L1 C/A, GPS L5, Galileo
E1 OS and GPS L1C, respectively, due to uncompensated code
Doppler and with the dwell times as defned in Table 4. For
BPSK-modulated signals (GPS L1 C/A and GPS L5), the shape
of the autocorrelation function becomes rounded and ofset
compared to the reference triangular curve. Te amplitude of
the maximum value of the correlation function is also reduced,
and the peak is shifed to the right for a negative Doppler. Te
result is a degradation of the probability of detection and a
potential missed detection due to the motion of the correlation
peak with time.
Even if the correlation functionpeak offset is not such a
problem for GPS L1 C/A due to its relatively slow chipping rate,
this can be a real problem for GPS L5, as seen in Figure 6 where
the correlation peak has moved by more than one chip over
the 217-millisecond dwell time. For Galileo E1 OS, the CBOC
modulations correlation function has a signifcantly reduced
amplitude and its shape becomes fat when the code Doppler
WORKING PAPERS
FIGURE 5 Autocorrelation function when considering code Doppler for GPS L1 C/A
on 126 ms
GPS L1 C/A
1
0.8
0.6
0.4
0.2
0
S
q
u
a
r
e
d

a
u
t
o
c
o
r
r
e
l
a
t
i
o
n

f
u
n
c
t
i
o
n
0.5 0 -0.5 -1 1 1.5 2
Code delay (chip)
0kHz
2kHz
4kHz
6kHz
8kHz
10kHz
FIGURE 6 Autocorrelation function when considering code Doppler for GPS L5 on
217 ms
GPS L5
1
0.8
0.6
0.4
0.2
0
S
q
u
a
r
e
d

a
u
t
o
c
o
r
r
e
l
a
t
i
o
n

f
u
n
c
t
i
o
n
0.5 0 -0.5 -1 1 1.5 2
Code delay (chip)
0kHz
2kHz
4kHz
6kHz
8kHz
10kHz
FIGURE 7 Autocorrelation function when considering code Doppler for Galileo E1
OS on 80 ms
Galileo E1 OS
1
0.8
0.6
0.4
0.2
0
S
q
u
a
r
e
d

a
u
t
o
c
o
r
r
e
l
a
t
i
o
n

f
u
n
c
t
i
o
n
0.5 0 -0.5 -1 1 1.5 2
Code delay (chip)
0kHz
2kHz
4kHz
6kHz
8kHz
10kHz
FIGURE 8 Autocorrelation function when considering code Doppler for GPS L1C on
50 ms
GPS L1C
1
0.8
0.6
0.4
0.2
0
S
q
u
a
r
e
d

a
u
t
o
c
o
r
r
e
l
a
t
i
o
n

f
u
n
c
t
i
o
n
0.5 0 -0.5 -1 1 1.5 2
Code delay (chip)
0kHz
2kHz
4kHz
6kHz
8kHz
10kHz
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
increases due to the presence of the side peaks. Tis can then cre-
ate a detection problem as several bins could trigger a detection.
If the slip between the received and the local spreading
codes exceeds one chip, then the correlation process no longer
makes sense because the power of the signal cannot be accu-
mulated since the correlator output is essentially noise. Figure
9 shows the linear relationship between the incoming Doppler
frequency and the time to the slip of one chip.
For the maximum incoming Doppler frequency considered
in this article (10 kilohertz), the slip of one chip occurs afer
154 milliseconds for a GNSS signal at L1 and afer only 12 mil-
liseconds for GNSS L5 signals. For GPS L5, for example, that
means the previously computed dwell time of 217 milliseconds
would not be realistic as it implies a slip of 18 chips. So, the code
Doppler clearly needs to be dealt with in a GPS L5 or Galileo
E5a/E5b receiver, and potentially in a GPS L1 C/A, GPS L1C,
or Galileo E1 OS receiver.
To complete this part of our investigation, Figure 10 presents
the losses on the maximum amplitude of the squared autocor-
relation function. Te maximum losses are for L5/E5 signals,
because these experience a slip of more than one chip (Fig-
ure 6). Te minimum loss is for GPS L1 C/A (1.9 decibel for a
code Doppler of 10 kilohertz), which is better than GPS L1C
(2.5 decibels) and Galileo E1 OS (4.5 decibels) due to its BPSK
modulation, even if the dwell time is longer (126 milliseconds
instead of 50 or 80 milliseconds).
Resulting probability of detection. Let us now consider the
resulting probability of detection taken in = 0 (Figure 11).
Clearly, for GNSS L5 signals, the probability of detection
decreases because the shif between the incoming and the local
signals is too large.
Performance of the Acquisition-to-Tracking
Transition
Once acquisition has been successful, the frequency estimate is
on the order of a few tens or hundreds of hertz, depending upon
the acquisition bin size. However, at the initiation of the track-
ing process, a refnement on the Doppler frequency is required
in order to ensure locking the phase lock loop (PLL).
Frequency tracking. One solution is to use a frequency
lock loop (FLL), which refnes the estimation of the Doppler
frequency. Tis is a critical stage in GNSS signal processing
because, if this transition is not well calibrated, even a success-
ful acquisition can lead to unsuccessful tracking, especially at
low received C/N
0
.
Te authors undertook a performance study for various
FLL schemes, which was described in the article by M. Fou-
cras et alia (2014d) listed in Additional Resources. Based on
the proposed test case, the probability of achieving FLL lock
was analyzed assuming a C/N
0
of 27 dB-Hz. Te four FLL dis-
criminators examined in the study are the cross-product (CP),
the decision directed cross product (DDCP), the diferential
arctangent (Atan), and the four-quadrant arctangent (Atan2).
We should mention that during this initial phase of GNSS sig-
nal tracking being studied, bit synchronization has not yet been
achieved.
FIGURE 9 Time for the slip of one chip in function of the incoming Doppler
frequency
2000
1500
1000
500
0
S
l
i
p

o
f

1

c
h
i
p

(
m
s
)
0 2 4 6 8 10
Incoming Doppler frequency (kHz)
GPS L1 C/A
GPS L1C
GPS L5
Galileo E1 OS
Galileo E5a
Galileo E5b
FIGURE 10 Losses on the autocorrelation function due to code Doppler
0
-5
-10
L
o
s
s

i
n

m
a
x

o
f

a
u
t
o
c
o
r
r
e
l
a
t
i
o
n

f
u
n
c
t
i
o
n

(
d
B
)
4 6 8 10
Doppler (kHz)
GPS L1 C/A
GPS L1C
GPS L5
Galileo E1 OS
Galileo E5a
Galileo E5b
FIGURE 11 Probability of detection considering the total signal power and code
Doppler
1
0.8
0.6
0.4
0.2
0
P
r
o
b
a
b
i
l
i
t
y

o
f

d
e
t
e
c
t
i
o
n
0 2 4 6 8 10
Incoming Doppler frequency (kHz)
GPS L1 C/A
GPS L1C
GPS L5
Galileo E1 OS
Galileo E5a
Galileo E5b
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
WORKING PAPERS
Table 7 summarizes the expressions,
linear regions, and characteristics of the
four candidate FLL discriminators. As
only two discriminators are bit sign
transition insensitive, this feature plays
a key role in the choice of the discrimi-
nator for the best FLL scheme.
Probability of Successful Transition.
The key figure of merit for the acqui-
sition-to-tracking process is the prob-
ability of successful transition (or con-
vergence) of the FLL, regardless of the
initial frequency error afer acquisition
(within the correct acquisition bin, thus
with a Doppler error within
hertz in the proposed case) as a func-
tion of the GNSS signal and the FLL dis-
criminator. Te convergence is assessed
by making sure that the loop is locked
afer 20 seconds of tracking. Te prob-
abilities are obtained based on 200 runs
per confguration.
For the simulations, the article by
M. Foucras et alia (2014d) (Additional
Resources) showed that it is better to
choose an FLL loop bandwidth B
L
that
is relatively reduced even though this
reduces the response time of the loop.
B
L
= 1 Hz is used in the following results.
Finally, for composite GNSS signals,
two techniques were investigated: the
first one consists of tracking only the
pilot component and the second one
consists of tracking both components
by computing a FLL discriminator based
on an average of the data and pilot dis-
criminators (thus using the whole avail-
able signal power).
Two fgures present the probabilities
of successful transition for a signal with
a C/N
0
equal to 27 dB-Hz. Figure 12 con-
siders the pilot-only cases whereas Fig-
ure 13 considers a scheme using the total
available power. As expected, successful
convergence depends upon the initial
frequency error (it is better to start close
to the correct value).
In the legend of each fgure, the mean
probability of successful transition in
the cell is provided. As can be observed,
for GPS L1 C/A, whatever the Doppler
initial frequency error, the FLL always
converges using the CP or Atan2 dis-
criminators, thus fnely dealing with bit
sign transitions.
- Foi GNSS composite signals, how-
ever, this is no longer the case:
- Foi the GPS L1C signal, due to the
Discriminator expression Linear region Characteristics
CP Linear region independent from SNR
DDCP Bit transition insensitive
Atan Bit transition insensitive
Atan2 Highest linear region
where U is the phase unwrapping function which maps the phase estimate to the interval and the Cross(k) and Dot(k) expressions are
defned by
TABLE 7. Frequency discriminators
FIGURE 12 FLL schemes results when using only pilot component
1
0.5
0
P
I
0 -10 -20 10 20
GPS L1C
Frequency (Hz)
DDCP : 0.93059
Atan : 0.94784
DDCP : 0.014706
Atan : 0.11941
1
0.5
0
P
I
0 -10 -20 10 20
Galileo E5
Frequency (Hz)
1
0.5
0
P
I
0 -50 50
Galileo E1 OS
Frequency (Hz)
DDCP : 0.93059
Atan : 0.94784
FIGURE 13 FLL schemes results when using total signal power
1
0.5
0
P
I
0 -10 -20 10 20
GPS L1C
Frequency (Hz)
DDCP : 0.95412
Atan : 0.954331
1
0.5
0
P
I
0 -100 -200 100 200
GPS L1 CA
Frequency (Hz)
CP : 1
Atan2 : 1
DDCP : 0.031176
Atan : 0.22529
1
0.5
0
P
I
0 -10 -20 10 20
Galileo E5
Frequency (Hz)
1
0.5
0
P
I
0 -50 50
Galileo E1 OS
Frequency (Hz)
DDCP : 0.93765
Atan : 0.9051
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
Hosted By
Co-Sponsors
TM
ROCK THE BAY
WHEN
Thursday,Sept,11,2014
8p.m.toMidnight
WHAT
LiveMusic,Karaoke,
Dancing,DoorPrize,
GreatFun!
WHERE
TampaMarriottWatersideHotelandMarina
(ION GNSS HeadquartersHotel)
700SouthFloridaAvenue
Tampa,Florida33602
ALL ION GNSS+ 2014 ATTENDEES WELCOME
Join us for
Open Mic Night Tampa
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
presence of 75 percent of the signal
power contained in the pilot compo-
nent, the diference between the two
schemes (considering pilot or both
components) is slight. The perfor-
mance for both bit transitioninsen-
sitive discriminators is similar and
around 0.95 in mean value.
For Galileo composite signals, it
appears preferable to use both com-
ponents. In this case, for the Galileo
E1 OS signal, the average value is 0.94
for the DDCP discriminator and has a
performance very similar to GPS L1C.
For the Galileo E5a or Galileo E5b sig-
nals (and GPS L5, not shown here), the
probabilities to get locked are very low
(mean value 0.23), which constitutes a
signifcant problem for the acquisition-
to-tracking transition. This can be
explained by the short integration time
(one millisecond) associated to these
signals which implies a high correlator
output noise variance for a signal with a
C/N
0
equal to only 27 dB-Hz.
Secondary code acquisition perfor-
mance. Te pilot component was initially
introduced to avoid the data bit transi-
tion problem on the data component.
Indeed, the pilot component is free of
transition once the secondary code is
demodulated. This leads to the use of
longer coherent integration for a more
robust tracking.
In an article by M. Foucras et alia
(2013), the authors provided a detailed
analysis on the probability of acquiring
the secondary code for several GNSS
composite signals. Te main conclusion
of this study was that the C/N
0
threshold
to acquire the secondary code with a very
high probability was much lower than 27
dB-Hz and should not be a problem.
Conclusions
Signal acquisition is a crucial processing
step in GNSS receivers. A useful signal
must be extracted from the incoming
signal that is assimilated in the back-
ground RF noise, and its parameters
should be estimated. Due to these con-
ditions, the acquisition process at low
received C/N
0
is a challenge.
We conducted a detailed analysis of
all the sources of acquisition degrada-
tions, treating each point separately as
described in this article, to understand
its specifc efect. Our emphasis was on
the probability of detection, voluntarily
putting aside the time-to-acquire fac-
tor, which is operationally of equivalent
importance. Te article also concentrat-
ed on a specifc test case, which was to
be able to acquire a GNSS signal with a
received C/N
0
of 27 dB-Hz with a prob-
ability of detection of 95 percent.
Te frst point that we addressed was
the degradation of signal acquisition
performance caused by the estimated
parameters uncertainty brought by the
size of the acquisition bin. A typical bin
size results in an average degradation of
the probability of detection on the order
of 5 to 20 percent in the test case that we
considered.
We then showed that the problem
of the bit-sign transition was not a big
issue for the acquisition of GPS L1 C/A.
Tis is because a data bit transition can
occur only every 20 spreading code
periods, and a good choice of the coher-
ent integration time enables a receiver
to limit the degradation of acquisition
performance.
However, for the new GNSS signals
considered in our research, a bit sign
(data or secondary code) transition can
occur at each spreading code period, and
the adverse efect on the acquisition per-
formance can become substantial. As a
consequence, we highly recommend use
of a transition-insensitive acquisition
technique for these signals even if they
are more computationally expensive.
We also showed that an uncompen-
sated code Doppler particularly afects
the acquisition performance for GNSS
L5 signals due to their high frequency
chipping rate. If not taken care of prop-
erly, this efect results in a correlation
function shape becoming rounded and
fattened, leading to a potentially poor
estimation of the incoming code delay.
Our research also showed that the BOC-
based signals are more inf luenced by
code Doppler due to the shape of their
correlation function. As a consequence,
if code Doppler is not taken into account
by the receiver, it becomes necessary to
limit the acquisition dwell time even if
this penalizes the acquisition perfor-
mance at low C/N
0
(it does anyway).
Finally, we described the use of FLL
for the carrier acquisition-to-tracking
process, with the main conclusions
being to use bit transitioninsensitive
discriminators for composite GNSS sig-
nals.
[1] Bastide, F., and O. Julien, C. Macabiau, and B.
Roturier, Analysis of L5/E5 Acquisition, Tracking
and Data Demodulation Thresholds, in Proceed-
ings of the 15th International Technical Meeting
of the Satellite Division of The Institute of Navi-
gation (ION GPS 2002), Portland, Oregon, USA,
2002, pp. 2196 2207
[2] Betz, J. W., and M. A. Blanco, C. R. Cahn, P. A.
Dafesh, C. J. Hegarty, K. W. Hudnut, V. Kasemsri,
R. Keegan, K. Kovach, L. S. Lenahan, H. H. Ma, J.
J. Rushanan, D. Sklar, T. A. Stansell, C. C. Wang,
and S. K. Yi, Description of the L1C Signal, in
Proceedings of the 19th International Technical
Meeting of the Satellite Division of The Insti-
tute of Navigation (ION GNSS 2006), Fort Worth,
Texas, USA, 2006, pp. 2080 2091
[3] Curran, J. T., Weak Signal Digital GNSS Track-
ing Algorithms, Ph.D. thesis, National University
of Ireland, Cork, 2010
[4] Foucras, M., (2012) O. Julien, C. Macabiau,
and B. Ekambi, A Novel Computationally Ef-
cient Galileo E1 OS Acquisition Method for GNSS
Software Receiver, in Proceedings of the 25th
International Technical Meeting of The Satellite
Division of the Institute of Navigation (ION GNSS
2012), Nashville, TN, USA, 2012, pp. 365 383
[5]Foucras, M., (2013) and O. Julien, C. Maca-
biau, and B. Ekambi, Probability of Secondary
Code Acquisition for Multi-Component GNSS
Signals, in Proceedings of the 6th European
Workshop on GNSS Signals and Signal Process-
ing (SIGNALS 2013), Neubiberg, Germany, 2013
[6] Foucras, M., (2014a) O. Julien, C. Macabiau,
B. Ekambi, and F. Bacard, Probability of Detec-
tion for GNSS Signals with Sign Transitions, IEEE
Transactions in Aerospace Electronic Systems,
submitted July 2014
[7] Foucras, M., (2014b) O. Julien, C. Macabiau,
B. Ekambi, and F. Bacard, Optimal GNSS Acqui-
sition Parameters when Considering Bit Transi-
tions, in Proceedings of IEEE/ION PLANS 2014,
Monterey, CA, USA, 2014
[8] Foucras, M., (2014c) O. Julien, C. Macabiau,
and B. Ekambi, Detailed Analysis of the Impact
of the Code Doppler on the Acquisition Perfor-
mance of New GNSS Signals, in Proceedings of
the 2014 International Technical Meeting of
The Institute of Navigation, San Diego, CA, USA,
2014, pp. 513 524
WORKING PAPERS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
[9] Foucras, M., (2014d) and U. Ngayap, J. Y. Li, O.
Julien, C. Macabiau, and B. Ekambi, Performance
Study of FLL Schemes for a Successful Acquisition-
to-Tracking Transition, in Proceedings of IEEE/ION
PLANS 2014, Monterey, California, USA, 2014.
[10] Jiao, X., and J. Wang, and X. Li, High Sensi-
tivity GPS Acquisition Algorithm Based on Code
Doppler Compensation, in IEEE 11th Interna-
tional Conference on Signal Processing (ICSP),
Beijing, China, 2012, pp. 241 245
[11] Kaplan, E. D., and C. Hegarty, Understanding
GPS: Principles and Applications, 2nd edition.
Artech House, 2005
[12]ODriscoll, C., Performance Analysis of the
Parallel Acquisition of Weak GPS Signals, Ph.D.
thesis, National University of Ireland, 2007
[13] Parkinson, B. W., and J. J. Spilker, Global
Positioning System: Theory and Applications,
Progress in Astronautics and Aeronautics., Vol.
I, 1996
[14] Psiaki, M. L., Block Acquisition of Weak
GPS Signals in a Software Receiver, in Proceed-
ings of the 14th International Technical Meet-
ing of the Satellite Division of The Institute of
Navigation (ION GPS 2001), Salt Lake City, UT,
USA, 2001, pp. 2838 2850
[15] RTCA, Inc., Assessment of Radio Frequency
Interference Relevant to the GNSS L1 Frequency
Band RTCA/DO-235B. 13-Mar-2008
[16] Van Diggelen, F. S. T., A-GPS: Assisted GPS,
GNSS, and SBAS, GNSS Technology and Applica-
tions Series. Artech House, 2009
Authors
Myriam Foucras received
her masters degrees in
mathematical engineer-
ing and fundamental
mathematics from the
University of Toulouse.
Since 2011, she has
been a Ph.D. student at the Signal Processing
and Navigation (SIGNAV) research group of the
TELECOM laboratory of Ecole Nationale de
lAviation Civile (ENAC). Funded by ABBIA GNSS
Technologies, in Toulouse, France, her work con-
sists in the development of a GPS/Galileo soft-
ware receiver.
Olivier Julien is the head
of the Signal Processing
and Navigation (SIG-
NAV) research group of
the TELECOM laboratory
of ENAC, in Toulouse,
France. His research
interests are GNSS receiver design, GNSS mul-
tipath and interference mitigation and GNSS
interoperability. He received his engineering
degree in digital communications from ENAC and
his Ph.D. from the Department of Geomatics
Engineering of the University of Calgary, Canada
Christophe Macabiau
graduated as an elec-
tronics engineer from
the ENAC in Toulouse,
France. Since 1994, he
has been working on the
application of satellite
navigation techniques to civil aviation. He
received his Ph.D. and has been in charge of the
TELECOM laboratory of the ENAC since 2011.
Bertrand Ekambi gradu-
ated from the University
of Toulouse with a mas-
ters degree in mathe-
matical engineering.
Since 2000, he has been
involved in the main
European GNSS projects: EGNOS and Galileo. He
is the founder manager of ABBIA GNSS Tech-
nologies, a French small/medium-sized enter-
prise working of the space industry, based in
Toulouse, France
Fayaz Bacard received his
master of science in
engineering, specializ-
ing in electronics and
computer engineering
from Ecole Nationale
Suprieure des Sciences
Appliques et Technologies (ENSSAT) in Lanion.
Since 2013, he has been a software engineer at
ABBIA GNSS Technologies.
Prof.-Dr. Gnter Hein
serves as the editor of
the Working Papers col-
umn. He is the head of
the EGNOS and GNSS
Evol ut i on Pr ogr am
Department of the Euro-
pean Space Agency. Previously, he was a full pro-
fessor and director of the Institute of Geodesy and
Navigation at the Universitt der Bundeswehr
Mnchen. In 2002, he received the Johannes
Kepler Award from the U.S. Institute of Navigation
(ION) for sustained and significant contribu-
tions to satellite navigation. He is one of the
inventors of the CBOC signal.
READ IT FIRST IN...
E-news and analysis
twice a month,
with exclusives
by Washington
correspondent
Dee Ann Divis and
Inside GNSS editor
Glen Gibbons
Subscribe now
www.insidegnss.com/
enews
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_______
A
series of ABI Research studies
released in recent months sketches
a promising future of robust growth in
GNSS OEM markets. .
Te London- and New Yorkbased
companys 2014 GNSS IC vendor
matrix released on June 30 concluded
that Qualcomm is the leading GPS IC
vendor, followed by Broadcom, and
for the frst time MediaTek in third
place afer another year of strong growth
and robust shipments as a result of its
targeted design strategy.
ABI Research predicts that the GNSS
IC market will break $2 billon by 2016,
with a host of secondary markets start-
ing to emerge. Senior analyst, Patrick
Connolly says that the firm expects
that consumer GNSS IC OEMs adopt-
ing technologies the produce sub-meter
accuracy, such as u-bloxs take on precise
point positioning (PPP) will open up
markets around unmanned aerial vehi-
cles (UAVs), machine control, timing
and synchronization, advanced driver
assistance systems (ADASs), driverless
cars, and other applications. As a result,
total GNSS IC revenues are forecast to
hit $2.75 billion in 2019, he says.
u-bloxs PPP technology uses carrier
phase tracking and ionospheric cor-
rection data from satellite-based aug-
mentation systems, such as WAAS and
EGNOS, and has been implemented in
the companys single-frequency NEO-7P
GNSS module.
INDUSTRY VIEW
Studies Predict 2019 GNSS
OEM Market to Hit $2.75 Billion
360 Degrees Communications Act
Continued from page 21.
receivers would have to meet as a way to limit interference
and expand the number of potential users.
Standards could stife innovation, wrote the Alliance and,
in the case of GPS, where the receivers can be baseball sized
or tiny enough to ft in a watch, engineering a solution for all
the receivers may be impractical.
One issue the committee has raised, but the Alliance did
not address, was changing the role of the National Telecom-
munications and Information Administration (NTIA). CTIA,
the association for the wireless industry suggested that the
FCC be put completely in charge of spectrum decisions.
CTIA recommends Congress consider changing NTIAs
role so that, consistent with national security concerns, spec-
trum use decisions are all made by the FCC, the association
wrote in its comments. Te NTIA would still play a crucial
function as an advisor to federal agencies, CTIA suggested,
and would be the organization that would request spectrum
from the FCC for federal users.
Changing the role of the NTIA could leave the GPS com-
munity with one less empowered potential advocate at the
table. Te organization played a key role in debate over Light-
Squared and in the decision not to go forward despite strong
support for the project in the White House and then-FCC
Chairman Julius Genachowski.
Given the momentum behind the project at the time, it
is not clear what would have happened if the FCC had been
able to proceed without the moderating efect of having to
work with another agency that was able to organize additional,
independent technical experts.
Politics versus Technology
Political idealism is what got us into this mess in the frst
place, rather than taking a hardnosed engineering view, said
Tim Farrar of TMF Associates, a consulting frm that closely
follows mobile communications industry.
Politics of a diferent sort are likely to add momentum to
the lawmakersupdate efort.
Upton took over the committee chairmanship in 2011.
Given the current six-year term-limit rules for House Repub-
licans, this means he will have to give up his committee lead-
ership during 2017. Walden does not appear to be in the best
position to assume the chair at least four other committee
members have more seniority. Although the next chairman
could be supportive, Upton and Walden will likely want to get
as much done as possible before the handover.
And Walden has his own reasons to be proactive. He is
not only the subcommittee chairman; he chairs the National
Republican Congressional Committee (NRCC), a group dedi-
cated to electing Republicans to the U.S. House. As a member
of the Republican leadership, he has input into the Houses
agenda and can be an advocate for overhauling the law. But
as the head of the NRCC, he will also be paying attention to
fundraising and political messaging. More commercial spec-
trum is broadly touted as a way to generate jobs and taking
action to support job growth is only smart during this election
year and the next.
Also smart is tackling issues bound to capture the atten-
tion and resources of well-heeled industries. According
to the watchdog group MapLight, which monitors money in
politics, the telephone utilities made more than $7.3 million
in campaign contributions during 2012 and 2013 while the
cellular systems and equipment frms dropped in more than
$2 million and the cable and satellite TV production and dis-
tribution wrote checks worth more than $7.7 million. Accord-
ing to Oregonlive.com and MapLight, Walden received more
than $109,000 from the cable companies, the highest amount
in Congress from that group for the two-year period.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
As single-purpose portable naviga-
tion device (PND) shipments continue
to fall, GNSS OEMs can expect new
opportunities in cycling, eyewear, golf,
and personal tracking to help the over-
all market to continue to grow. Te total
market for dedicated GPS devices in
2013 was estimated at 35 million units,
which is expected to grow almost 30
percent over the forecast period, despite
accelerated decline in Western PND
markets.
ABI Vice-President and practice
director Dominique Bonte adds, From
an automotive perspective, the connect-
ed car, infotainment, insurance, telemat-
ics/diagnostics, safety, will all increase
adoption of GNSS. We can expect to see
cars fitted with multiple GNSS ICs in
the future. Bonte says strong signals
suggest that some GNSS systems, such
as Galileo and GLONASS, will be man-
dated for regional automotive applica-
tions such as emergency calling.
ION GNSS+, IGM
Partner on Show Pub
T
he Institute of Navigation is partner-
ing with the publisher of Inside GNSS
magazine (IGM) to produce, publish, and
distribute an expanded Show Daily pub-
lication at this years ION GNSS+ 2014
event taking place September 912 in
Tampa, Florida. Te alliance will leverage
the magazines many years of publishing
expertise to advance the business interests
of exhibitors and attendees.
Inside GNSS will expand the Show
Daily offering to include richer prod-
ucts and company news sections as well
as insider previews on content tracks
planned for this years conference. Inside
GNSS will also produce a digital version
that will be sent to more than 20,000
GNSS professionals around the world.
ION Executive Director Lisa Beaty
comments, ION GNSS exhibitors,
advertisers and conference attendees will
have much to gain with the expanded
format, reach and in-depth reportage of
the new Show Daily. ION is committed
to bridging the divide between science
and business, academia and industry,
and we believe that Inside GNSS, led by
Glen Gibbons and Richard Fischer, can
help us achieve that.
Inside GNSS, which reaches 30,000
readers worldwide with its print edi-
tion and thousands more with its digital
version and twice-monthly GNSS SIG-
NALS e-newsletter, is in its ninth year of
publication. Before taking on expanded
responsibility for this years production,
Gibbons Media & Research LLC, pub-
lisher of the magazine, provided editorial
support for the ION show daily for the
past eight years
Companies interested in having a
presence in the ION GNSS+ Show Daily
may contact Fischer by e-mail <rich-
ard@insidegnss.com> or phone 1-732-
741-1964 (office) or 1-609-240-1590
(mobile).
P R E C I S E T I M E A N D T I M E I N T E R V A L M E E T I N G
SEAPDRT BDSTDN HDTEL BDSTDN, ASSACHUSETTS

TECHNtCAL EETtNG: DEC 2 ~ 4, 2014 TUTDRtALS DEC 1, 2014 SHDW DATES: DEC 2-3, 2014
Exhibit booths available.
ION PTTI 2014 SESSION TOPICS
Advanced Atomic Frequency Standards Applications | Advanced Clocks | Enhancing Resilience of Timing and Critical
Infrastructure | GNSS Present and Future | Industrial, Commercial and Military PTTI Systems, Applications and Technologies
| New Commercial Products for PTTI Systems | Precise Networking Timing Standards, Requirements and Applications |
PTTI Systems Calibration | Signals of Opportunity | Space PTTI Applications | Time and Frequency Laboratory Activities and
Updates | Time Scales, Algorithms and Methods | Traditional and Alternate Time and Frequency Transfer Methods
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
___
______________
See additional listings at
WWW.INSIDEGNSS.COM/EVENTS
ADVERTISERS INDEX
Company Page Number
Antcom 18
CAST Navigation 11
ComNav Technology Ltd. 51
Forsberg 41
GPS Networking 35
Honeywell 25
IFEN 13
Intergeo 34
ION GNSS+ 2014 47
ION PTTI 2014 81
KVH Industries 3
LabSat 8
Locata 29
NavCom Inside Back Cover
NavtechGPS 42, 77
NovAtel, Inc. Back Cover, 23, 3233
Rohde & Schwarz Inside Front Cover
Sensonor 1617, 19
Septentrio 43
Spectracom 45
Spirent 67
Terrastar 21
Thales 27
Topcon 9
Trimble 5, 39, 59
GNSS TIMELINE
September
SEPTEMBER 1-3
ARAB INSTITUTE OF NAVIGATION (AIN)
MELAHA CONFERENCE 2014
Alexandria, Egypt
Resilience Navigation is the theme of the 2014
Arab Institute of Navigation Conference and
Exhibition, MELAHA. <http://www.ainegypt.org/>
SEPTEMBER 8-9
CGSIC MEETING
Tampa, Florida USA
The 54th meeting of the Civil GPS Service Interface
Committee features in-depth updates on GPS
from USAF, DoT, Coast Guard, State Department,
Homeland Security and others. It is free and open
to the public. <gps.gov/cgsic/meetings/2014/>
SEPTEMBER 9-12
ION GNSS+ 2014
Tampa, Florida USA
The 27th international technical meeting of the
Institute of Navigation features a plenary session
with explorer Tristan Gooley, plus 5-minute
lightning talks with GNSS leaders on whats hot -
whats not. <ion.org/gnss/>
October
OCTOBER 7-9
INTERGEO 2014
Berlin, Germany
Geodesy, Geoinformation and Land Management
trade fair plus conference. It attracts 16,000
visitors from 92 countries who work in the
surveying, geoinformation, remote sensing and
photogrammetry elds. The conference language is
German. <intergeo.de/intergeo-en/>
OCTOBER 8
ERA-GLONASS 2014
Moscow, Russia
Government and industry event on current status
and future plans for the State Accident Emergency
Response System. Sponsored by GLONASS Union, a
consortium of Russias navigation and information
services companies. <http://congress-era-
glonass.com/>
OCTOBER 911
ASIA OCEANIA REGIONAL WORKSHOP ON
GNSS
Phuket, Thailand
Workshop on Multi-Asia GNSS (MGA) collaborative
experiments using multiple GNSS constellations
in the region. The main organizer is JAXA, the
Japanese Space Agency. <multignss.asia/
workshop.html>
OCTOBER 21-24
ISGNSS 2014/KGS 2014
Jeju, Korea
The International Symposium on Global Satellite
Navigation Systems and Korea GNSS Society
Conference will be held jointly. The theme is Cloud
PNT in IoT. <http://isgnss2014.org/>
November
NOVEMBER 3-5
TRIMBLE DIMENSIONS 2014
Las Vegas, Nevada USA
The Trimble International User Conference covers
products using GPS technology - among many
others- developed and sold by its network of
companies. <trimbledimensions.com/>
NOVEMBER 9-14
ICG-9
Prague, Czech Republic
Ninth Meeting of the International Committee
on GNSS, a UN-backed group of GNSS and
augmentation provider countries. <oosa.unvienna.
org/oosa/en/SAP/gnss/icg/meetings.html>
NOVEMBER 2021
UPINLBS 2014
Corpus Christi, Texas USA
The third IEEE international conference on
Ubiquitous Positioning, Indoor Navigation
and Location-Based Services will concentrate
on innovative, state-of-the-art solutions and
techniques that provide PNT capability anywhere,
anytime. <http://upinlbs.tamucc.edu/>
December
DECEMBER 14
PTTI 2014
Boston, Massachusetts USA
The 46th systems and applications meeting for
Precise Time and Time Interval managers, system
engineers and program planners, will be held at the
The Seaport Hotel in Boston. <http://www.ion.org/
ptti/index.cfm>
DECEMBER 15
UN/ ICG WORKSHOP: GNSS FOR
SCIENTIFIC APPLICATIONS
Trieste, Italy
A weeklong workshop on GNSS and its scientic
applications in low-latitude regions of the world
will be held in Trieste, Italy at the Abdus Salam
International Centre for Theoretical Physics (ICTP).
<http://www.oosa.unvienna.org/oosa/en/SAP/
act2014/trieste-gnss/index.html>
DECEMBER 35
NAVITEC 2014: ESA WORKSHOP ON
SATELLITE NAVIGATION TECHNOLOGIES
AND GNSS SIGNALS AND SIGNAL
PROCESSING
Noordwijk, The Netherlands
Era of Galileo IOV is the theme of the 7th ESA
Workshop on Satellite Navigation Technologies.
< http://www.congrexprojects.com/2014-
events/14c12/introduction>
DECEMBER 811
NAVTECHGPS WINTER GNSS TRAINING
San Diego, California, USA
NavtechGPS Winter GNSS Training 2014 will take
place at the Bay Club Hotel and Marina in San
Diego, California. <http://www.navtechgps.com/
events/location/>
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
_______________________
NovAtel CORRECT
with TerraStar
Decimetre-level positioning. Anywhere, anytime.
No local RTK base station or support infrastructure required.
LIMITED TIME OFFER
Trade in your legacy NovAtel hardware for
a FlexPak6
, antenna and 1-year TerraStar

subscription for only $4,995.00
OR
Current OEM6
customers, buy a 1-year

TerraStar subscription and get your
authorization code upgrade for free
Call: 1-800-NovAtel (1-800-668-2835) E-mail: sales@novatel.com
Visit NovAtel.com/terrastar-promo for more information.
q
M
M
q
q
M
M
qM
Qmags
IGNSS
q
M
M
q
q
M
M
qM
Qmags
IGNSS
_____________

Inside GNSS Magazine: Off The Shelf and Into Space, Volume 9 Number 4 July/August 2014

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Inside GNSS Magazine: Off The Shelf and Into Space, Volume 9 Number 4 July/August 2014

Uploaded by

Copyright:

Available Formats

July/August 2014 | Engineering Solutions from the Global Navigation Satellite System Community | www.insidegnss.

346: GPS/GNSS Operations for Engineers and Technical

541: Using Advanced GPS/GNSS Signals and Systems.

336: GPS Fundamentals, Enhancements and Intro to

L1/L2/L5/E5a/E5b/E5 AltBOC code/carrier tracking

Patented Galileo AltBOC tracking

APME, RAIM and LOCK+ tracking

SEAPDRT BDSTDN HDTEL BDSTDN, ASSACHUSETTS

, antenna and 1-year TerraStar

customers, buy a 1-year

You might also like