gnss_20090501_may_2009

INSIDE GNSS DIGITAL

The print edition

Inside GNSS

plus

READ

REVIEW

EXPLORE

move

enlarge

Search

download

the first thorough analyses of the new L5 signal and initial observations of SVN49 signal

problems

expanded versions of two important articles:

Working Papers, Architecture for a Future C-Band/L-Band GNSS

Synthetic Aperture GPS Signal Processing: Concept and Feasibility Demonstration

live links to sources, references, companies and more.

The following are a few tips for navigating the digital edition once you've opened it:

To from page to page, use the forward and back buttons in the toolbar at the top of the

screen.

To text for easier reading, click once anywhere on the page. Click again to bring the text

back to its original size.

The function allows you to conduct a keyword search within the entire issue, or across

all available issues.

For offline reading, a fully functional copy of the digital edition to your hard drive by

selecting the Save function from the toolbar at the top of the screen.

We encourage you to forward this email to friends and colleagues with whom you would like to share

this digital edition.

If you are viewing the Web browser version, just click the Share function on the toolbar.

Sincerely,

Glen Gibbons

Editor and Publisher



IGNSS



IGNSS

GPS | GALILEO | GLONASS | COMPASSEngineering Solutions from the Global Navigation Satellite System Community

May/June 2009 www.insidegnss.com

GNSS SOLUTIONS:Vector Tracking Loops Benets & Drawbacks

WORKING PAPERS:Can C-Band Work in the GNSS World?

L5 FIRST LOOKS:Two Views of the New GPS Civil Signal

GPS SARMaking the Invisible, Visible

Contents | Zoom in | Zoom out Search Issue | Next PageFor navigation instructions please click hereFor navigation instructions please click here

Contents | Zoom in | Zoom out Search Issue | Next PageFor navigation instructions please click hereFor navigation instructions please click here



IGNSS



IGNSS

_________________

_________



IGNSS



IGNSS

4 InsideGNSS M A Y / J U N E 2 0 0 9 www.insidegnss.com

TECHNICAL ARTICLES

TABLE OF CONTENTS

MAY/JUNE 2009 VOLUME 4NUMBER 3

TOC BY THENUMBERS

8 Thinking Aloud

10 360 Degrees

13 GNSS Hotspots

ARTICLES

16 GNSS SolutionsVector Tracking Loops

22 First Look: GPS L5 Signal

30 GPS Modernization Milestone: L5 on the Air

37 Synthetic Aperture GPS

47 Working PapersC-Band for GNSS: Part 1

DEPARTMENTS

57 Industry View

58 Advertisers Index

58 GNSS Timeline

COVER STORY

37 Synthetic Aperture GPS Signal Processing

Concept and Feasibility DemonstrationAndrey Soloviev, Frank van Graas, Sanjeev Gunawardena, and Mikel Miller

In addition to positioning, navigation and tim-ing, GPS turns out to be a useful sensor for other purposes. Synthetically generated, phased-array antennas can process GPS signals to create large antenna apertures. The resulting narrow-beam generation capabilities mitigate interfer-ence and jamming and produce high-resolution radar images which could lead to interesting civil and military applications.

DoD

Phot

o by

MSG

T Pau

l N. H

ayas

hi

22 First LookObserving the GPS L5 Test Transmission from SVN49 Using Software Radio ProcessingSanjeev Gunawardena, Zhen Zhu, and Michael Braasch

The test signal from SVN49, the latest Block IIR-M GPS satellite, has different characteristics than the operational L5 signal specications. A group of Ohio University researchers discovered the discrepancies when they tracked the new L5 signal using a commercial antenna and a modi-ed software GPS receiver of their own design.

30 Modernization Milestone

Observing the First GPS Satellite with an L5 PayloadGrace Xingxin Gao, Liang Heng, David De Lorenzo, Sherman Lo, Dennis Akos, Alan Chen, Todd Walter, Per Enge, Bradford Parkinson

Using a modied navigation payload on a GPS Block IIR-M satellite, the U.S. Air Force switched on the eagerly awaited, modernized L5 signal on April 10. Researchers in California and Colorado report their Initial observations of L5 broad-casts, including an anomaly in SVN49s L1 signal.



IGNSS



IGNSS



IGNSS



IGNSS

________________

ENGINEERING SOLUTIONS FROM THE GLOBAL NAVIGATION SATELLITE SYSTEM COMMUNIT Y

May/June 2009 Volume 4/Number 3

EDITORIALEditor & Publisher Glen Gibbons [email protected]

Art Director Tim JordanGraphic Artist Gwen Rhoads

Circulation Director Peggie KegelContributing Editor for Working Papers: Gnter Hein

[email protected]

Contributing Editor for GNSS Solutions: Mark Petovello [email protected]

Technical Editor Hans J. Kunze [email protected] Writers/Copyeditors Eliza Schmidkunz,

Melody Ward LeslieWeb Designer/Developer Mike Lee

Web Editor Sierra RobinsonIT Technical Support Elijah Buck

Circulation Assistant Anna Liv Gibbons

MARKETING AND PUBLIC RELATIONSDirector/Partner Eliza Schmidkunz [email protected]

[email protected]

Telephone: 408-216-7561 Fax: 408-216-7525

PUBLISHED BY GIBBONS MEDIA & RESEARCH1574 Coburg Road No. 233

Eugene, Oregon, 97401-4802 USATelephone: 408-216-7561

Fax: 408-216-7525

Copyright 2009 Gibbons Media & Research LLC. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical (including by photocopy, recording, or information storage and retrieval), without permission in writing from Gibbons Media & Research. Authorization is granted to photocopy items, with attribution, for internal/educational or personal non-commercial use. For all other uses, contact Glen Gibbons.

INSIDE GNSS (ISSN 1559-503X) is a controlled circulation magazine, published six times a year. Inside GNSS is a registered trademark of Gibbons Media and Research LLC. Postage paid at Lebanon Junction MPO, KY 40150-9998, Mail Permit #473. INSIDE GNSS does not verify any claims or other information in any of the advertisements or technical articles contained in the publication and cannot take responsibility for any losses or other damages incurred by readers in reliance on such content.

Editorial Advisory Council

VIDAL ASHKENAZINottingham Scientic Ltd., Nottingham, United Kingdom

JOHN BETZMITRE Corporation, Bedford, Massachusetts, USA

PASCAL CAMPAGNEFrance Developpement Conseil, Vincennes, France

MARIO CAPORALEItalian Space Agency, Rome, Italy

PER ENGEStanford University, Palo Alto, California, USA

MARCO FALCONEEuropean Space Agency, Noordwijk, The Netherlands

SERGIO GRECOThales Alenia Space, Rome, Italy

JEAN-LUC ISSLERCNES, Toulouse, France

CHANGDON KEESeoul National University, Seoul, Korea

MIKHAIL KRASILSHCHIKOVMoscow Aviation Institute, Moscow, Russia

SANG JEONG LEEChungnam National University, Daejon, Korea

JULES MCNEFFOverlook Systems Technologies, Inc., Vienna, Virginia, USA

PRATAP MISRAMITRE Corporation, Cambridge, Massachusetts, USA

BRAD PARKINSONStanford University, Palo Alto, California, USA

TONY PRATTProfessor and Consultant, United Kingdom

SERGEY G. REVNIVYKHFederal Space Agency, Korolyov, Russian Federation

MARTIN RIPPLEThales ATM, Melbourne, Australia

CHRIS RIZOSUniversity of New South Wales, Sydney, Australia

TOM STANSELLStansell Consulting, Rancho Palos Verdes, California, USA

RAYMOND J. SWIDEROfce of the Assistant Secretary of Defense for Network and Information Integration , Washington D.C. USA

A.J. VAN DIERENDONCKAJ Systems, Los Altos, California, USA

JRN TJADENEuropean Space Agency, Korou, French Guiana

FRANTISEK VEJRAZKACzech Technical University, Prague, Czech Republic

PHIL WARDNavward Consulting. Garland, Texas, USA

CHRISTOPHER WILSONTele Atlas, Redwood City, California, USA

LINYUAN XIASun Yat-Sen University, Guangzhou, China

AKIO YASUDATokyo University of Marine Science and Technology, Tokyo, Japan

Subscribe OnlineFREE 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. Plus change your address, renew, or unsubscribe do it all online:

WWW.INSIDEGNSS.COM/SUBSCRIPTIONSERVICES



IGNSS



IGNSS

_____________

______________________

__________________

_______________

______________

______________

____________________________

www.insidegnss.com M A Y / J U N E 2 0 0 9 InsideGNSS 7

16 GNSSSolutionsWhat About Vector Tracking Loops?Mark Petovello with Matthew Lashley and David Bevly

47 WorkingPapers ArchitectureforaFuture

C-Band/L-BandGNSSPart 1: C-band Services, Space and Ground

Segments,Overall Performance

Andreas Schmitz-Peiffer, Lars Stopfkuchen, Jean-Jacques Floch, Antonio Fernandez, Rolf Jorgensen, Bernd Eissfeller, Jose Angel Rodri-guez, Stefan Wallner, Jong-Hoon Won, Marco Anghileri, Berthold Lankl, Torben Schler, Oliver Balbach, and Enrico Colzi

8 Thinking AloudInection PointsGlen Gibbons

10 360 DegreesNews from the world of GNSS

SVN49 & L5: Mixed ResultsGalileo: GSA Renamed, Contract TalksCompass Gets New GEO SVGAO Warns of GPS Satellite Gap

57 Industry View58 Advertisers Index 58 GNSS Timeline

Calendar of Events

COLUMNS DEPARTMENTS

TABLE OF CONTENTS

Is C-band in GNSSs Future? See Working Papers on page 47



IGNSS



IGNSS

_____________

________________


Ihave never been one who thought there was a fundamental contradiction between physics and metaphysics.Arguably, an understanding of

Newtonian (and Einsteinian) laws of motion (and relativity) may lead inevitably to a younger person giving up her seat on the bus to an elder but probably most of us did not arrive at that sensibility (if, indeed, we have) by the route of science.

Nothing about the Big Bang or evolution intrinsically stands in opposition to the creation narratives of faith only a failure of imagination on either side can make these ideas antagonistic. Now, as to human interpretations of scientific and religious concepts well, thats another matter entirely.

Nonetheless, as long as we accept the sound science of the former and the metaphors and poetry of the latter with tolerance, curiosity, and, when necessary, a modest element of humor, physics and metaphysics can live and thrive together and learn from each other.

For instance, in the 20th century a movement arose known as ecumenism, from the Greek oikoumene, for the whole inhabited world. It was, first, a search for common ground among factions within a particular faith, but later expanded to seek cooperation and mutual understanding among faiths.

Eventually, the idea has arisen of a tolerant primacy of ones own faith in a multicultural world.

But ecumenism is far from a done deal. Crusade and jihad, apocalypse and rapture are much more thrilling prospects than cooperation, it seems. Defining who we are by sketching the shadows of strangers is often easier than filling in the outline of our own qualities and aspirations on which to build the basis of mutual interests.

And thats where GNSS enters into this little homily.

The world now has four global navigation satellite systems in existence or struggling to be born. (If reports that India now plans to expand its regional efforts into a full GNSS, make that five . . . and counting.) All of the providers have joined that most ecumenical of organizations, the UN-backed International Committee on GNSS (ICG), dedicated to the proposition that the systems should be compatible and interoperable.

But, as frequently occurs in the realm of religions, each GNSS provider continues to exhibit nationalist tendencies as well. Recent GNSS examples: even as his country prepares to host the fourth meeting of the ICG, Roskosmos chief Anatoly Perminov

is making the argument that all cars imported to Russia should have GLONASS navigation systems.

Chinas Bureau of Surveying reminds (foreign) cell phone owners that the unlicensed GPS function should be turned off to avoid running up against the agencys crackdown on illegal mapping. Europe would like to tax GNSS chips but has still not finalized an ICD for its Open Service that would allow manufacturers to build receivers.

As for the United States, the original GNSS operator with a complete system and enormous installed base, it sometimes only has to not step up to some issue to advance its particular interests.

The search for unique qualities, secure signals, and comparative

advantages vis--vis the other systems must coexist beside the demand of rapidly growing numbers of GNSS users to have interoperable systems.

Otherwise, well have no way to deal with such things as the trend toward fielding higher power GNSS signals that raise the noise floor in limited L-band frequencies like water

from a broken pipe flooding the basement.

Each system faces a series of internal challenges and inflection points, where providers must decide to turn one way or another in their GNSS program development. And they all have the same external inflection point: protectionism or free trade, dominance and exclusivity or primacy with mutual benefit.

GLEN GIBBONS, JR.Editor

Dening who we are by sketching the shadows of strangers isoften easier than lling in the outline of our own qualities andaspirations on which to build the basis of mutual interests.

THINKING ALOUD

Inection Points



IGNSS



IGNSS



IGNSS



IGNSS

____________

____________


SVN49 & L5: Mixed Results

Want the good news first, or the bad?Well, in the order they occurred, the good news is that a successful launch of a Block IIR-M(20) satellite on March 24 and the beginning of transmissions have secured the primacy of the GPS civil L5 signal centered at 1176.45 MHz.

L5, the third civil GPS signal, will eventually support safety-of-life applica-tions for aviation and provide improved availability and accuracy to users. The GPS program had faced an August 26, 2009, deadline in order to secure its filing for the frequency with the International Telecommunications Union (ITU).

The bad news? Signal anomalies characterized by the U.S. Air Force as out of family transmissions will

keep the latest GPS satellite from being declared healthy for an indeterminate amount of time.

News of the problem with IIR-M(20) now that its on orbit, known as space vehicle number (SVN)49 was announced May 4 to the European Navi-gation Conference in Naples, Italy, by the GPS Wings chief engineer, Col. (Select) David Goldstein.

Although a healthy L5 signal began transmitting on April 10, other GPS signals being broadcast by the satellite particularly those at the L1 frequency are demonstrating larger than expect-ed pseudorange errors that appear to be elevation-dependent, Goldstein said. That is, they vary with the varying eleva-tion angle of the satellite as it rises and sets in the sky.

(Articles by two teams of research-ers appear in this issue of Inside GNSS, beginning on page 22 and page 30, and confirm the phenomena.)

The Air Force detected the anomalies

on April 9, the day before the L5 broad-casts began. That suggests the out-of-family problem may stem from a fun-damental mechanical difficulty, which cant be fixed now that the spacecraft is in orbit. However, the GPS Wing has a large military/civilian team of satellite experts working on the issue.

Goldstein predicted that several more months would pass before we finish all of the trouble-shooting.

Because 18 other Block IIR satellites built by Lockheed Martin are on orbit and performing well, a substantial like-lihood also exists that something about the L5 payload is creating the problem.

The problem poses no immediate problem for GPS users. Thirty GPS sat-ellites are currently in orbit and broad-casting healthy signals; however, the full operational constellation for GPS only requires 24 satellites.

Were in no hurry because of our overpopulated constellation, Goldstein told the ENC audience. SVN49 official-

360 DEGREES



IGNSS



IGNSS

______________________ ___________________________


ly remains in the early orbit check-out phase. However, the Air Force har-vested the pseudorandom noise (PRN) code number (PRN01) of a GPS satellite already in orbit to use on SVN49. That reduced the constellation by one.

The uncertainty created by the signal anomalies will probably delay launches of the last modernized Block IIR satel-lite, now scheduled for August, and the first next-generation Block IIF, which had been expected to go up late this year or early in 2010.

Galileo: GSA Reorganized, Contract Talks

Against the backdrop of nego-tiations to build and deploy the Galileo system, a proposal now before the European Par-liament and Council of the European Union would complete the transforma-tion of the European GNSS Supervisory Authority (GSA) from the leading execu-tive agency for the Galileo program into a diminished subsidiary of the European Commission (EC).

That pre-eminent role, envisioned under the strategy of a public-private partnership (PPP) abandoned more than two years ago, would have seen the GSA sign and oversee a contract with a private consortium building and operat-ing the Galileo system and its precursor European Geostationary Navigation Overlay Service (EGNOS).

Instead, under the terms of EC Com-munication 139 released March 24, the GSA would be renamed the GNSS Agen-cy with the EC holding veto power over its administrative board and the agen-cys primary mission reduced to market research and promotion of Galileo as well as conducting security audits.

On April 1, ownership of the EGNOS infrastructure was transferred to the EC, and a Brussels-based company, the European Satellite Services Provider (ESSP SaS), was entrusted with opera-tion of the system.

Buying and Building a GNSS. Galileo is

now being developed as a fully public procurement with a 3.4 billion budget. The European Space Agency (ESA) is acting as the technical design authority and prime contractor.

The EC is currently negotiating with a short list of 11 industry teams compet-ing for prime contracts in six so-called work packages (WPs) to build the fully operational capability (FOC) Galileo

system. ESA and the EC say that an FOC constellation of 30 satellites is still planned, including four in-orbit valida-tion (IOV) spacecraft, although some participants have called for a reduced set of satellites and services.

Its not just about sending up a bunch of satellites and building a ground infrastructure, said Michel Bosco, an EC representative who spoke to the Euro-



IGNSS



IGNSS

______________________


360 DEGREES

pean Navigation Conference on May 4. We are now moving from infrastruc-ture to applications with the objective of promoting the quickest and broadest uptake of EGNOS and Galileo.

Some work packages are further along than others, with some in the competitive dialog phase with bid-ders and others approaching the best and final offers phase. The system sup-port WP, for example, may be signed this summer, but EC and ESA officials hope to have all contracts signed by the end of 2009.

The negotiations are hard; the play-ers know their jobs, he added. But the signs are positive.

Nonetheless, ESA and industrial sources involved in the process say that the initial bids for WP4, space segment (satellites), were 40 percent over the bud-geted allocations and the offers for all six WPs, several hundred million euros over budget.

ESA reportedly asked the competing WP4 teams, led by EADS Astrium and OHB-System for three options in their proposals: one 28-satellite batch (appar-ently including two spares), two sets (16 and 12) of space vehicles (SVs), and another for sets of 8 and 12. The latter option suggests that ESA might award two contracts of 8 each, ensuring a dual source for the space segment and reserv-ing the option of choosing between the two designs for the final 12 satellites.

The first IOV satellites will begin launching next year from ESAs facility in Korou, French Guiana, using Russian Soyuz rockets. A Soyuz facility is nearing completion in Korou. The launch strat-egy for the FOC satellites will probably use a combination of Ariane 5 launch-ers (four SVs per launch) and Soyuz (two SVs per launch).

Lessons for Leadership. Observers generally believe that the current leaders of the Galileo program are doing better than their predecessors in the EC and the Galileo Joint Undertaking (GJU), an EC/ESA-guided agency from which the GSA took over responsibility in 2007. The GJU failed to complete a contract with the private consortium under the PPP model and has recently come under criticism from the European Court of Auditors.

Albany, New York; Madison, Wisconsin

Laying Down the Law:In May, the New York Court of Appeals ruled 4 to 3 that warrantless GPS surveillance isnt legal. Oregon and Washington courts agree. Meanwhile, a Wisconsin appeals court panel okd secret police use of a GPS tracking device, because it didnt involve search or seizure. Wonder when the Feds will chime in

In focusing on the development and validation phase during 20036, the auditors concluded that the Galileo program is five years behind schedule and facing a current overrun of 2.25 billion (US$3.06 billion) above the 2000 cost projection of 3.33 billion for the definition, development and validation, and deployment phases.

The auditors attribute many of the problems to an unclear mandate and con-flicted governance structure of the GJU, established in September 2003, adding that the EC also failed to provide ade-quate leadership during that phase.

A New Role for the GNSS Agency. The EC characterizes its recent communication as an effort to remove contradictions and ambiguities between the 2004 regulation

360 DEGREES



IGNSS



IGNSS

____________________

_____________


establishing the GSA and a 2008 regu-lation outlining the public procurement for Galileos deployment.

Although the 2008 regulation implic-itly and comprehensively amended the Supervisory Authoritys responsibilities, it had no impact on its internal organisa-tion, and the Commissions influence in this area continues to be very limited, says the March 24 communication.

The communication says the reor-ganized GSA would continue as a Community Agency that is, one under the control of European mem-ber states rather than the EC, the EUs executive branch, but the far-reaching changes that it proposes would pro-foundly alter the agencys political and legal status.

These changes include giving the EC representative on the GSA/GNSS Agen-cys administrative board a vote equal to half of that bodys votes, with the 27 rep-resentatives of the member states having the other half.

Moreover, the term of the agencys executive director would be shortened to four years from the current five and cur-tails that role to carrying out his duties under the supervision of the Admin-istrative Board in accordance with the guidelines provided to the Agency by the Commission.

The communication calls for aboli-tion of the GSAs Scientific and Techni-cal Committee, which had been charged with delivering opinions on technical questions or on proposals involving major

changes in the design of the European GNSS system and making recommenda-tions on modernization of the system.

Those responsibilities have been moved along with many of the GSA technical staff to the EC or to the ESA Evolutions project that is investigating the technical design for a next-genera-tion system with 105 million in fund-ing from ESA members over the next two years.

The ECs proposal would also replace the GSAs System Safety and Security Committee with a Security Accredita-tion Committee for European GNSS Systems, chaired by an EC representative and with members from the EU nations. With oversight from this committee, the GNSS Agency would be charged with

360 DEGREESGNSS Hotspots

Moscow, Russia

Pay to Play. The headof Roscosmos, Russiasspace agency, has askedthe government to makeit prohibitively expensiveto import cars that cantuse GLONASS. Businessnewspaper Vedomosti saidnot many Russian cars havebuilt-in navigation systemsnow, and only 10,000 ofnearly two million importscan use the Russian GNSS.

Trieste, Italy

GNSS for Africa 50scientists from 15 sub-Saharan universitiesconsulted with GNSSexperts and even builtLEGO Mindstorm robots at the rst SatelliteNavigation and Technologyfor Africa workshop in April.Why? GNSS infrastructuremeans better maps, safertransportation, managednatural resources andfood supplies, improvedemergency services majorgoals on the continent.

Washington, DC

Backup The land-basedradio navigation system,Loran-C and its eLoranmodernization, has beencut from the 2010 federalbudget. Key members ofthe Senates HomelandSecurity and Science andTransportation committees worried about the GAOsreport on a falteringGPS question killingan interoperable butindependent PNT backup.

Thiruvananthapuram, India

Ready to Go? In just three years, says the directorof the Vikram Sarabhai Space Centre (VSSC), IndiasRegional Navigation Satellite System (IRNSS) will beup and running, delivering 10 meter accuracy to thesubcontinent using three GEOs and four IGSO satellites.Could be the precursor to a full-edged Indian GNSS.

GLONASS

Compass

Military

Breaking

Policy

SignalLaunch

Other SystemsGalileo

GPS

Bright Idea

Technology

Commercial

Consumer

Satellite

History

Glitch

Conference

Middle Earth Orbit

L5: A Mixed Bag. The GPSsatellite carrying the newsafety-of-life civil signalis under investigation. AnL5 signal transmitted onApril 10 was healthy. Butsignals on the L1 frequencyare not meeting spec. Largerthan expected pseudorangeerrors, says the GPS Wingschief engineer. The L5 signalitself could be a cause.



IGNSS



IGNSS


setting up a Galileo Security Center that would begin operation in 2012.

The new regulation has had its first reading in the European Parliament and been referred to committee.

GAO Warns of Potential GPS Satellite Gap

In what is not necessarily its worst-case scenario, a recent U.S. Government Accountability Office (GAO) report suggests that a two-year delay in the production and launch of the first and all subsequent GPS III satellites would reduce the probability of maintaining a 24-satellite constellation to about 10 per-cent by around fiscal year 2018.

That would be fewer than the mini-mum number of satellites to which the U.S. government has committed for national and international user com-munities. It would also surely diminish the quality of GPS service, particularly for military, safety-critical, and urban applications.

However, 12 months into a 72-month schedule from contract award to first launch, the GPS IIIA program is still on track for a 2014 launch, according to officials at the GPS Wing and Lock-heed Martin, which won the $1.4-billion development and production contract in May 2008.

The GPS Wing, part of the Space & Missile Systems Center (SMC) at Los Angeles Air Force Base, California, suc-cessfully completed an integrated base-line review of the GPS IIIA program last October 31. An overall segment-level preliminary design review (PDR) was scheduled to take place in mid-May.

The U.S. government plans to invest more than $5.8 billion from 2009 through 2013 in GPS space and ground control segments. Nonetheless, in a report issued April 30 under the title, Global Positioning System Sig-nificant Challenges in Sustaining and Upgrading Widely Used Capabilities, Cristina Chaplain, the GAOs director of acquisition and sourcing management

and primary author of the report, indi-cated that considerable risk still exists that the schedule may not be met.

Noting GPS IIIAs highly com-pressed timeline, the GAO report said that the schedule is shorter than most other major space programs we have reviewed, adding no major satellite program undertaken in the past decade has met its scheduled goals.

A two-year delay in the launch schedule would translate into 5 years when the U.S. government would be operating a GPS constellation of fewer than 24 satellites, and a 12-year period during which the government would not meet its commitment to maintaining a constellation of 24 operational GPS sat-ellites with a probability of 95 percent or better.

Even before the 2014 first GPS IIIA launch, however, the GAO warns that a 20 percent chance will arise in 2011-2012 that the constellation could drop below 24 space vehicles (SVs) as older satellites begin failing faster than they can be replaced.

However, the projections of an impending gap are primarily based on design life estimates, not so much on-orbit performance. None of the GPS Block IIR satellites on orbit now, for instance, have failed, even though the first was launched more than a decade ago with a design life of 7.5 years.

Aside from possible launch delays and constellation decline, the GAO report criticized the failure of the GPS program to synchronize the acquisi-tion and development of the next gen-eration of GPS satellites with the corre-sponding timelines of the ground control segment and military user equipment. The result: a likelihood that the mod-ernized military signal (M-code) will be available for more than a decade before user equipment will be fielded that can take strategic advantage of it.

The report attributed the problems to a variety of causes: a bungled acqui-sition reform introduced in the 1990s, turnover in military program leader-ship, diffuse responsibilities for GPS system development, reprogramming of GPS funds to other DoD program (or from GPS control segment and user

equipment programs to backfill cost overruns in the space segment), fielding of immature technologies, post-contract engineering changes, and so on.

Compass Gets New GEO SV

Launch of a second modernized Compass (Beidou 2) satellite on April 14 this one a geostation-ary spacecraft marks the return of China to its launch program two years after the initial venture into space to build a full-fledged GNSS.

Designated Compass G2 reflecting the geostationary nature of its intended orbital position about 22,300 miles above the equator, the satellite lifted off at 16:16 UTC aboard a Long March 3C rocket from the Xichang launch base in south-western Chinas Sichuan province, accord-ing to Chinas state news agency Xinhua.

The satellite is the first of 10 that China has previously announced it plans to launch over the next year and a half. The second Compass satellite was developed by the China Academy of Space Technology, which is part of the China Aerospace Science and Technol-ogy Corporation.

An April 17 news report by Xinhua quoted Cao Chong, chief engineer of the China Electronics Technology Group Corporation, as estimating that the first phase of Compass, scheduled to complete a regional capability by the end next year, would cost more than US$1.46 billion.

I think the Compass system might cost China several dozen billion yuan, said Cao, who works with the China Satellite Navigation Engineering Cen-ter responsible for building the Compass system. The first phase alone could cost more than 10 billion yuan, Cao said.

China launched its first Compass middle-earth-orbiting (MEO) satellite in April 2007, joining several geosta-tionary Beidou-1 satellites that have been launched since 2000. Program officials says that the system will be complete by 2015.

360 DEGREES



IGNSS



IGNSS



IGNSS



IGNSS

____________________

16 InsideGNSS M A Y / J U N E 2 0 0 9 www.insidegnss

GNSS Solutions:

V ector tracking loops are a type of receiver architecture. The difference between traditional receivers and those that use vector tracking algorithms is the man-ner in which they process the received GNSS satellite signals, and how they determine the receivers position and velocity.

Vector-based tracking loops com-bine the two tasks of signal tracking and position/velocity estimation into one algorithm. In contrast, traditional or scalar tracking methods track each satellites signal(s) independently; both of each other and of the position/velocity solution.

Vector tracking has many advan-tages over scalar tracking loops. The most commonly cited advantage is increased immunity to interference and jamming. The minimum carrier-to-noise power density ratio (C/N0) at which the receiver can operate is low-ered by processing the signals in aggre-gate instead of separately.

Vector tracking algorithms also have the ability to bridge signal outages and immediately reacquire blocked sig-nals. Moreover, vector tracking loops have a greater immunity to receiver dynamics than scalar tracking loops.

A final advantage: The vector track-ing architecture allows the receivers motion to be constrained in different dimensions, which can be exploited by receivers whose motion occurs primar-ily in one or two directions, such as ships or automobiles, for example.

The primary drawbacks of vector tracking loops relative to traditional approaches are their processing load and complexity. The Kalman filter used by the vector tracking architecture (more details to follow) must be iterat-ed on a time scale commensurate with

the integrate-and-dump period used by the algorithm (~ 50 Hz). The numeri-cally controlled oscillators (NCOs) in each channel also must be controlled directly by the central Kalman filter.

Another drawback of vector track-ing is that the presence of a fault in one channel will affect all the other chan-nels, possibly leading to receiver insta-bility or loss of lock on all satellites.

Before discussing how vector track-ing loops operate, lets first review how a traditional receiver operates. Figure 1shows a block diagram of a typical GPS receiver.

In the traditional GNSS receiver, scalar tracking loops are used to esti-mate the pseudoranges and pseudor-ange-rates for the available satellites. A delay lock loop (DLL) is generally used for estimating the pseudoranges, and either a Costas loop or frequency lock loop (FLL) is used to estimate the pseu-dorange-rates or carrier Doppler. (A phase lock loop can also be implement-ed, although it is not strictly required for signal tracking).

The pseudoranges and pseudo-range-rates are fed forward to the navigation processor, which solves for the receivers position, velocity, clock bias, and clock drift (i.e., the naviga-tion states). The navigation processor is typically an iterative least squares algorithm or a Kalman filter.

In Figure 1, note that the flow of information in the receiver is strictly from left to right. Each channel of the receiver tracks its respective signal independent of the other channels. In addition, no information from the navigation processor is fed back to the tracking loops.

The only exception to this may occur when the navigation solution is used to initialize the acquisition pro-cess for a particular satellite. Although this may reduce acquisition time, it does not improve the receivers satellite tracking capability.

By its very nature, the traditional receiver architecture does not exploit the inherent relation between signal

GNSS Solutions is a regular column featuring

questions and answers about technical aspects of GNSS. Readers are invited to send their questions to

the columnist, Dr. Mark Petovello, Department of

Geomatics Engineering, University of Calgary, who will nd experts to answer

them. His e-mail address can be found with

his biography at the conclusion of the column.

What are vectortracking loops,

and what aretheir benets and

drawbacks?



IGNSS



IGNSS


tracking and navigation state estima-tion. In particular, recall that the basicconcept of GNSS is that the signaltracking information (i.e., pseudor-anges and pseudorange-rates) can beused to estimate the desired navigationstates (i.e., position, velocity, and clockinformation).

In contrast to traditional receiv-ers, vector tracking algorithms exploitthe inherent coupling between signaltracking and navigation solution com-putation, and combines them into asingle step. In other words, in a vec-tor tracking approach, the navigationprocessor is used to perform both tasksand eliminates the need for intermedi-ate tracking loops.

Figure 2 shows a block diagram ofa receiver employing a vector delay/frequency lock loop (VDFLL). In thisarchitecture, the pseudoranges andpseudorange-rates are predicted by the

navigation processor (in this case anextended Kalman filter (EKF)) for eachsignal that is to be tracked. This predic-tion is performed using the estimatednavigation states and the computedsatellite position and velocity.

Each channel of the receiver thenproduces pseudorange and pseu-dorange-rate residuals (differences)

relative to the predicted pseudorangeand pseudorange-rates. In turn, theEKF uses the residuals to update itsestimates of the receivers navigationstates. In the VDFLL, the vector track-ing loop is closed through the EKF.

For a VDFLL, the typical statesused in the EKF are shown in equation(1).

FIGURE 1 Traditional Receiver Architecture

Antenna

IF SignalPseudoranges,

pseudorange-ratesRF Front-endProcessing

TrackingLoops

Channel 1

Channel 2

Channel j

NavigationProcessor

Position,Velocity,

TimeTracking

Loops

TrackingLoops



IGNSS



IGNSS

_________

________


above a value represent its time derivative). The receivers clock error is denoted as t and the letter c represents the speed of light. The terms ax,j, ay,j, az,j are the elements of a unit vector pointing from the receivers estimated position to the j-th satellite.

Equation (2) is very important in that it shows how the channels of the receiver are coupled. The pseudoranges are tied together through the three position states and one clock bias state. Similarly, the pseudorange-rates are coupled through three velocity states and one clock drift state. The position and velocity states are related to the residuals by the line-of-sight vectors.

We should note that the phase of the received carrier sig-nals can also be tracked using the vector tracking approach. This is referred to as a vector phase lock loop (VPLL). The VPLL requires an alternate formulation of the central EKF due to the fact that the carrier phases of the received signals cannot be predicted unambiguously from the filter states shown in (1).

The VPLL is not as common as the VDLL and VFLL because the carrier frequencies and code phases can be tracked at lower C/N0 ratios than the carrier phases. In gen-eral, vector tracking is used specifically for situations where low C/N0 ratios are encountered.

The advantage of vector tracking over scalar tracking loops stems from the number of unknowns that the two algo-rithms are attempting to estimate, and how the unknowns are related to the available measurements. A traditional receiver uses N scalar DLLs to estimate N pseudoranges. In contrast, a VDLL uses N pseudorange residuals to estimate four states (three position and one clock bias). Similar num-bers apply to the VFLL case as well and are therefore not provided here.

To illustrate this point, consider the situation where Npseudorange residual measurements are available, as shown in (3).

Higher order derivative states can be appended to (1) but are not necessary for the VDFLL to function. The residuals produced in the j-th channel are related to errors in the states of the EKF by equation (2).

In (2), the symbol denotes an error in a state. The receiv-ers Cartesian coordinates are represented by x, y, and z (dots

GNSS SOLUTIONS

FIGURE 2 Vector Tracking Receiver Architecture

Antenna

IF Signal

Pseudoranges,pseudorange-

rateResiduals

PredictedPseudoranges,

pseudorange-rates

RF Front-end

Processing

TrackingLoops

Channel 1

Channel 2

Channel j

ExtendedKalman

Filter

TrackingLoops

TrackingLoops



IGNSS



IGNSS

_____________________________


In this equation, the pseudorange residuals (denotedwith a tilde) are assumed to consist of the true residuals pluswhite noise. In a manner analogous to using scalar DLLs, thepseudoranges are estimated using the equations in (3) withweighted least squares. The weighted least squares estimate ofthe pseudoranges ( ) and associated covariance are shownin (4).

Examining equation (4) reveals an important drawbackof scalar tracking loops. As the number of available pseudo-ranges increases, the variance of the estimated pseudorangesremains constant. This is a direct result of the pseudorangesin (3) being modeled as completely uncoupled.

Now, consider using the N pseudorange residuals to firstestimate three position errors and one clock bias error. Thisis analogous to the VDLL approach. Equation (5) relates theposition and clock errors to the residuals.

The weighted least squares estimate of the vector X andits associated covariance are shown in (6).

The vector X is related back to the estimated pseudor-anges by Equation (7).

Therefore, the covariance of the estimated pseudorangesfrom the vector tracking approach are:



IGNSS



IGNSS

_____________


In other words, the variance of individual pseudoranges is determined by multiplying the appropriate diagonal element of the matrix W by the noise variance .

Comparing the pseudorange covariances in (4) and (8), the vector tracking approach will yield smaller pseudorange variances when the diagonal elements of W are less than one. In the case of four satellites, the pseudorange covariances in (4) and (8) are equal (assuming H has full rank).

In a case where N exceeds four, the pseudorange vari-ances from the vector tracking method in (8) will generally be less than those in (4). This is the main benefit of vector-based tracking.

Time (Hours)

Posi

tion

Dilu

tion

ofPr

ecis

ion

(PDO

P) 14

12

10

8

6

4

2

0

Num

ber o

f Sat

ellit

es

12

10

8

6

4

2

00 2 4 6 8 10 12 14

Time (Hours)

Effe

ctiv

eGai

n in

C/N

0(d

B)

8

7

6

5

4

3

2

1

00 2 4 6 8 10 12 14

Maximum GainMinimum Gain

FIGURE 3 PDOP and Number of Satellites FIGURE 4 Maximum and Minimum Effective Gain from Vector Tracking

GNSS SOLUTIONS



IGNSS



IGNSS

________________


Equation (8) also shows that theperformance of vector tracking is afunction of how many satellites areavailable and their geometry. To deter-mine the relative performance advan-tage of the vector-tracking algorithmfor a typical GPS receiver, the visiblesatellite constellation was recordedevery minute for about 14 hours atAuburn University.

For each satellite geometry, theeffective gain in C/N0 ratio was deter-mined by examining the maximumand minimum diagonal elements ofthe matrix W in (8). A nominal C/N0ratio of 45 dB-Hz was assumed for allof the available satellites. At 45 dB-Hz,the noise variance is 34.1 m2.

The reduction in C/N0 ratio neededto make the largest pseudorange vari-ance equal to 34.1 m2 is defined as theminimum gain in effective C/N0 ratio.Conversely, the reduction in C/N0 rationeeded to make the smallest pseu-dorange variance equal to 34.1 m2 isdefined as the maximum gain in effec-tive C/N0 ratio.

Figure 3 shows the position dilutionof precision (PDOP) and number of vis-ible satellites over the 14-hour period.

The maximum and minimum gainin effective C/N0 ratio over the 14-hourperiod are shown in Figure 4.

The maximum gain in C/N0 ratiovaries from 2 to 6.5 decibels and hasa mean of 5.1 decibels. The minimumgain in C/N0 ratio varies from nearly0 to 2.8 decibels and has a mean of 1.1decibels. Figure 4 demonstrates thatthe vector approach can significantlyimprove a receivers ability to track thereceived signals.

In conclusion, vector trackingalgorithms combine the operations ofsignal tracking and navigation stateestimation. The performance improve-ment brought about by vector trackingis contingent on the number of avail-able satellites and their geometry. Theonly major drawbacks of vector track-ing are their complexity and computa-tional loads.MATTHEW LASHLEY AND DAVID M. BEVLY

Matthew Lashley is aPh.D. candidate atAuburn University. He isan employee ofNavigation TechnologyAssociates, Inc. and amember of the GPS and

Vehicle Dynamics Lab (GAVLAB). His researchinterests are currently vector tracking and deepintegration for GPS receivers.

David M. Bevly received his B.S. from TexasA&M University, an M.S. from MassachusettsInstitute of Technology, and a Ph.D. from

Stanford University inmechanicalengineering. Bevlydirects AuburnUniversitys GPS andVehicle DynamicsLaboratory (GAVLAB),

which focuses on modeling, navigation, andcontrol of vehicles.

SUGGESTED ARTICLES:[1] Benson, D., Interference Benets of a VectorDelay Lock Loop (VDLL) GPS Receiver, in Proceed-ings of the 63rd Annual Meeting of the Institute ofNavigation. Cambridge, Massachusetts, Instituteof Navigation, April 2007[2] Petovello, M., and G. Lachapelle, Comparisonof vector-based software receiver implementa-tions with application to ultra-tight GPS/INS inte-gration, in Proceedings of ION GNSS 2006. FortWorth, Texas, Institute of Navigation, September2006.[3] Spilker, J. J., Fundamentals of Signal Track-ing Theory, in Global Positioning System: Theoryand Applications, Vol. I. Progress in Astronauticsand Aeronautics, Volume 163, AIAA, Washington,D.C., 1996

Mark Petovello is an AssistantProfessor in the Department ofGeomatics Engineering at theUniversity of Calgary. He has beenactively involved in many aspectsof positioning and navigation since1997 including GNSS algorithmdevelopment, inertial navigation,sensor integration, and softwaredevelopment.

Email: [email protected]



IGNSS



IGNSS

________________________________

______________


First Look

Using a commercial antenna and a modied software GPS receiver of their own design, researchers at Ohio University tracked the new L5 signal from the latest Block IIR-M GPS satellite. Among other things, they discovered that the test signal being transmitted from SVN49 has some different characteristics than the signal specications dened for the operational L5 signal.

US A

ir Fo

rce

SANJEEV GUNAWARDENA, ZHEN ZHU, AND MICHAEL BRAASCHAVIONICS ENGINEERING CENTER,OHIO UNIVERSITY


Observing the GPS L5 Test Transmission

from SVN49 Using Software Radio Processing



IGNSS



IGNSS


The much-anticipated first GPS satellite with an L5 test payload was launched from Cape Canaveral on March 24, 2009. On April 10, at approximately 11:58 UTC, the L5 test transmission was turned on by the GPS Control Segment.

This event marked a significant milestone for GPS: 31 years after the launch of NAVSTAR 1 (space vehicle number 1 or SVN01), GPS SVN49 is now transmitting on a completely new, third navigation frequency; something the GPS forefathers probably could never have imagined back in February 22 of 1978 when that Atlas 64F rocket carrying NAVSTAR 1 lifted off the launch pad.

Arguably even more significant is the fact that a U.S. Department of Defense (DoD) program developed primarily for military use during the Cold War era is now transmitting a signal dedicated for civil use; free of any military signal modu-lations.

Times have changed.The GPS L5 transmission, like L1, is allocated in the inter-

nationally protected aeronautical radio navigation services (ARNS) band, clearing one more hurdle for GPS-based dual-frequency systems to be certified for safety-of-life services such as aviation. The L5 transmission also demonstrates the final prong of the three-pronged approach to GPS modernization: transmit stronger signals; implement longer, faster, and more sophisticated pseudorandom noise (PRN) codes; and add fre-quency diversity to make GPS more robust and resistant to interference for civilian and military users alike.

The implementation of the GPS modernization program started in September of 2005 with the launch of the first Block IIR-M satellite SVN53, which gave civil users direct access to L2 for the first time via the L2C code not to mention the sophisticated M-code on L1 and L2 for military users.

Significance aside, an operational GPS L5 constellation wont exist for yet a few more years until a significant number of Block IIF satellites replace the existing constellation. For the GPS Wing, the event of April 10 meant that the US wont lose its International Telecommunications Union (ITU) filing sta-tus for L5, which came uneasily close to an August 26, 2009, deadline due to launch delays.

For the thousands of individuals who comprise the com-munity of researchers and equipment manufacturers that use (or cater to users of) GPS for aviation and other safety-of-life, atmospheric studies, space weather monitoring, RTK, and countless other applications, it means the first-time availability of an unrestricted wideband civil GPS signal-in-space to help bring their research to fruition.

Certainly, GPS researchers at the Ohio University Avion-ics Engineering Center are among these thousands. During the past 18 months we upgraded our instrumentation-quality L1/L2 GPS software receiver to include L5. April 10, 2009 will forever be etched in our minds as that Good Friday we collected terabytes of data and spent an entire gorgeous Easter weekend stuck indoors, unrelentingly processing and processing till we finally saw the L5 correlation peak from GPS SVN49. What follows is our first look of the signal from Athens, Ohio.

GPS L5 Signal StructureThe most complete description of the GPS L5 signal can be found in the interface specification IS-GPS-705, referenced in the Additional Resources section near the end of this article. What follows is a quick summary for the purposes of this article.



IGNSS



IGNSS

____________________


FIRST LOOK AT L5

In general, the L5 transmission received at time t with power level PL5 from the ith GPS satellite can be described as:

Where fL5 is the L5 carrier frequency of 1176.45 MHz, fD is the carrier frequency offset due to satellite line-of-sight motion and satellite-and-receiver combined clock frequency error, L5is a phase offset and n(t) is the received noise.

As shown, the L5 transmission is quadrature BPSK (bi-phase shift keying) modulated, with the in-phase component contain-ing an SV-specific 10,230-chip long PRN code G1 chipping at a rate of 10.23 106 chips/sec, a 10-bit Neumann-Hoffman code NHi at 1000 chips/sec, and a rate 1/2 convolutionencoded C-NAV data stream D5 at 100 symbols/sec.

The quadrature component is modulated with a different 10,230-chip long SV-specific PRN code GQ and a 20-bit Neu-man-Hofman code NHQ with chipping rates the same as the I-channel: CI, CQ, NHI, NHQ, and D5 {1, - 1}, NHI = [1 1 1 1 - 1 - 1 1 - 1 1 - 1}, and NHQ= [1 1 1 1 1 - 1 1 1 - 1 - 1 1 - 1 1 - 1 1 1 - 1 - 1 - 1 1]. The 10.23 MCPS code chipping rate results in a null-to-null bandwidth of 20.46 MHz: identical to the band-width of the legacy P(Y) code modulation on L2.

The Neuman-Hofman codes effectively lengthen the 1-mil-lisecond, periodic L5 PRN codes to 10 and 20 milliseconds for the in-phase and quadrature-phase channels, respectively, to provide improved cross-correlation performance.

Referred to as the pilot channel, the quadrature-phase com-ponent contains no data modulation and, hence, enables long coherent integration by the user equipment integration that is primarily limited by dynamics estimation errors (i.e., inertial drift) and the stability of the receivers (and to a lesser extent the satellites) reference oscillator. The L5 pilot channel thus allows the implementation of robust and certifiable high-sensitivity and anti-jam processing.

According to IS-GPS-705, the guaranteed minimum received signal power PL5, min, to a user at the surface of the earth measured at the output of a 3 dBi linearly polarized antenna from an SV at elevation above five degrees is stated as -157.9 dBW: 0.6 dB stronger than the -158.5 dBW minimum specified for the legacy L1 C/A code signal.

Assuming the said antenna is at the standard ambient tem-perature of 295K, the guaranteed received minimum carrier-to-noise ratio (C/N0) measured at the output will be 46.9 dB-Hz.

However, C/N0 values measured by a GPS receiver connected to a typical circularly polarized antenna would be three to six decibels lower due to the antenna gain pattern and receiver implementation losses.

The L5 Test TransmissionIt is important to realize that the GPS L5 transmission from SVN49 has significant deviations from the specification of IS-GPS-705. As the article by T. Powell et alia discusses (Addi-tional Resources), the primary goal of integrating an L5 dem-onstration payload into a Block IIR-M satellite was to satisfy the bring into use deadline of August 26, 2009, a date which was set when the United States filed with the ITU Radiocom-munication Sector (ITU-R) to transmit on the L5 frequency.

The following is a listing of these deviations: SVN49 transmits only the dataless quadrature component

of the L5 signal specification. (i.e., in the L5 signal equation presented in the previous section, GI,i(t) = 0).

It is hardwired to generate only one PRN code on L5: L5-Q PRN63. This applies no matter what PRN is assigned to SVN49s primary GPS mission. (Currently, the primary mission assignment is PRN1.)

Most likely, the transmitted power on L5 is lower than required to meet the guaranteed minimum received sig-nal strength specified in IS-GPS-705. This is because the demonstration payload is occupying the auxiliary payload capability of the Block IIR-M spacecraft, which is probably not designed to handle the power requirements of an addi-tional transmission at full-spec.

Most likely, the antenna gain pattern of SVN49s L5 trans-mission may not be optimal for full earth surface coverage because the L5 demo mission is secondary to the IIR-M vehicles primary mission of sustaining the GPS constella-tion.The first two factual constraints mean that the L5 test trans-

mission is non-operational, but allows acquisition and tracking as an experimental signal good enough to meet the condi-tions of the ITU filing and to enable basic research on triple-fre-quency GPS. The latter two deviations in the preceding list are educated guesses based on the authors initial observations of the L5 signal.

L1/L5 RF Front-End and Data Collection SystemIn anticipation of the first L5 signals in space, the Ohio team upgraded its Transform-Domain Instrumentation GPS Receiv-er (TRIGR), described in the article by S. Gunawardena et alia (2007), to include L5. TRIGR represents a breakthrough set of GPS receiver technologies that had been developed at the Ohio University Avionics Engineering Center during the last six years.

The technology encompasses instrumentation-quality RF front-ends, high-fidelity wideband multi-bitsampled interme-diate frequency (IF) data collection and post-processing, and high-performance realtime transform-domain GPS baseband

The GPS L5 transmission from SVN49 has signicant deviations from the specication of IS-GPS-705 [because] the primary goal of integrating an L5demonstration payload into a Block IIR-M satellite was to satisfy the ITUs bring into use deadline.



IGNSS



IGNSS


processing engines implemented on field programmable gate array (FPGA) pro-cessors.

The current version under develop-ment represents the third generation of TRIGR technology and is being tar-geted for a variety of next-generation multi-channel and multi-frequency GNSS applications. Figure 1 shows the block diagram and frequency plan of the L1/L5 RF front-end used for L5 signal analysis.

Figure 2 shows a close-up view of the L5 section of the front-end. The perfor-mance of the L5 section of the front-end was verified previously using L5 signals-in-space from WAAS PRNs 135 and 138. Results of this work and additional details of the front-end are covered in the article by S. Gunawardena et alia (2008).

As shown in Figure 1, the L1/L5 sig-nals downconverted to the 70 MHz IF are bandpass sampled at 56.32 MSPS to yield a digital baseband signal with its IF centered at 13.68 MHz. To maxi-mize options for high-dynamic range GNSS signal processing (such as in the midst of interference), the analog IF is sampled at 14-bit resolution without the use of automatic gain control circuitry. For data collection purposes, these 14-bit samples are then reduced to 8, 4, 2 or 1-bit per sample inside the FPGA and subsequently streamed to a RAID stor-age array.

The ability to stream 8-bit L1/L5 samples continuously (~113 MBytes/sec sustained transfer rate) and to do so for up to five hours continuously (storage limit of the array) enabled us to capture pristine sets of L1/L5 data for the morn-ing and afternoon visibility periods of SVN49. We postprocessed the data to produce the results presented in this article.

Data Collection SetupFigure 3 shows the test setup that was used to collect live L1/L5 data for the results presented in this article.

We used a commercial antenna that covers the L1, L2 and L5 bands. Because this model is a passive antenna, we incorporated a high-quality low noise

amplifier (LNA) with a noise figure below 1 decibel and approximately 40-decibel gain. The LNA is placed as close to the antenna port as possible to obtain

the lowest possible system noise figure. The RF signal is fed to each front-end channel via an active splitter built into the four-channel RF front-end.

GPS L5

GPS L1

Reference Clockfref : 10.0 MHz

IF DatafIF :

13.68 MHz

fs : 56.32 MHz

fLO1 : 1106.45 MHz

fs : 56.32 MSPS

RF/IF Digital Gain Control

Parallelinterface

to Virtex-4FPGA

CH2: GPS L5 RF/IF Section

BPF1: 1176.45 MHz, 2-pole, Bw: 20 MHzBPF2: 1176.45 MHz, 3-pole, Bw: 20 MHzBPF3: 70 MHz, SAW, Bw: 20 MHz

FIGURE 1 L1/L5 RF front-end and frequency plan

FIGURE 2 L5 channel of TRIGR RF front-end

FIGURE 3 Test setup for live GPS L1/L5 RF data collection and spectrum analysis

RF-In Bias-T LNA BPF1 BPF2

IF-In IF DVGA IF SAW Filter IF-Out Manual-Mode DVGA Control

RF DVGA Mixer LO-In Mixer IF-Out

NovAtelGPS-704X

(L1, L2, L5)

Reference Clock

LNAGain: 40 dB

NF:


Acquiring the GPS L5 SignalAs described previously, the L5 test trans-mission may not necessarily adhere to the guaranteed minimum received signal level of -157.9 dBW (~47 dB-Hz). Even though we suspected it, the necessity for more care when acquiring the L5 signal (relative to the easy acquisition of L1 C/A) was not immediately apparent.

Our initial analysis of the dataset containing the L5 turn-on event yield-ed no discernible correlation peak. This had us puzzled: was the L5 turn-on event postponed? Did we have a bug in our L5 code generator? Our initial method used one millisecond of coherent integration (to sidestep having to perform NHQ code wipeoff) and a handful of non-coherent integrations (to increase signal-to-noise ratio at the expense of squaring loss).

Soon we realized this would not work if the signal were significantly weaker than expected. Further, what if the NHQ code transition happened close to the middle of our one-millisecond data blocks? This would kill most of the cor-relation energy and yield no peak. A bit of refinement was in order!

Because we had recorded both the L1 and L5 coherently sampled data streams in the same file, we could initialize the L5 processing by first acquiring SVN49s L1 C/A-code signal (PRN01) and use it to determine both the Doppler frequen-cy offset at L5 (fD,L5 = (fL5 /fL1)fD,L1) and, more importantly, the 20-millisecond L1 C/A navigation data bit epochs. The latter, by definition, are the same as the

20-bit NHQ code epochs (ignoring the few samples offset due to hardware and atmospheric relative delays).

While we were at it, we also aligned our one-millisecond data block to coin-cide with the start of the C/A-code (i.e., code phase zero alignment). With the code-phase and Doppler frequency off-sets nailed, we tried 100 milliseconds of coherent integration with NHQ code wipeoff. (The carrier frequency needs to be within 10 Hz for 100-millisecond coherent integration).

The resulting L5 correlation peak was a sight to behold! Figure 4 shows the acquisition results. Animations of the L5 correlation space can be viewed at .

Observed Strength of L5 Test SignalAs of the time of this writing, at least two groups from Europe have reported GPS-L5 received signal strength being stronger than expected. (See Septentrio and Javad GNSS news releases cited in Additional Resources.) This may be due to the fact that SVN49 reaches higher elevations in Europe.

In the U.S. Midwestern region SVN49 peaks at elevations of approxi-mately 37 and 34 degrees during its two daily passes. At the peaks of these passes, the observed signal strengths are approximately 35 dB-Hz; relatively weak compared to the guaranteed minimums of IS-GPS-705.

As alluded to earlier, the test pay-loads L5 antenna beam pattern appears to have a narrow main beam. Observa-tions at multiple locations would aid in inferring the antenna pattern, which would be useful for numerous research applications.

We are presenting the carrier-to-noise (C/N0) versus time and elevation profiles from our initial observations of the L5 test signal from Athens, Ohio, (N 39 12 33.14520, W 82 13 25.93487) as a first step towards inferring the actual gain pattern of the L5 test signal. We trust this information will be espe-cially helpful for those researchers who plan to use regular L1/L5 GPS antennas (i.e., standard patch-types as opposed to high-gain dish antennas) to get a sense of the signal strengths to expect from a location similar to ours.

For the results shown in this article, the L1 and L5 front-end channels were sampled at an effective resolution of four bits per sample and continuously streamed to the TRIGR RAID array for about four hours, generating files of approximately 850 GB each for the two passes.

Because its prohibitively slow to postprocess the entire files with the soft-ware we were using (requiring about 30 minutes to process a second of data on a fast desktop computer), observations were made every five minutes for a dura-tion of one second per observation.

Figure 5 shows measured C/N0 from SVN49s L1 C/A and L5 Q transmissions

FIRST LOOK AT L5

FIGURE 4 Left: SVN49 L1 C/A code correlation space computed using 100milliseconds of coherent integration (includes 5-bits data wipeoff). Right: Cor-relation space of L5 pilot channel computed using 100-millisecond coherent integration.

SVN49 L1 C/A Code Correlation Space (PRN1)

Carrier Doppler Offset (Hz) Code Offset

(Samples)

Corr

elat

ion

Mag

nitu

de

0

1 23 4

5-20 -40 -60-80

4

3

2

1

x 106

x 104

SVN49 L5-Q Channel Correlation Space (PRN63)

DopplerOffset (Hz) Code Offset

(Samples)

Corr

elat

ion

Mag

nitu

de

0

1 23 4

5-20 -40 -60-80

1.5

2.5

1.5

0.5

x 105

x 104



IGNSS



IGNSS

_________


during the morning visibility period from Athens, Ohio, on April 23. Figure 6 shows the same for the afternoon vis-ibility period.

Figures 7 and 8 show the measured C/N0 as a function of azimuth and elevation angles (i.e., skyplot) from our location for the morning and afternoon visibility periods, respectively. Figure 9shows C/N0 as a function of elevation angle for both periods of visibility.

From Figures 5 and 6, it can be seen

that, on average, the power differential between SVN49s L1 and L5 transmis-sions is approximately 15 decibels. This confirms our initial assessment that the L5 test transmission has significantly lower signal strength than the 46.9 dB-Hz guaranteed minimum specified in IS-GPS-705. The observed minimum C/N0 from our location is closer to 27 dB-Hz.

Assuming that the L1 and L5 gain patterns of the reception antenna are

consistent, as was verified from the antenna manufacturers datasheet, the L1 C/N0 contours from Figures 5 and 6 could be used as a baseline to deduce the L5 antenna gain pattern variation. Using this observation, the figures show two faintly distinguishable minima that may correspond to nulls in the antenna pattern. The time spacing between these nulls is approximately two hours for the morning visibility period and three hours for the afternoon.

030

60

90

120

150180

210

240

270

300

33060

50

40

30

20

10

L5 C/N0 (dB-Hz) versus azimuth & elevation (degrees)

SVN49 Visibility from Athens, Ohio on Apr 23 2009 Morning

030

60

90

120

150180

210

240

270

300

33060

50

40

30

20

10

L5 C/N0 (dB-Hz) versus azimuth & elevation (degrees)

SVN49 Visibility from Athens, Ohio on Apr 23 2009 Afternoon

FIGURE 7 Received C/N0 from L5 test transmission as a function of azi-muth and elevation at Athens, Ohio: morning visibility period

FIGURE 8 Received C/N0 from L5 test transmission as a function of azi-muth and elevation at Athens, Ohio: afternoon visibility period

SVN 49 Signal Strength, Apr-23-2009

CNR,

dB-

Hz

6 6.5 7 7.5 8 8.5 9 9.5

55

50

45

40

35

30

25

20

15

Time of Day, hr

L1L1-polynomial fitL5L5-polynomial fit

FIGURE 5 L1 and L5 received C/N0 as a function of local time (EDT) at Athens, Ohio: morning visibility period

FIGURE 6 L1 and L5 received C/N0 as a function of local time (EDT) at Athens, Ohio: afternoon visibility period

SVN 49 Signal Strength, Apr-23-2009

CN0,

dB-H

z

14.5 15 15.5 16 16.5 17 17.5 18

55

50

45

40

35

30

25

20

15

Time of Day, hr

L1L1-polynomial fitL5L5-polynomial fit



IGNSS



IGNSS


In terms of elevation angle, these minima correspond to an elevation of between 22 to 24 degrees, as distinguishable from Figure 9. This reinforces the notion that SVN49s antenna pattern has a narrower-than-specified main lobe. Besides this initial assessment, we plan to do more detailed analysis of the L5 test transmission during the coming months.

Data Available For DownloadThe Ohio University research team is making available raw L1/L5 sampled IF datasets containing the L5 transmission from SVN49 (naturally includes the WAAS GEO L5 transmissions as well). We hope this data will help other researchers to develop and test their own GPS L5 signal processing techniques.

The data can be downloaded at . The site contains data files sampled at 1, 2, 4, or 8 bits. In addition, file format information and a MATLAB data visual-ization script are also provided. Coherently sampled L1/L2/L5 datasets from our next-generation TRIGR instrument currently under development will also be made available shortly.

Summary and ConclusionsAt long last, after numerous launch delays and just 20 weeks before the ITU filing deadline, an L5 signal is transmitting from a GPS satellite. Even though the L5 signal from SVN49 is non-operational, it is nevertheless useful for many dual and triple-frequency research applications, receiver development, and testing.

The Ohio University research team collected coherently sampled L1/L5 software radio data from its TRIGR instrument of the L5 turn-on event as well as complete data sets for both visibility periods of SVN49 for several days thereafter. The first-look results presented here were obtained by post-processing this data.

We presented data about the visibility of the signal in terms of C/N0 as a function of elevation and azimuth angles,

as observed using a typical multi-frequency patch-type antenna from our location in Athens, Ohio, USA. We observed that the signal is on the average approximately 15 decibels weaker than specified in IS-GPS-705. Moreover, we showed that the signal strength varies significantly with elevation angle from our loca-tion where SVN49 peaks around 35 degrees in elevation.

Our observations contrast stronger-than-expected-signal reports from Europe where SVN49 rises near zenith. Obser-vation of L5 C/N0 minima with respect to time and elevation angle seems to indicate an antenna pattern with a narrower-than-expected main lobe.

As reported by Inside GNSS (see Additional Resources) on May 4, 2009, in addressing the European Navigation Conference, Lt. Col. David Goldstein of the GPS Wing indicated the presence of an anomaly on SVN49 that is responsible for elevation-angle-dependent range biases on the L1 and L2 transmissions.

The availability of long L1/L5 sampled data records from our TRIGR instrument, even days before the L5 turn-on event, enables us to perform a detailed independent study of this anom-aly and perhaps will shed some light as to what to expect for those still intending to do research using what is for now the only triple-frequency, albeit unhealthy, GPS satellite. The results of such a study will be presented in the months ahead.

AcknowledgementsThe authors thank Frank Lorge of the Federal Aviation Admin-istrations William J. Hughes Technical Center for discussions related to the L5 test transmission. Curtis Cohenour with the Ohio University Avionics Engineering Center is acknowledged for providing SVN49 orbit data used in the skyplots. Previous generations of TRIGR technology was developed under fund-ing from the FAA LAAS and WAAS programs.

ManufacturersThe L1/L5 antenna was a GPS-704X from NovAtel, Inc., Cal-gary, Alberta, Canada. We used a 3 GHz spectrum analyzer from Agilent Technologies, Santa Clara, California, USA, for initial spectrum observations. TRIGR uses FPGAs from Xilinx Inc., San Jose, California, USA., and runs on the Windows XP platform from Microsoft Inc., Redmond, Washington, USA. The data was postprocessed using MATLAB from the Math-works, Inc., Natick, Massachusetts, USA.

Additional Resources[1] ARINC Engineering Services LLC, Navstar GPS Space Segment/User Segment L5 Interfaces: IS-GPS-705, September 22, 2005, available at , accessed April 30, 2009[2] Gunawardena, S. (2007), and A. Soloviev and F. van Graas, Wideband Transform-Domain GPS Instrumentation Receiver for Signal Quality and Anomalous Event Monitoring, NAVIGATION: Journal of the Institute of Navi-gation, Vol. 54 No. 4, November 2007[3] Gunawardena, S. (2008), and Z. Zhen and F. van Graas, Triple Frequency RF Front End for GNSS Instrumentation Receiver Applications, Proceedings of the 21st International Technical Meeting of the Satellite Division of the Institute of Navigation, September 1619, 2008, Savannah, Georgia, USA

SVN49 L5 C/N0 versus Elevation Angle

AMC/

N 0, d

B-Hz

5 15 25 35 25 15 5

403530252015

Elevation, deg

L5 C/N0Polynomial fit

PMC/

N 0, d

B-Hz

10 20 30 40 30 20 10

403530252015

L5 C/N0Polynomial fit

FIGURE 9 LReceived C/N0 from L5 test transmission as a function of eleva-tion angle for both visibility periods at Athens, Ohio

FIRST LOOK AT L5



IGNSS



IGNSS

___

__________________________________


[4] Inside GNSS (on website), GPS SVN49 and L5Signal: A Success with Problems, , May 5, 2009[5] Javad GNSS Inc., Our Customers Now Track GPSL5 Signal, News Release, April 24, 2009; availableat , accessed April 30, 2009[6] NovAtel Inc. GPS-704X: GNSS WidebandAntenna, available at , accessed April30, 2009.[7] Powell, T., and C. Edgar, H. Ozisik, M. McFad-den, D. Reigh, R. Spieth, and J. Irvine, The L5Demo Payload on GPS Mission IIR-20, Proceed-ings of the 21st International Technical Meeting of the Satellite Division of the Institute of Naviga-tion, September 1619 2008, Savannah, Georgia,USA[8] Septentrio Satellite Navigation NV, Dual-Constellation Live L5 Tracking with SeptentrioPolaRx3G, news release, April 16, 2009, availableat , accessed April 30, 2009

AuthorsSanjeev Gunawardena is asenior research engineerand co-principal investi-gator with the Ohio Uni-versity Avionics Engi-n e e r i n g C e n t e r .Gunawardena earned

Ph.D. and M.S. degrees in electrical engineeringfrom Ohio University, and was the 2007 recipientof the RTCA William E. Jackson Award. Gunawar-denas research interests include RF systemsdesign, digital systems design, reconfigurablecomputing, and all aspects of GPS receivers andsignal processing.

Zhen Zhu is a seniorresearch engineer withthe Ohio University Avi-onics Engineering Cen-ter, and an adjunctassistant professor withthe School of Electrical

Engineering and Computer Science, Athens,Ohio, USA. He received a Ph.D. in electrical engi-neering from Ohio University. His research inter-ests include GPS and augmentation systems,software radio technology, GPS interference andmultipath, computer vision, and laser-basednavigation, and automatic navigation and guid-ance. He has also been actively involved inresearch of articial intelligence, neural net-works, and machine learning.

Michael Braasch, Ph.D., is Thomas Professor ofElectrical Engineering and director of the Avionics

EngineeringCenter(AEC)at Ohio University. Hewasoneofthepioneeringr e s e a r c h e r s w h oi n v e s t i g a t e d t h eapplication of softwareradio and transform-

domain signal processing techniques for GPSreceivers. His work also includes research in high-precision GPS positioning through differential

carrier-phase processing. Braasch isinternationally recognized for his work oncharacterizing the effects of GPS multipath onboth pseudorange and carrier-phase basebandsignal processing and is one of the originators ofthe integrated multipath-limiting antenna forGPS. In addition to GPS-related research, he hasalso worked on other navigation systems includingINS, ILS, MLS, VOR, Loran-C, and DME.



IGNSS



IGNSS

__________

___________

_______________

A t 5 a.m. PDT (Pacific Daylight Time) on April 10, 2009, GNSS reached a new milestone as the first GPS L5 signal was turned on and transmitted from a GPS satel-lite. Eagerly anticipated, this is the first broadcast of a GPS signal in a frequency band dedicated solely for civilian use.

The latest signal represents the first step into a new era for GNSS users, particularly for safety-of-life applica-tions. The L5 signal gives these users two civil signals (L1 being the other) in a protected aeronautical radionaviga-tion services (ARNS) band. This allows for ionospheric corrections using only ARNS bands.

After a lengthy delay in satellite launches caused by a variety of technical problems, the new civil GPS signal L5 began transmitting a demonstration signal on April 10. Using a modied navigation payload on a GPS Block IIR-M satellite placed in orbit only two weeks earlier, the U.S. Air Force switched on the eagerly awaited, modernized signal. The article reports the initial observations of L5 broadcasts by researchers in California and Colorado, including the appearance of an anomaly in SVN49s L1 signal.

GRACE XINGXIN GAO, LIANG HENG, DAVID DE LORENZO, SHERMAN LO STANFORD UNIVERSITYDENNIS AKOS UNIVERSITY OF COLORADOALAN CHEN, TODD WALTER, PER ENGE, BRADFORD PARKINSON STANFORD UNIVERSITY

Above, left to right: The 1.8-meter dish antenna of Stanford GNSS Monitoring Station (SGMS), Stanford, California; ITS 18-meter parabolic antenna, Table Mountain, Colorado; SRI Internationals 45.7-meter radio telescope antenna in the Stanford hills, Stanford, California.




IGNSS



IGNSS


Additionally, the higher chipping rate on L5 enables more precise code phase measurements than available on other civilian GPS frequencies.

This article will trace the history of L5s origin within the GPS program modernization initiative, the signals characteristics as designed, and the results of observations of the initial transmission of L5 and other GPS sig-nals from three research facilities.

A Short History of L5Development of the L5 signal has its roots in the increasing use of GPS by civil users, particularly in aviation. As the system approached full operational capability, GPS system managers and user communities recognized that a sec-ond civil frequency would provide great benefits. Eventually, a second civilian signal became part of GPS moderniza-tion planning in the late 1990s.

In fact, a select panel of GPS experts recommended that two additional civil frequencies be broadcast, L2C and L5. An anecdote from that period charac-terizes the recommendation as arising from a debate over strategy: the civil L2 signal was thought to pose a lower risk approach to getting Department of Defense (DoD) buy-in, as there was already a GPS P(Y) broadcast on L2 (at 1227.6 MHz). However, a civil signal on L5 (centered at 1176.45 MHz), located in an ARNS band, would be more useful to safety-of-life users.

Rather than deciding to ask for one or the other, the panel recommended both new signals. After discussion and analysis, the D

gnss_20090501_may_2009

Documents

cover search issue

hereprevious page contents

entire issue

keyword search

new l5 signal

gnss world

sharethis digital edition

cband work