IMPROVING RESILIENCE TO EMERGENCIES THROUGH … · GSA. Galileo increases the accuracy of location...

IMPROVINGRESILIENCETOEMERGENCIESTHROUGH

ADVANCEDCYBERTECHNOLOGIES

ReportonEGNSSIntegration

Deliverable ID D3.4

Work Package Reference WP3

Issue 1.0

Due Date of Deliverable 30/11/2017

Submission Date 15/11/2017

Dissemination Level1 PU

Lead Partner ISMB

Contributors -

Grant Agreement No 700256

Call ID H2020-DRS-1-2015

Funding Scheme Collaborative

I-REACT is co-fundedby theHorizon2020FrameworkProgrammeof theEuropeanCommissionundergrantagreementn.700256

1PU=Public,PP=Restrictedtootherprogrammeparticipants(includingtheCommissionServices),RE=Restrictedtoagroupspecifiedbytheconsortium(includingtheCommissionServices),CO=Confidential,onlyformembersoftheconsortium(includingtheCommissionServices)

Improving Resilience to Emergencies through Advanced Cyber Technologies

Project: I-REACT Report on EGNSS Integration Deliverable ID: D3.4 Grant Agreement: 700256 Call ID: H2020-DRS-1-2015 Page: 2 of 34

Preparedby Reviewedby Approvedby

G.Marucco W.Stemberger C.Rossi

Issue Date Description Author(s)

0.01 13/10/2016 Provisionoftemplate W.Stemberger

0.1 15/11/2016 TableofContent G.Marucco

0.2 14/03/2017 FirstVersion G.Marucco

0.3 08/11/2017 DraftFinalVersion G.Marucco

0.4 10/11/2017 Internalreview M.Pini

0.5 13/11/2017 FirstDeliveryversion G.Marucco

1.0 14/11/2017 FinalVersion G.Marucco



TABLEOFCONTENTS1 INTRODUCTION...........................................................................................................................5

1.1 PurposeoftheDocument....................................................................................................5

1.2 StructureoftheDocument..................................................................................................5

1.3 Acronymslist.......................................................................................................................6

1.4 Referenceandapplicabledocuments.................................................................................7

2 EGNSS..........................................................................................................................................9

2.1 EGNOS..................................................................................................................................9

2.2 Integrity.............................................................................................................................11

2.3 EDAS..................................................................................................................................14

2.4 Galileo................................................................................................................................15

3 AUGMENTATIONMODULE.......................................................................................................18

3.1 Overview............................................................................................................................18

3.2 EDASDecoder....................................................................................................................19

3.3 AugmentedPVTandIntegrityComputation.....................................................................22

4 GNSSCHIPSETSELECTION.........................................................................................................24

4.1 GNSSData..........................................................................................................................24

4.2 Integration.........................................................................................................................25

4.3 MultiConstellationGNSSReceivers..................................................................................25

5 LOCALINTEGRITY......................................................................................................................27

5.1 Stateoftheartandlatestdevelopments..........................................................................27

5.2 proposedalgorithm...........................................................................................................27

5.3 Validation...........................................................................................................................30

6 CONCLUSIONS...........................................................................................................................33



LISTOFFIGURES

Figure2-1:EGNOSsystemarchitecture...........................................................................................10

Figure2-2:ProtectionLevels............................................................................................................13

Figure2-2:EDASdataprovision.......................................................................................................14

Figure2-3:EDASarchitectureandservices......................................................................................15

Figure3-1:MainsoftwarecomponentsandflowsoftheAugmentationmodule...........................19

Figure3-1:EDASstreamdecodingandstorage...............................................................................21

Figure3-2:Augmentationalgorithmflowchart...............................................................................23

Figure5-1:Resultsofdatacollections for thealgorithmtuning. (a)and (b)showresidualssinglevalues,theiraverageandaquadraticfittotheaveragevaluesinthecaseofavegetatedroadonthehillsrespectivelytowardssatelliteelevationandC/N0.Augmentationalgorithmflowchart..........30

Figure5-2:A25kmlongtrackzoomedinthemostcriticalpoint(i.e.worstoff-track).Greenarrowpointstothecomputedpositionwhiletheredarrowpointstotherealposition.Realpositionfallscorrectlyinsidetheellipse...............................................................................................................31

Figure5-3:A10kmlongurbantrackzoomedinthemostcriticalpoint.Greenarrowpointstothecomputedpositionwhile the redarrowpoints to the realposition.Realposition falls inside theellipse. Inthiscasethedatawerecollectedusingabikepassingmorethanoncealongthesamestreets:thisexplainsthehighellipsesdensity.................................................................................31

LISTOFTABLE

Table2-1:Message4085subtypes..................................................................................................20



1 INTRODUCTION

1.1 PURPOSEOFTHEDOCUMENT

Thegoalofthisdocumentistodescribethedesign,implementationandtheintendedexploitationofGlobalNavigationSatelliteSystem(GNSS)withinI-REACT,withparticularreferencetotheaspectsconcerningtheEuropeansystems(i.e.EGNOSandGalileo).

Fromthealgorithmicpointofview,mostof theeffortwason thestudyofanovelapproachtoestimate the position confidence, which is not only based on the information provided byaugmentationsystems,butalsoleveragesonthecharacterizationoftheenvironmentsurroundingtheantennathroughtheprocessingofthereceivedincomingGNSSsignals.ThisconceptisemerginginLocationBasedService(LBS)andisreferredas“localintegrity”,becauseittakesintoaccountsthelocalimpairments(i.e.:multipath,interferingsignals,attenuations,etc.)thatactuallyhavethemostsignificanteffectsonthepositionaccuracyand,inturn,onthereliability.

1.2 STRUCTUREOFTHEDOCUMENT

Thedocumentisorganizedasinthefollowing:

• Chapter1isthisintroductionanddescribesthedocumentcontent;• Chapter2isadescriptionofEGNSSandrelatedservices.Itincludesadescriptionofthe

integrityconcept;• Chapter3describestheaugmentationmodule,asdesignedandimplementedwithinI-

REACT;• Chapter4focusesontheselectionofproperGNSSchipsetstobeintegratedintheI-REACT

wearabledevices;• Chapter5describesthedevelopmentcarriedoutwithinI-REACTforthelocalintegrity

computationandassessment;• Chapter6containstheconclusions.



1.3 ACRONYMSLIST

AL Alarm Limit

AM Augmentation Module

APC Antenna Phase Centre

ARP Antenna Reference Point

ASQF Application Specific Qualification Facility

ATC Air Traffic Control

ATPL Along Track Protection Level

C/N0 Carrier to Noise ratio COSPAS-SARSAT

COsmicheskaya Sistyema Poiska Avariynikh Sudov - Search And Rescue Satellite-Aided Tracking

CRC Cyclic Redundancy Check

CS Commercial Service

CTPL Cross Track Protection Level

EDAS EGNOS Data Access Service

EGNOS European Geostationary Navigation Overlay Service

EMS Emergency Management Service

EO Earth Observation

ESA European Space Agency

EU European Union

EWAN EGNOS Wide Area Network

EWS Early Warning System

FOC Full Operational Capability

FTP File Transfer Protocol

GCC Galileo Control Centres

GEO Geostationary satellites

GIVE Grid Ionospheric Vertical Error

GLONASS Global'naja Navigacionnaja Sputnikovaja Sistema (Russian GNSS)

GNSS Global Navigation Satellite System

GPS Global Positioning System

GSA European GNSS Agency

GSS Galileo Sensor Stations

HPL Horizontal Protection Level

ICAO International Civil Aviation Organization

IOC Initial Operational Capability

JSON JavaScript Object Notation

LBS Location Based Services

MCC Mission Control Centres

MOPS Minimum Operational Performance Standard

NLES Navigation Land Earth Stations

Ntrip Networked Transport of RTCM via Internet Protocol

OS Open Service

PACF Performance Assessment and Checkout Facility

PL Protection Level

PRN Pseudo Random Number (code)

PRS Public Regulated Service

PVT Position Velocity and Time



RAIM Receiver Autonomous Integrity Monitoring

REST REpresentational State Transfer

RIMS Ranging Integrity Monitoring Stations

RTCA Radio Technical Commission for Aeronautics

RTCM Radio Technical Commission for Maritime

RTK Real Time Kinematic

SAR Search And Rescue service

SARPS Standards and Recommended Practices

SBAS Satellite Based Augmentation System

SL (EDAS) Service Layer

SIS Signal in Space

SISNet Signal in Space over Network

SL Service Level

SoL Safety of Life

TOW Time Of Week

TTA Time To Alarm

UDRE User Differential Range Error

UTC Universal Time Coordinated

VANET Vehicular Ad hoc NETwork

VPL Vertical Protection Level

WAAS Wide Area Augmentation System

1.4 REFERENCEANDAPPLICABLEDOCUMENTS

ID Title Revision Date

[RD01] European GNSS Service Centre https://www.gsc-europa.eu/

Online: accessed Nov.2017

2017

[RD02] EGNOS Portal https://www.egnos-portal.eu/


2017

[RD03] European Space Agency: Galileo http://www.esa.int/Our_Activities/Navigation/Galileo/What_is_Galileo


2017

[RD04] International Standards and Recommended Practices – Annex 10 to the Convention on International Civil Aviation – Aeronautical Telecommunications – Volume 1 Radio Navigation Aids

6th ed. July 2006

[RD05] DO-229D Minimum Operational Performance Standards for Global Positioning System/Wide Area Augmentation System Airborne Equipment

D 13/12/2006

[RD06]

Don Jewell, Protect, Toughen, Augment: Words to the Wise from GPS Founder http://gpsworld.com/protect-toughen-augment-words-to-the-wise-from-gps-founder/


15/04/2014

[RD07]

CEN/CENELEC, Space - Use of GNSS-based positioning for road Intelligent Transport Systems (ITS) - Part 1: Definitions and system engineering procedures for the establishment and assessment of performances

DRAFT EN 16803-1

October 2014

[RD08] E.D. Kaplan and C.J. Hegarty. Understanding GPS: Principles and Applications. Artech House. 2nd ed. 2006



[RD09] EGNOS Data Access Service https://www.gsa.europa.eu/egnos/edas


2017

[RD10] https://developer.android.com/guide/topics/sensors/gnss.html accessed Nov.2017 2017

[RD11] GSA. Galileo goes live. https://www.gsa.europa.eu/newsroom/news/galileo-goes-live

Online: accessed Sep.2017

2016

[RD12] GSA. Galileo increases the accuracy of location based services https://www.gsa.europa.eu/news/results-are-galileo-increases-accuracy-location-based-services

Online: accessed Sep.2017

2017

[RD13]

The Local Integrity Approach for Urban Contexts: Definition and Vehicular Experimental Assessment. Margaria D., Falletti E. Sensors (Basel); 16(2):154. doi: 10.3390/s16020154.

- 26/01/2016

[RD14]

Cosmen-Schortmann, J.; Azaola-Saenz, M.; Martinez-Olague, M.A.; Toledo-Lopez, M. Integrity in Urban and Road Environments and its Use in Liability Critical Applications. In Proceedings of the IEEE/ION Position, Location and Navigation Symposium, Monterey, CA, USA,; pp. 972–983.

8 May 2008

[RD15]

Pullen, S.;Walter, T.; Enge, P. SBAS and GBAS Integrity for Non-Aviation Users: Moving Away from “Specific Risk”. In Proceedings of the 2011 International Technical Meeting of The Institute of Navigation, San Diego, CA, USA; pp. 533–545.

26 January 2011

[RD16] K. Borre. Gps easy suite ii - RAIM. Inside GNSS, 4(4):48–51 2009

[RD17] D. Margaria and E. Falletti. A novel local integrity concept for GNSS receivers in urban vehicular contexts. In Position, Location and Navigation Symposium, Monterey, pages 413–425. IEEE/ION

2014

[RD18] https://www.novatel.com/products/span-gnss-inertial-systems/ Online: accessed Nov.2017

2017



2 EGNSS

This Chapter recalls the fundamentals of the European Global Navigation Satellite Systems(EGNSS)[RD01] and their related services. It also includes a terse description of the standardintegrityconceptusedintheaviationdomain.Suchaconceptisthestartingpointforallthenewalgorithms (developed by researchers for LBSs working in other domains) that propose thecomputationofaconfidenceintervalontopoftheestimatedGNSSpositions.

TheacronymEGNSSisusedtoaddresstwodifferentEuropeansystems:EGNOS[RD02]andGalileo[RD03].Whilethefirstisaregionalaugmentationsystemconceivedtoimprovetheexploitationofthe American Navstar Global Positioning System (simply known as GPS), the second is anindependentcivilGNSSsetupbytheEuropeanUnion(EU)andtheEuropeanSpaceAgency(ESA).

2.1 EGNOS

TheEuropeanGeostationaryNavigationOverlayServicewasbornastheEuropeanversionoftheAmericanWideAreaAugmentationSystem(WAAS).Boththesesystemsareconceivedspecificallytosupporttheautomaticnavigationofaircraftsprovidingsupplementalinformationviasatellites:forthisreasontheyarecalledSatelliteBasedAugmentationSystems.WAASprovidessupportforNorthAmericawhileEGNOSforEurope;bothofthemarecompliantwithtwosetsofInternationalStandards that enable their use by Civil Aviation Authorities: the Standards and RecommendedPractices(SARPS)StandardforSBASsystems[RD04]andtheMinimumOperationalPerformanceStandard (MOPS) DO229 [RD05]. The SARPS has been established and controlled by theInternational Civil Aviation Organization (ICAO). It provides standards regarding the type andcontentofdata,whichmustbegeneratedandtransmittedbyanSBASsystem.Ingeneral,theSBASprovidershallbroadcastaSBASSignalinSpace(SIS)complianttothisstandardintermsofradio-frequency characteristics, and data content and format. The MOPS has been established andcontrolledbytheUSRadioTechnicalCommissionforAeronautics(RTCA)anditprovidesstandardsforSBASreceiverequipment.

EGNOSwasimplementedbyEUROCONTROL,ESAandECtoaugmentthepotentiality(inparticularintermsofsafety)ofGPSovertheEuropeancontinentandthisisachievedwiththeprovisiontotheusersof:

• WideAreaDifferentialcorrections;• Integrityinformation.

EGNOSwasimplementedwithaspecificarchitecture(Figure2-1)dividedinthreesegments:

• Space segment: the geostationary satellites (GEO) broadcasting information vis GPS-likesignals;

• Groundsegment:itistheinformationsource.Itisacomplexsystemcomposedbyseveralelementsthatcollectmeasurements,computereal-timecorrectionsandestimateandsend



themtothegeostationarysatellitesforbroadcasting.Thesystemiswidelydistributed,highlyredundantandperformscontinuousself-checks.

• Usersegment:userswithproperreceiversabletoprocesstheinformationprovided.

Figure2-1:EGNOSsystemarchitecture.

TheEGNOSGroundSegmentcomprises:

• 39RangingIntegrityMonitoringStations(RIMS)• 4MissionControlCentres(MCC)• 6NavigationLandEarthStations(NLES)(2foreachGEO)• TheEGNOSWideAreaNetwork(EWAN)whichprovidesthecommunicationnetworkforall

thecomponentsofthegroundsegment.• 2additionalfacilitiesarealsodeployedaspartofthegroundsegmenttosupportsystem

operationsandserviceprovision:o thePerformanceAssessmentandCheckoutFacility(PACF)o theApplicationSpecificQualificationFacility(ASQF)

EGNOS is a Safety of Life (SoL) system. For such systems, some parameters are used for theperformanceevaluation:

• Availability:abilityof thesystemtoperform its functionat the initiationof the intendedoperation.

• Continuity:abilityofthetotalsystemtoperformitsfunctionwithoutinterruptionsduringtheintendedoperation.



• Accuracy: degree of conformance between the computed user position and the trueposition.

• Integrity:abilityofthesystemtoprovidetimelywarningstouserswhenitmaynotbeusedtonavigate

2.2 INTEGRITY

Integrity is the most stringent requirement for SoL systems, because it relates to the systemreliability.Aninformal,butveryeffective,definitionwasprovidedbyDr.BradParkinson[RD06]:

“IknowI’mgettingthisaccuracy,thesystemisnotlyingtome…”

However,theformaldefinitionofpositionintegrity[RD07]is:

“Positioningintegrity”(orsimply“integrity”)canbedefinedasageneralperformancefeaturereferringtotheleveloftrustausercanhaveinthevalueofagivenpositionorvelocityasprovided

byalocationsystem.“

Althoughintegrityisacomplexframework,itsultimategoalistoassociateaconfidenceintervaltoanypositionestimateproducedbythelocationsystem,providedthatthisconfidenceintervalcanbecomputed,inthehypothesisthattheoperationalconditionsareproperlymonitored,modelledorestimated.Theconfidenceintervalandtheprobability inherentlyassociatedtoitaretypicallymappedtotheconceptsof“protectionlevel”and“integrityrisk”.

TheintegrityofthenavigationsolutionsisthebasisforSoLapplications,startingfromaviation.Infact,algorithmsandsystemshavebeenfirstlydevelopedtoenablesafeaircraftsoperationswiththeuseofGNSS.Atthereceiverside,thefirstmethodstoprovideintegrityinformationbelongtotheReceiverAutonomousIntegrityMonitoring(RAIM)algorithmfamily[RD16].TheyenableproperGNSSreceiverstodetectsignalsproblems,whichcanimpairpositioningperformance.Ontheotherhand,atsystemlevel, thefirstreceiver independentsystemdevelopedwastheaforementionedWAAS,followedbyEGNOS.

EGNOSprovidesintegrityinformationintwodifferentways:

1. TheGEOsatellitebroadcastscoarseintegritydataonly;2. TheGEOsatellitebroadcastscoarse integritydatapluswideareacorrectionsandrelated

integrityinformation.

Inthefirstcase,thecoarseintegritydataincludeuse/don’t-useinformationonallsatellitesinviewoftheapplicableregion.ThereceiverincludesinthePVTcomputationonlythesatellitesthatareflaggedasusable.

Inthesecondcase,foreachsatelliteoftheconstellation,theSBASprovidesinformationtocorrectsatelliteclockparameters,positionofthesatellite,ionosphereeffects,whiletheuseofamodelisforeseen to compensate for the effect of the troposphere. At the same time, supplemental



informationisprovidedtocomputethereliabilityofsuchcorrectionsandmodels.Obviously,SBASdonotprovideneitherinformationnorcorrectionstomitigateerrorsduetothepropagationofthesignalintheenvironmentsurroundingthereceivingantenna.

Suchsupplementalinformationreferstotheestimationoftheresidualerrorsafterapplicationofthecorrections,whichareexpressedinformofvariances:

• σ2UDREisthevarianceofaNormaldistributionassociatedwiththeUserDifferentialRangeError(UDRE)forasatelliteaftertheapplicationoffastcorrectionsandlongtermcorrections,excludingatmosphericeffects.Itisprovidedforallthesatellitesusedforpositioning.

• σ2GIVEisthevarianceofaNormaldistributionassociatedwiththeresidualGridIonosphericVerticalError(GIVE)anditisassociatedwithaspecificpoint(belongingtoadefinedgrid).Itis provided for all the grid points for the specific service area (for EGNOS it roughlycorrespondstoEurope)usedforpositioning(throughaspecificinterpolationmethod).

Forthecomputationoftheconfidenceintervalontopoftheestimatedposition(i.e.:thatintheaviation domain is called Protection Level), four parameters, for each satellite i included in thepositionsolution,areneeded.Theyare:

• σ2i,fltwhichtakesintoaccountσ2UDREanditsdegradationintime.

• σ2i,UIREwhichtakesinaccountσ2GIVEinterpolatedfortheuserposition.

• σ2i,tropo which takes into account the residual error following the application of thetroposphericcorrectionprovidedbyaspecificmodel.

• σ2i,airwhichtakesintoaccounttheoperationalenvironment(airstandsforaviation)andtheGNSSreceivercharacteristics.

ThedataforthecomputationofthefirsttwotermsarebroadcastbyEGNOS,whilethefollowingtwoaredefinedin[RD05].

Asdefinedin[RD05]andexplainedinseveralbooksonGNSS(e.g.:[RD08]),forthesatellitei,σ2i=σ2i,flt+σ2i,UIRE+σ2i,air+σ2i,tropoprovidesthepseudorangemeasurementresidualerrorvarianceaftertheapplicationofEGNOScorrections.

ThecombinationofthesevalueswiththesatellitegeometryandanaviationspecificrequirementcalledintegrityriskleadstothesocalledProtectionLevels(PLs).Thesearedefinedinthehorizontalplane (HPL) andalong the vertical (VPL) andprovide anupperbound to theerror affecting thenavigationon-boardequipment.ThePLscomparedtothemaximumallowederrorsforeachphaseofflighttelltheuserifthenavigationsystemistrustableornot.IfthePLisbiggerthantheallowederror(thesocalledAlarmLimit)analarmshallberaisedwithinthesocalledTimeToAlarm(usually6s).

ThetwoPLsdefineacylindercentredintheuserpositioncomputedbytheGNSSreceiver,whilethetruepositionisassured(withtheprobabilityrequestedbythespecificapplication)tobecontainedinthecylindervolume.



Figure2-2:ProtectionLevels

What reported above stands for aviation applications. Aeronautical GNSS signal propagationmodelsarenotusableastheyareinotherapplicationsbecause:

• theyassumeopen-skysatellitevisibility;• theyassumewingsandtailmultipathorgrounddiffusedmultipath.

Intheotherterrestrialapplications,itisstillpossibletoexploittheinformationmadeavailablebyEGNOS, because it is related to the satellite clocks and orbits and signal disturbances due toionosphericandtroposphericpropagation.Thisisindependentfromthespecificenvironment.

However,thelocalenvironmenthasimportanteffectsonthesignalpropagationand,inturn,onthefinalaccuracyandreliabilityofthecomputedposition.Usuallyoneofthemostdetrimentalsourceof error is theaforementionedmultipath,which refers to the receptionof the samenavigationsignals after one ormore reflections fromnearby objects: in the case of an aircraft the effectsproducedbymultipatharetakenintoaccountbymodelsspecificallydevelopedforaviation.Ifotherapplicationsareconsidered,aviationspecificmodelsareunfitforthelastpartofthesignalpath,dealingwithcomplexenvironmentslikeurbanoneorwherevegetationispresent.Thismeansthatproper strategies able to evaluate the degradations due to the local environment shall beimplemented. This is an open frontier of research, especially for safety-related terrestrialapplicationsthatestimatetheuser’spositionwithGNSS.

ThesameconceptisexploitedinI-REACT.Althoughtheapplicationcannotbeconsideredsafety-critical, the knowledge of a confidence interval on top of the estimated position is consideredvaluableforthewholeservice.ThiswillbematterofChapter5onLocalIntegrity.



2.3 EDAS

EGNOSprovidesthreeservices,twoviatheSpacesegment:

• OpenService:integrityisnotguaranteed;• SafetyofLife:integrityisprovided;

whilethethirdisprovidedviainternet:thisistheEGNOSDataAccessService[RD09].

Figure2-3:EDASdataprovision

The main reasons behind the development of this service were the EGNOS signal receptionproblemsduetothegeostationaryorbitofthetransmittingsatellites.Infact,whileGNSSnavigationsatellitesarealmostevenlydistributedinthesky,sothatasubsetofthemisusuallyvisible,it’snotunusualtohavealltheEGNOSsatellitesobstructedbyabuilding(oranyothersurroundingobject),inparticularathighlatitudes.

Inadditiontoofferingground-basedaccesstoEGNOSdata,EDASisthesinglepointofaccessforthedatacollectedandgeneratedbytheEGNOSinfrastructure.

ThemaintypesofdataprovidedbyEDASare:

• GPS,GLONASSandEGNOSGEOdatacollectedbytheentirestationsnetworkofEGNOS• EGNOSaugmentationmessagesidenticaltothosebroadcastviaGeostationarySatellites• AntennaphasecentrecoordinatesforeachEGNOSreferencestation



Figure2-4:EDASarchitectureandservices

Figure2-4depictstheEDASarchitectureandthedifferentservicesavailable.Theseareclassifiedin:

• MainDataStreamServicesdeliverrawdatavia:

– ServiceLevel0(SL0):itisneededtoeithertransmitdatainrawformat,ortransmittheminaformatthatallowsacompletereconstructionafterdecoding.

– ServiceLevel2(SL2):itisusedtotransmitdatainRTCM3.1standard.

• DataFilteringserviceallowsEDASuserstoaccessasubsetoftheSL0orSL2datatoreducebandwidthconsumption.

• FTPserviceenablesEDASuserstogetEDAS/EGNOShistoricaldataindifferentformatsanddatarates.

• SISNeT serviceprovides access to the EGNOSGEO satellitesmessages over the InternetthroughtheSISNeTprotocol(definedbyESAin2002).

• Ntrip service provides data from the EGNOS network through the Ntrip protocol whichrepresentthestandardfordifferentialcorrectiondistribution.

2.4 GALILEO

Galileo is the EuropeanGlobalNavigation Satellite System. It has been conceived for twomainpurposes:theimprovementoftheperformanceswithrespecttoexistingGNSSandmakingEuropeindependentfromothernon-civiliansystems.



TheprogrammeisdesignedtobecompatiblewithallexistingandplannedGNSSandinteroperablewith GPS and GLONASS. In this sense, Galileo is positioned to enhance the coverage currentlyavailable–providingamoreseamlessandaccurateexperienceformulti-constellationusersaroundtheworld.

Satellitepositioninghasbecomeanessentialservicethatweoftentakeforgranted,butthisisnottrueforotherGNSS,whileGalileoissettoguaranteeavailabilityoftheserviceunderallbutthemost extreme circumstances. This is fundamental in a world where the use of satellite-basednavigation systems continues to expandand consequently the implicationsof a potential signalfailurebecomeevengreater. TheadditionofGalileo to the globalGNSS constellationsnotonlyminimisestheserisks,butalsoensuresbetterperformanceandaccuracyfortheend-user.

TheGalileosystem,oncefullyoperational,willofferfourhigh-performanceservicesworldwide:

• OpenService(OS):Galileoopenandfreeofchargeservicesetupforpositioningandtimingservices.

• Commercial Service (CS): a service complementing the OS by providing an additionalnavigationsignalandadded-valueservicesinadifferentfrequencyband.TheCSsignalcanbeencryptedinordertocontroltheaccesstotheGalileoCSservices.

• Public Regulated Service (PRS): service restricted to government-authorised users, forsensitiveapplicationsthatrequireahighlevelofservicecontinuity.

• SearchandRescueService(SAR):Europe’scontributiontoCOSPAS-SARSAT,aninternationalsatellite-based searchand rescuedistressalertdetection system.Satellitesare thereforeequippedwitha transponder,which isable to transfer thedistresssignals fromtheusertransmitters to regional rescue co-ordination centres,whichwill then initiate the rescueoperation.Atthesametime,thesystemwillsendaresponsesignaltotheuser,informinghimthathissituationhasbeendetectedandthathelpisontheway.Thislatterfeatureisnewandisconsideredamajorupgradecomparedtotheexistingsystem,whichdoesnotprovideuserfeedback.

ThefullydeployedGalileosystemwillconsistof24operationalsatellitesandupto6activespares,positioned in threecircularMediumEarthOrbitplanes.Eachorbithasanominalaverage semi-majoraxisof29600km,andaninclinationof56degreeswithreferencetotheequatorialplane.

AnInitialOperationalCapability(IOC)phaseisforeseentobebasedon18satellites.Atthisstage,theOpen Service, Search and Rescue and Public Regulated Servicewill be availablewith initialperformances.Thenastheconstellation isbuilt-upbeyondthat,newserviceswillbetestedandmadeavailabletoreachFullOperationalCapability(FOC).

Having18satelliteinorbit(11available,4undercommissioning,2undertestingand1notavailable),inDecember2016theavailabilityofInitialServiceshasbeendeclared.ThismeansthattheGalileosatellitesandgroundinfrastructurearenowoperationallyready.

TwoGalileoControlCentres(GCCs)havebeenimplementedonEuropeangroundtoprovideforthecontrolofthesatellitesandtoperformthenavigationmissionmanagement.Thedataprovidedby



aglobalnetworkofGalileoSensorStations(GSSs)aresenttotheGalileoControlCentresthrougharedundantcommunicationsnetwork.TheGCCsusethedatafromtheSensorStationstocomputetheintegrityinformationandtosynchronisethetimesignalofallsatelliteswiththegroundstationclocks. The exchange of the data between the Control Centres and the satellites is performedthroughup-linkstations.

Galileowillprovide severaladvantages inparticular through theexploitationof theCommercialServicethatwillenablePrecisePointPositioning,howeverbeingthedeploymentofsuchserviceforeseenafter2018,theI-REACTprojectwillpointontheexploitationGalileoandGPSaugmentedbyEGNOSandEDAS,plusGLONASS,thatevennotaugmentedcanhelpin improvingpositioningavailability(seeparagraph4.3)



3 THEAUGMENTATIONMODULE

WithAugmentationModule(AM)wedefineanexternalsoftwarecomponentrunningexternallytotheI-REACTORbackend,whichprovidesto itapreciseandreliablepositioningserviceexploitingEDASandimplementingtheaugmentationasforeseenbythestandards[RD04]and[RD05]plusthecomputationofintegrityasdescribedinChapter5.

Through the I-REACTORbackend, theAM receives rawGNSSmeasurements fromtheusersandprovidestheusers’PVT,correctedwithEDASdata.Italsoprovidesaconfidenceinterval,basedonthe“localintegrity”concept.

An AMwith basic functionalities was already developed by ISMB in the frame of the FLOODISproject.TheI-REACTprojectreviewed,completedandupdatedsuchoriginalversion.Inparticularanew algorithm for the computation of the “Local Integrity” has been added to the componentdevotedtothepositioncomputation(i.e.:ComputePVTandlocalintegritymodule).

3.1 MODULEOVERVIEW

TheAMcanbeseenascomposedbytwomainparts the firstmanagesthe interfacewithEDASincludingEDASdatastorage,thesecondcontainthecorealgorithmsforthepositioncomputationandaugmentationandmanagestheinterfacewiththeI-REACTORbackend.

The positioning augmentation process starts with a usermaking a request to the RESTful webservicesexposedbytheAM.GNSSreceiversparameters(TOWandrawdatacomingfromtheGNSSreceiver)arereceivedthroughthewebservices(JSONformatisused).Aftervalidatingtherequest,theAMretrievestherequireddata(originatingfromEDAS)fromthecloudstorageandrunstheaugmentationalgorithms. It returns theaugmentedposition including integrity information.TheoverallprocessofacquiringthedataandprovidingtheserviceissketchedinFigure3-1.

Inthefollowingparagraphs,theelementscomposingtheAMaregropedintwofunctionalpartsanddescribedindetail.



Figure3-1:MainsoftwarecomponentsanddataflowsoftheAugmentationModule

3.2 EDASSERVICESELECTOR,DECODERANDSTORAGE

ThissoftwarecomponentrunsontheAzurecloudasawebjob,whichisabletoconnecttotheEDASClientSWserverandcontinuouslyreceivetheEDASstreamdatafromtheEDASClientSWSL2inRTCMformatthroughinternetsocketconnection.

TheserviceselectorconnectstotheEDASClientSW(partofEDAS)torequesttheSL2,consequentlytheEDASdatadecoderalgorithmparsesthecontenttogettheEDAScorrectiondatatobestored.DatastorageisimplementedwithAzureTablewhichisaservicethatstoresstructuredNoSQLdatainthecloud,providingakey/attributestorewithaschemalessdesign.Thisdesignguaranteesagooddegreeofflexibilityallowingfastadaptationtothealgorithmicdesignevolutions.

Thedecoderalgorithmworksstepbystepasfollows:

• ReadsthebinaryRTCMfileo Thefirst8bitsarethepreamble;o Thenext6bitsindicatingthemessagetype;o Thelast24bitsareassignedtotheCyclicRedundancyCheck(CRC).

• Basedonthemessagetype,identifiestheEGNOSdatacontained.Thelengthofbitsshouldbe takenvariesbasedon themessage type.Afterdecoding, theextracted information isstoredintointernaldatastructure.HerearesomeSL2messagetypes:

o Message1004.ExtendedL1&L2GPSRTKobservables.o Message1005.StationaryRTKreferencestationARP.o Message1007.Antennadescriptor.o Message1010.ExtendedL1-onlyGLONASSRTKobservables.



o Message1013.Systemparameters.o Message1019.GPSephemerides.o Message1020.GLONASSephemerides.o Message4085whichcanbeofdifferentsubtypes:

§ Subtype0.GPS/GLONASS/GEOEphemeris.§ Subtype1.GEOObservations.§ Subtype2.NLESCyclicFeedback.§ Subtype3.ATCInformation.§ Subtype4.RIMSAPCdata.§ Subtype5.IonosphericandUTCdata.§ Subtype6.GPSAlmanac.§ Subtype7.RIMSStatus.

MessagesexploitedbytheAMareMT4085(subtype2)forEGNOScorrectionandMT1019togetGPSsatelliteclockandephemeris information. InSL2,allEGNOSdefinedmessagesareencodedwithMessageNumber4085(EDASproprietarymessage),andcanbedistinguishedbythemessagesubtypeasthetablebelow.

Message4085

Subtype2:NLEScyclicfeedback

EGNOSMessageTypes Messagenames

1 PRNMask

2-5 Fastcorrections

6 Integrityinformation

7 Fastcorrectiondegradation

10 Degradationfactor

18 IGPmask

24 Longtermandfastcorrections

25 Longtermcorrections

26 Ionosphericdelaycorrections

27 Servicemessage

Table3-1:Message4085subtypes.

TheflowchartinFigure3-2explainshowastreamofEDASdataisdecodedandstoredinthetable.



Figure3-2:EDASstreamdecodingandstorage



3.3 AUGMENTEDPVTANDINTEGRITYCOMPUTATION

Theaugmentationalgorithmruns inAzurecloudasacontinuouswebjob. It isusedtocomputeuserspositionsusingleastsquarealgorithmorKalmanfilteranditisabletoaugmentpositionusingEDAScorrectiondata.Itisalsousedtodeterminetheconfidenceintervalsaroundthecomputedpositions.

Thealgorithmrequirestwoinputs,namelyastreamofGNSSrawdata(i.e.:pseudorange,Dopplermeasurementsandsignalcarriertonoiseratio)andEDAScorrectiondata.InrealtimeprocessingtherawGNSSdatacomeinJSONformatfromtheI-REACTORbackendand,inthesameformat,theresultsaresentbacktotheI-REACTORbackend.DependingofthereceivedGNSSdata(includingtime)matchingEDASdataareretrievedfromtheAzureTable.

ThealgorithmfordeterminingPVTusingEDASaugmentationworksstepbystepasfollows:

• GNSSRawdatareceptioninJSONformatanddecoding;• retrieval of the EDAS correction messages from the Azure Table storage based on

timestamps;• exclusion of pseudoranges from satellites declared by EGNOS as Do Not Use and Not

Monitored;• determination of satellite clock corrections and the satellites positions from the GPS

ephemerisdataprovidedbyEDAS;• computationofEGNOScorrectionsusingthethreesetsofcorrections,whicharebroadcast

tousers:o Fast corrections – used to compensate short-term disturbances in GPS signals,

generallyattributabletosatelliteclocks;o Long-termcorrections-usedtocompensateforthelong-termdriftinsatelliteclocks

andtheerrorsinthebroadcastsatelliteorbits;o Ionosphericcorrections–broadcastasverticaldelaysforagridofpointsfromwhich

it is possible to determine a slant correction to be applied on each rangemeasurement to compensate for thedelayexperiencedby the signal as itpassesthroughtheionosphere;

• application of the corrections on the pseudoranges and satellite position for each GPSsatellite;

• PVT computationusing correctedpseudoranges andother data selected from theAzureTable(ephemeris,clockparameters).ThecomputationcanbedoneusingKalmanfilterorleastsquaremethod;

• computationofsatelliteselevations:thesedataisusedtogetherwiththeCarriertoNoiseratio(C/N0)estimatedbytheGNSSreceiveranddeliveredalongtherawdata,areusedtoestimate the signal reliability used to compute PLs.This step is fundamental, because itprovidestheinputforimplementingthelocalintegrityalgorithm(seethenextstep);

• computationofthelocalintegrity.Thissteprepresentstheaddedvaluewithrespecttothestandard implementation (i.e. aviation integrity). The output are confidence levels, here



namedHPLandVPLalongtheaviationnomenclature.Incasetheusermotionisrelevant,instead of the circle defined by the HPL, an ellipse with its axes oriented along andperpendicularlytotheuserpathcanbedefined.Theseaxesarethecrosstrackandalongtrack PLs and are computed taking into account user motion direction and satellitesgeometryandreliability;

• deliveryofaugmentedpositionandintegrityinformationinJSONformattotheI-REACTORbackend.

TheflowchartshownbelowexplainsthewholeflowofPVTandPLcomputation.

Figure3-3:Augmentationalgorithmflowchart.



4 GNSSCHIPSETSELECTION

TheI-REACTprojectforeseesthedevelopmentofawearabledevicetoequipon-fieldprofessionalsenrolledinemergencymanagementservices.ThischoiceassuredtheavailabilityofGNSSrawdataandthusthepossibilitytodeveloptheAMasdescribedinChapter3.NeverthelessitisimportanttohighlinethattheI-REACTAMwasconceivedinordertoprovideintegrityinformationtoeveryuser,evenifwithdifferentlevels,thatdependontheequipmentavailable.

After the project has been conceived, several smartphone producers started to delivermodelsprovidinginoutputGNSSrawdata[RD10].ThisfurtherwidenedtheexploitabilityoftheI-REACTAM.InfacttherawdataavailabilityenablestheapplicationofEGNOScorrections,henceabetterpositioningaccuracy.

In any case the wearable device still holds its role within the project from the positioningperspective: an accurate selection of the GNSS chipset and the antenna guarantees betterperformanceswithrespecttotheonesachievablewithsmartphones.

ApreliminaryanalysisofGNSSchipsetsavailableonthemarkethasbeenaddressedtoselecttheproperonestobeintegratedintheI-REACTwearabledevices.

Themaindriversforthechipsetselectionhavebeentheavailabilityofspecificdataandthepreviousexperienceconcerningtheintegrationaspects.

Asitwillbebetterexplainedinthefollowingtheuseofamulti-constellationreceiverabletotrackGalileohasbeenconsideredasmandatory.

4.1 GNSSDATA

Inordertoreachtheobjectivespursuedbytheproject,ithasbeennecessarytoexploitawidersetofdatawithrespectthemerepositionprovidedbyeveryGNSSchipsetonthemarket.

Firstly,thecomputationofpositions,exploitingtheEGNOSaugmentationdataprovidedviaEDAS,cannotbeperformeddirectlybyaGNSSreceiver(orbetter:nomassmarketdeviceonthemarkethasthisfeature).Inordertodothis,itisnecessarytohaveaccesstothereceiverrawdata.Secondly,data are also needed to provide an estimation of the position reliability, i.e. for the integritycomputation.Inthiscase,asitwillbedetailedinChapter5,somedatahavebeenexploitedduringthealgorithmdevelopmentphase,whileothersareneededtorunit.

Datarequiredfromthereceiverduringoperations:

• Pseudoranges• CarriertoNoiseratio• Dopplermeasurements

Datarequiredfromthereceiverduringalgorithmtuning:



• Pseudorangeresiduals1• Satelliteazimuthandelevation• CarriertoNoiseratio

GiventhepreviousexperienceofISMBwithu-bloxreceiverssomeevaluationkitshavebeenusedtoperformsometestsindifferentenvironmentsinordertocollectstheaforementioneddata:

• C94–MP8• EVK-M8U• EVK-M8QCAM

4.2 INTEGRATION

VerificationoftherequirementsfortheintegrationofthechipsetandtheantennaontheI-REACTwearabledevice.

Themaindrivertakenintoconsiderationhasbeentheprovisionofapositioninformationwhichisbetterbyfarwithrespecttotheoneprovidedbyagenericsmartphone.ApartfromthespecificI-REACT implementation, the greatest difference can be done by the antenna used by theGNSSreceiver:usuallyhand-helddevicesnotspecificallydesignedasnavigatorsareequippedwithalinearantenna(i.e.withlinearpolarization)havinganomnidirectionalpatternandasmallgroundplane,whileatypicalGNSSpatchantennaisrighthandcircularlypolarizedwithanhemisphericalpatternwhichprovideasensibleadvantage;furthermore,inthecaseofanexternalpatchantenna,abiggergroundplanecanbeeasilyadded,theonlylimitationgivenbyitsoverallsize.

AsitwillbeclearinChapter5,thealgorithmdevelopmenthasbeenbasedonthedatacollectedwithau-bloxreceiverabletodeliveralsoinformationaboutpseudorangeresiduals,whileforthedevice operation theparameters usually providedby theGNSS receivers (carrier to noise ratio,azimuthandelevation)areenough.

4.3 MULTICONSTELLATIONGNSSRECEIVERS

ThelastdecadesawthegrowthofGNSSalternativeorbettercomplementarytoGPS.TheRussianGNSS,GLONASShasbeenbroughtbacktoitsfullpotential,Galileo,theEuropeanGNSS,isinthemiddle of its deployment [RD11], but it presents new interesting features, while the Chineseconstellation,namedBeidou,itisalmostcomplete.

GNSSreceiversproducersareexploitingthepossibilitiesprovidedbythecontemporarypresenceofthese systems. Consumer grade GNSS receiver manufactures found a cost benefits balance in

1 Pseudorange residual: the difference between the expectedmeasurement and the observedmeasurement. Theexpectedmeasurementisthedistancebetweenthesatelliteandthecomputedposition.Seeparagraph5.2forfurthersdetails.



exploitingasinglefrequencyfromtwoormoresystems.Themainbenefitofthisapproachistobeabletoexploitabiggernumberofsatellitesallowingweightingorselectionstrategies,inordertoavoidtheuseofsatellitesmoreaffectedbyerrors.Moresimply,incaseofnaturalorurbancanyonstheavailabilityofmoreconstellationsenablesthepositioning itself,when, forexample, theGPSsatellitesinviewarenotenough(atleastfoursatellitesareneededforpositioncomputation,ifnoadditionalinformationisprovidedbyothersources,asanbarometricaltimeter).AstudyconductedbytheEuropeanGNSSAgency(GSA)andRxNetworksshowsthatwhenusedjointlywithGPSand/orGLONASS,Galileosignificantlyimprovesthepositionaccuracyinchallengingenvironments,suchasurbancanyonorindoor[RD12].ForsuchreasonsareceiverincludingGalileoreceptioncapabilitieshasbeenconsideredmandatoryintheframeofI-REACT.

Basically,theadvantagesprovidedbytheexploitationofmoreGNSSsareduetoredundancy,butthisdoesn’tprovideanyguaranteeaboutpositioning reliability: fromhere theneedof integrityinformationwhichhasbeenthemainabjectofthedevelopmentscarriedoutforwhatconcernspositioningwithinI-REACT.



5 LOCALINTEGRITY

ThepresentchapterdescribesthedevelopmentsperformedwithinI-REACTforthelocalintegritycomputation. The idea at the basis of the development is to the joint use EGNOS information(throughEDAS)with local informationextractedfromtheprocessingof thereceivedsignal.Thisallows for the computation of a confidence interval of the computed position. Following thegeomatics fundamentals (i.e.:dataandmeasurementsassociatedtoageo-localizedpointof theenvironment), the confidence interval introduces an estimate of the level of reliability of geo-localizedmeasurements.

5.1 STATEOFTHEARTANDLATESTDEVELOPMENTS

Integritycanbeconsideredas“stateoftheart”onlyintheaviationfield:aquickoverviewaboutithasbeenprovidedinSection2.2.

The work performed in the frame of I-REACT has been based on the local integrity conceptintroducedbytheauthorsin[RD13].Initsturn,thisstudy([RD13])wastriggeredbypreviousworks([RD14],[RD15]) about the limits of applicability of the aviation-born integrity to othertransportationfieldsandtheneedofadeepreconsiderationoftheapproachtoeffectivelyexploititinnon-aviationoperations.Thisnovelcooperativeintegritymonitoringconceptwasconceivedtobe suitable to automotive applications in urban scenarios. The idea is to take advantage of acollaborative Vehicular Ad hoc NETwork (VANET) architecture in order to perform aspatial/temporal characterization of possible degradations ofGlobalNavigation Satellite System(GNSS)signals.Suchcharacterizationenables thecomputationof theso-called"LocalProtectionLevels", taking into account local impairments to the received signals. Starting from theoreticalconcepts,thispaperdescribestheexperimentalvalidationbymeansofanintensemeasurementcampaign and the real-time implementation of the algorithm on a vehicular prototype. A livedemonstration in a real scenario was carried out successfully, highlighting effectiveness andperformanceoftheproposedapproach.

5.2 PROPOSEDALGORITHM

Thiswork[RD13]hasbeenusedasastartingpointforthedevelopmentofanewalgorithmbasedon the characterization of the behaviour and performances of GNSS receivers in differentenvironmentalconditions,whilethefirsthasbeenbasedonthecharacterizationoftheoperationalenvironment. Inotherwords, theobjectiveof theoriginalalgorithmwasthedeterminationofareliability index (σ2i) for themeasurements obtained by any receiver for any specific time andlocation.Thiswaspossiblebytheexploitationofaverypopulateddatabase,composedofthedatacollectedbyseveralGNSSreceiversmountedon-boardthevehiclestravellingaround.Intheframeofanemergencyscenario,thisapproachtothelocalintegritymaybenotapplicablebecausetheoperationstheatresortheareasofinterestcanbealmosteverywhereandanextensivepreventive



mapping (and creation of a corresponding data base) of the signals quality is unfeasible. Theapproachfollowedforthedevelopmentofthisgeneralizedversionofthelocalintegrityalgorithmforesawfourmainphases:

• tocollectabigamountofdataasdoneinthe[RD13]case.However,now,datasetsarenotassociatedwithaspecifictimeandlocation,buttheyarejustassociatedtodifferentgroupsofenvironmentalconditions.Theassumptiontobeverifiedisthattheeffectsonthesignalscanbeassociatedtotheenvironmentwheretheuserisoperating;

• to classify the kindsof environmenton thebasisof theelevationof the satelliteswhichsignalsarereceived;

• todefineasignaldegradationmodelforeachenvironmenttocomputeeffectiveprotectionlevels;

• to verify the goodness of the proposed approach checking if errors are bounded by theconfidenceintervals.

Theobjectiveof thenewalgorithm isaprobabilisticevaluationof themeasurements reliability,which isbasedonparametersasC/N0,azimuthandelevation thatarecommonlyavailable.Theintendedoutputisthevarianceoftheerrorcomponentduetothelocalenvironmentsurroundingtheantenna (i.e.:aσ2local) tobeused jointlywith theothervariances (i.e.: σ2) fromtheEGNOSestimatesconcerningGNSSsystemsandatmosphericerrors.

Thedevelopmentof thealgorithmstarted fromtheanalysisofsomedatasets taken indifferentenvironments(opensky,country,hillmostlycoveredwithvegetationandurban)withaconsumergradeGNSSreceiver(au-bloxLEO8U).Amongthemeasurementsmadeavailablebysuchareceivertheanalysiswasaddressedtothesocalledpseudorangeresiduals.

Inprincipleitispossibletocomputesuchvaluesoncethereceiverhascomputedaposition.ItisworthtorecallthatthepseudorangesaretherawmeasurementperformedbythereceiveroneachoftheincomingGNSSsignals.Thesearerangingmeasurements,buttheyarecalledpseudorangesbecauseofacommontermaffectingallthembythesameamount;thisbiasisduethemisalignmentofthereceiverclockwithrespecttotheGPStime(tiedtotheUniversalTimeCoordinated).AnyGNSSreceivercancomputeapositionusingtheknownpseudorangesfromatleastfoursatellitesandsolvinganon-linearsystemoffourequations,inordertodetermineitscoordinatesanditstimebias (unknowns). In casemoremeasurements are introduced, the achieved redundancy can beexploitedtoimprovethepositionestimation;thisisobtainedminimizingtherootmeansquareofthedifferencesbetweenthepseudoranges(oncethetimebiashasbeenremoved)andthedistancebetweentheestimatedpositionandtherelatedsatellites.Suchdifferencesarethepseudorangeresiduals which are related to the errors affecting the measurements [RD16]. Ideally, all theresidualsshouldbeequaltozero,howeverthedifferenterrorsourcesintroducesomebiasestothemeasurements,whichaffectthepositioncomputation.Therefore,itisimportanttohighlightthatthetruepseudorangeerrordoesn’tequaltheresidual:infactthefirstisthedifferencebetweenthereal satellite-receiverdistanceand themeasurement,while the second,beingestimatedby thereceiver,derivesfromthecomputedpositionwhichisaffectedbythepseudorangeerrors.Inany



casethepseudorangeresidualsprovideausefulinformationaboutthemeasurementsquality:whentheirvaluesarelowitislikelytohaveagoodpositionestimate,whereas,ingeneral,highvaluesaresignoferrorsinthemeasurementsandconsequentlyofworsepositionquality.

Theideaatthebasisofthealgorithmusestheinformationaboutresidualstoestimatetheerrorvariance of the user position. For each epoch, the set of residual variances is then combinedaccording to the satellitesgeometry inorder toprovideanestimateof thepositionqualityandconsequentlytheprotectionlevelsalongthedifferentdirections(alongtrackandcrosstrack).

InsteadofresortingtogenericanalyticalmodelsforaUserEquivalentRangeError(UERE)(providingσUERE,ivaluesforeachi-thsatelliteasithappensinthecaseofSBAS),withouttakingintoaccounttherealeffectsofnoiseandmultipathinaspecificconditions,itispossibletodefinean“effectiveUERE” parameter (σUERE,eff) as anensemble average of several “instantaneous” estimates of thecovarianceoftheresiduals.Asdetailedin[RD17],thevectorwcontainingthepseudorangeresidualscanbeputinrelationwiththe(non-observable)pseudorangeerrorsεthroughaprojectionmatrixS:

w=Sε

ItcanbedemonstratedthatitispossibletoobtainanestimateoftheeffectiveUEREthrough

𝜎"#$#,&''( =diag Σ/0 Σ/𝑁234 − 4

(1)

whereNsatisthenumberofsatellitesusedtocomputethepositionandΣ/isanestimatoroftheresidualscovariancematrixdefinedas

Σ/ =1𝑀

𝐰:𝐰:0

;

:<=

(2)

whereMisthenumberofindependentobservationsandwnisthen-thobservationsvector.

Startingfrom(1),confidenceintervalscanbecomputedtakingintoaccountthesatellitegeometryandthemotiondirectionoftheuserasdescribedin[RD17].TheyaretheAlongTrackProtectionLevel(ATPL)thatboundtheerrorinthedirectionofmovement,theCrossTrackProtectionLevel(CTPL)andtheVerticalProtectionLevel.Thesevaluescanbeusedtodefineanellipsecanteredatthecomputeduserpositionandcontainingthetrueuserpositionwithacertainprobability. It isorientedaccordingtothedirectionofmotion.

Atthispointitisworthtohighlighttwofactsthatdrovethealgorithmpracticalimplementation.The first is that the information about residuals is seldomprovided byGNSS receivers (at leastconsumergradeones), fromherethenecessitytoexploitother informationaboutsatellitesandsignalsmoreeasilyavailable.For this reasononeof themainactivitiesperformedtosetup thealgorithmwasaddressedtocharacterizedifferentenvironmentsusingasinputstheelevationofthesatellitesabovethehorizonandthesignalstrength,namelytheC/N0andmakingresidualensembleaveragesforasetofelevationandC/N0valuesintervals.Forexamplealltheresidualvaluescollectedforallthesatellitesinaspecificenvironmentandwithanelevationof10°areaveragedasdescribedby (11), note that satellite elevation resolutionusuallyprovidedbyGNSS receivers is 1°.During



currentoperationthealgorithmreceivesininputsatelliteselevationandC/N0valuesandassociatestoeachsatellitetheaveragedresidualvalueforthespecificenvironment;giventhissetofresiduals,itispossibletocomputethePLsasdescribedin[RD13]blendingtheresidualvariancescomingfromtheevaluationbasedontheelevationandontheC/N0.

(a)

(b)

Figure5-1:Resultsofdatacollectionsforthealgorithmtuning.(a)and(b)showresidualssinglevalues,theiraverageandaquadraticfittotheaveragevaluesinthecaseofavegetatedroadonthehillsrespectivelytowardssatelliteelevationandC/N0.Augmentationalgorithmflowchart.

Figure 5-1 shows how two different data collections yielding averaged values of residuals fordifferentsatelliteelevations(a)andC/N0values(b).

Thesecondfactthatdrovethealgorithmpracticalimplementationislessmanifestbeingrelatedtothepossibilityofsystemdisruptionsthatoriginfromproblemsatsystemlevel(global):theseeventseventhoughextremelyrarecanhaveabigimpactonpositionintegrity.Indeed,SBASareconceivedtoprovidetimelywarning incaseoftheseeventsthatcan involveoneormoresatellites: this isachievedthroughtheconstantreal-timemonitoringoftheGNSSconstellations.Thedevelopmentof the algorithm was addressed to the estimation of signal degradations at local level. Thesedegradationsareindependentfromsystemlevelproblemsthat,ifpresent,arefilteredout.Infact,thecharacterizationofsignalreceptioninthedifferentenvironmentalconditionswasbasedontheaveragingof the residualsvariancesoverextendeddatacollections.For these reasons localandglobaleffectsare taken intoaccountsimultaneouslyas it isdone in theclassical integrity,usingEGNOSinformation(throughEDAS),butreplacingthevariableswhichtakesintoaccountthelocaleffects(σ2air)withthenewones(σ2local).

5.3 VALIDATIONOFTHEALGORITHM

Inordertovalidatethealgorithm,fourtrialswereperformedalongfourpathrepresentativeofthedifferent environments: urban, hillwith vegetation, country and open sky. They lasted from30minutesto1houreach.ThreeGNSSreceiverswereusedforthevalidation:



• u-bloxLEO8T(configuredtouseGPSandGalileo)• u-bloxLEO8U(configuredtouseGPSandGalileo)• NV08C-CSM(configuredtouseGPSandGLONASS)

Furtheractivitywillbeperformedwiththeprototypewearabledeviceinitsfinalversion.

Thevalidationhasbeenperformedusingareferencepath(groundtruth)computedwithaNovatelSPAN-CPT receiver [RD18], which integrates GNSS measurements with an inertial sensor andprovide a sub-decimetric accuracy and using Google Earth orthophotographs to have a visualinformation.

Figure5-2:A25km long track zoomed in themost criticalpoint (i.e.worstoff-track).Greenarrowpoints to thecomputedpositionwhiletheredarrowpointstotherealposition.Realpositionfallscorrectlyinsidetheellipse.

Figure5-3:A10kmlongurbantrackzoomedinthemostcriticalpoint.Greenarrowpointstothecomputedpositionwhiletheredarrowpointstotherealposition.Realpositionfallsinsidetheellipse.Inthiscasethedatawerecollectedusingabikepassingmorethanoncealongthesamestreets:thisexplainsthehighellipsesdensity.

Thedifferencebetweenthecomputedpositionwiththegroundtruewascomputedandcomparedwiththeconfidenceintervalvalues.TheorthophotographsshowsthateachpointbelongingtothepathcoveredduringthetestiscontainedwithintheellipsedefinedbythecomputedpositionandthePLsasshowninFigure5-2inthecaseofthevegetatedhillandinFigure5-3inthecaseoftheurbantrack.



Table5-1reportsanexampleoftheresultsusedforthealgorithmvalidation.

Opensky Country Wood(hill) Urban

Duration(@1Hz)[s] 1800 2451 2367 2842

Averageerror[m] 1.2 1.8 2.5 4.3

Maximumerror[m] 5.1 7.5 8.2 14.9

AverageHPL[m] 5.56 6.5 9.01 20.26

MaximumHPL[m] 21.5 27.81 35.32 54.75

MinimumHPL[m] 4.64 5.02 5.21 10.67

SVinview(average) 13 12 10 9.6

Minimumerror-PLdifference[m] 1.8 1 0.41 3.6

Table5-1:Datawithublox8T.

Theprocessingofthedataacquiredduringthevalidationtestdidn’tshowanycriticalpoint.Fromtheanalysisofdataitispossibletoseethatinthecaseofthetestalongthevegetatedtrack(wood)theerrorwasboundedbytheconfidencelevelbyonly41cm(seeFigure5-2),whileintheothercases themargin ismore consistent,meaning that there is still some possibility to reduce theconfidencelevelsvalues.



6 CONCLUSIONS

TheworkcarriedoutintheframeofTask3.6oftheI-REACTprojectrepresentsanapplicationofageneralizedversionof theLocal integrityalgorithmdescribed in [RD13]. Itproved toprovideaneffectivemeasureofapositioningsystemreliabilityinmostcircumstances,whichcanbeassociatedalso toemergencyevents . The jointexploitationof the local integrity conceptwith theEGNOSintegrityallowstoprovideaguaranteealsoinparticularsituationsinvolvingGNSSsignalproblemsthatarenotconnectedwiththelocalenvironmentlikeseveresignaldegradationduetoionosphericpropagationorproblemsatsystemlevel.CurrentlysystemsotherthanGPSarenotmonitoredbyEGNOS,buttheycanbeeasilyincludedinthealgorithmtoincreaseitsreliabilityonceEGNOSwillincludeotherGNSS.



END OF THE DOCUMENT

Date post:	20-Jul-2020
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

IMPROVING RESILIENCE TO EMERGENCIES THROUGH … · GSA. Galileo increases the accuracy of location...

Documents