419
1 INTRODUCTION
An uninterrupted information about the user’s
position can be obtained generally from specialized
electronic position‐fixing system, in particular,
satellitenavigationsystem(SNS).Atthetimeofthis
writing(January2017)twoglobalSNSs,theAmerican
GPS and the Russian GLONASS, are fully
operational,two next, also global, G
alileo in Europe
and BeiDou in China are under construction
(www.globaln‐ews.ca, www.gpsworld.com,
www.insidegnss.com).
InthecaseofeachSNStheaccuracyoftheuser’s
position determined using a single standalone
receiverisaffectedbythethreemainfactors:
position’s accuracy of each satellite used for
calculation,
accuracyofpseudorangemeasurements,
act
ualsatellitegeometry.
Theuser’spositionerrorisafunctionofboththe
pseudorange error called UERE (User Equivalent
RangeError)anduser/satellitegeometryexpressedby
Dilution Of Precision (DOP) coefficient. Assuming
that the measurements errors for all satellites are
identicalandindependent,theUEREmaybedefined
as the root square of the va
rious errors and biases.
MultiplyingtheUEREbytheappropriateDOPvalue
produces the expected precision of the GPS
positioningatthe one sigma level(el Rabbany 2016,
Misra & Enge 2006). The best guideline for position
erroranalysesispseudorangeerrorbudget.
2 USEREQUIVALENTRANG
EERROR
Overall UERE is the ranging error along the vector
betweentheuser’sreceiverandaparticularsatellite.
Thiserrorisdecomposedintotwotypesoferrors:the
Sources of Error in Satellite Navigation Positioning
J
.Januszewski
GdyniaMaritimeUniversity,Gdynia,Poland
ABSTRACT:Anuninterruptedinformationabouttheuser’spositioncanbeobtainedgenerallyfrom satellite
navigation system (SNS). At the time of this writing (January 2017) currently two global SNSs, GPS and
GLONASS,arefullyoperational,twonext,alsoglobal,GalileoandBeiDouareunderconstruction.IneachSNS
the accuracy of the user’s position is affect
ed by the three main factors: accuracy of each satellite position,
accuracyofpseudorangemeasurementandsatellitegeometry.Theuser’spositionerrorisafunctionofboththe
pseudorangeerrorcalledUERE(UserEquivalentRangeError)anduser/satellitegeometryexpressedbyright
Dilut
ionOfPrecision(DOP)coefficient.Thiserrorisdecomposedintotwotypesoferrors:thesignalinspace
rangingerrorcalledURE(UserRangeError)andtheuserequipmenterrorUEE.ThedetailedanalysesofURE,
UEE,UEREandDOPcoefficients,andthechangesofDOPcoefficientsindifferentdaysarepresentedinthi
s
paper.
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 11
Number 3
September 2017
DOI:10.12716/1001.11.03.04
420
signalinspacerangingerrorcalledURE(UserRange
Error)or SISRE(SignalInSpace Ranging Error) and
theuserequipmenterrorUEE.
22
UERE URE UEE
(1)
URE accounts for pseudorange error due to the
spacesegmentandterrestrialsegmentwhileUEEfor
effectsofpropagationandreceiverprocessing.Inthe
caseofUREdominatethreedifferenterrorsources–
clock error (estimation, prediction, curve fit and
stability),ephemeriserror(estimation,predictionand
curvefit)and
satellitegroupdelayerror.Additionally
beforeMay2,2000UREofGPS(oneonlyglobalfully
operational SNS at that time) was dominated, very
significantly, by Selective Availability (SA), the
intentional distortion of the civil signal’s clock
correctionandephemerisparameters.
Ephemeriserrorsresultwhennavigationmessage
does not transmit the
correct satellite location. It is
typical that the radial component of thiserror is the
smallest:thetangentialandcross‐trackerrorsmaybe
larger by an order of magnitude. Fortunately, the
largercomponents do not affect rangingaccuracyto
thesamedegree.FundamentaltoSNSistheone‐way
rangingmeasurementthatultimatelydependsonthe
clockpredictability.Thesesatelliteclockerrorsaffect
boththeC/AandPcodeusersinthesameway.
In the case of UEE dominate five distinct error
sources–ionospheric,tropospheric,multipath,group
delay and noise and interference [Beitz 2016]. To
eliminate ionospheric error
single frequency SNS
receivers apply model of ionosphere with eight
coefficients transmitted in navigation message by
satellites.IftheGPSreceivercanworkindifferential
modethe PseudoRangeCorrections(PRC)
transmitted by reference stations can be used. The
corrections to the clock, ephemeris and ionospheric
information can be provided
by geostationary
satellites of SBAS (Satellite Based Augmentation
Systems) also. That’s why the number of integrated
SNS/SBAS receivers increases since few years
incessantly.DualfrequencySNSreceiverscanusethe
simultaneouslymeasurementsofpseudorangeontwo
transmitted satellite frequencies (Januszewski 2012,
Misra&Enge2006).
Unlike the ionosphere, the troposphere is
non‐
dispersive medium for all frequencies carrier of all
SNSandSBAS.Therefractiveindexndoesn’tdepend
upon the frequency of the signal. As this index is
larger than unity, the speed of propagation of SNS
and SBAS signals is lower than in free space and,
therefore, the apparent
range to a satellite appears
larger.Thephaseandgroupvelocitiesonallsatellites
frequencies,themeasurementsofcodeandcarrierof
allthesefrequenciesexperienceacommondelay.The
disadvantages is that an elimination of the
tropospheric refraction by dual frequency methods,
unlikeionosphericdelayisnotpossible(Januszewski
2013,
Misra&Enge2006).
Multipathis mainly caused byreflectingsurfaces
near the user’s receiver, a satellite emitted signal
arrives at the receiver by more than one path.
Secondaryeffectsarereflectionsatthesatelliteduring
signaltransmission.Incodetrackingtheresultcanbe
tensorevenhundredsof
metersoferror.Toreduceor
estimatethemultipatheffects,variousmethodswere
developed – antenna‐based mitigation, improved
receiver technology and signal and data processing
(Beitz2016,Hofmann‐Wellenhof2008).
Groupdelayerrorsoccurintheuser’sreceiver.In
the case of CDMA (Code Division Multiple Access)
receiver (currently GPS,
in the future Galileo and
BeiDou) these errors are very small (fractional
nanosecond)whenallmeasurementsaremadeusing
asinglesignaltype.InthecaseofFDMA(Frequency
Division Multiple Access) receiver (currently
GLONASS) the signals from different satellites have
different center frequencies and consequently
experience different group delays
that can be
significant.
Noiseandinterferenceerrorsproducejitterinthe
estimatesofasignal’stimeofarrival(Beitz2016).
FinallywecansaythatinthecaseofGPSsystem
duringlastyearsUREdecreasedconsiderably,i.e.in
2001–1.6m,in2014–0.7m.Inthis
periodStandard
PositioningService(SPS) Performance Standard (PS)
changedalso;before2008itwas6mRMS,lateritwas
considered that equivalent RMS decreased to 4 m
only. Additionally in 2008 the worst of any healthy
satelliteitwas7.8m(95%),currentlyitisabout3m
(95%)
only(Munchen2016).
3 DILUTIONOFPRECISIONCOEFFICIENTS
Satellitenavigation system (SNS) positioning
accuracy is measured by the combined effect of the
unmodeledmeasurementerrorsandtheeffectofthe
satellitegeometry.Thelasteffectcanbemeasuredby
a single dimensionless coefficient called the dilution
ofprecision(DOP).Alow
DOPcoefficientrepresents
abetterpositionalprecisionduetothewiderangular
reparative between the satellites used to calculate a
user’sposition.Thelowervalueofthiscoefficient,the
better the geometric strength, and vice versa. DOP
value is computed based on the relative user’s
receiver–satellitegeometryat
anyinstance,thatis,it
requiresthe availability of both the receiver andthe
satellitecoordinates.Duetotherelativemotionofthe
satellites and the user the value of the DOP will
changeovertime(Forsell2008).
The various DOP forms are used, depending on
the user’s need. GDOP (Geometric
Dilution of
Precision) coefficient determines the multiplication
factor of the estimated distance measurement error
fortheestimationoftotalpositionandtimeerrors.Its
value is the root sum square of the variances along
the coordinate axes, that’s why the quantity can be
regardedasthedistancebetweentwopoints
inspace
and,consequently,itisindependentoftheselectionof
coordinatesystem(Prasad&Ruggieri2005).
TheGDOPrepresents the combined effect of two
coefficients: PDOP (Position Dilution Of Precision)
andTDOP(TimeDilutionOfPrecision).PDOPcanbe
broken into two components: HDOP (Horizontal
DilutionOfPrecision)coefficientand
VDOP(Vertical
Dilution Of Precision) coefficient. PDOP represents
the contribution of the satellite geometry to the 3D
positioning accuracy and HDOP the contribution to
421
the2Dpositioningaccuracy.Becausethereceivercan
track only those satellites above horizon (H
min > 0
O
),
VDOPwillalwaysbelargerthanHDOP.
In order to know the distribution four DOP
coefficients,VDOP,HDOP,PDOPandGDOP,values
of all four global mentioned above SNSs author’s
simulatingprogramwasused.Thecalculationsbased
on reference ellipsoid WGS−84 were made for the
observer at latitudes 50 – 60
O
for masking elevation
angle H
min = 5
O
(the most frequently used value in
SNS receiver) and H
min = 25
O
(representative for the
positioning in restricted area where the visibility of
satellites can be limited). The parameters of spatial
segment, the number of MEO satellites and time
intervalofconstellationrepeatabilityofallfourSNSs
are presented in the table 1. The geographical
longitudeofascendingnodeandargumentof
latitude
ofallGPSandGLONASSsatellitesweretakenfrom
currentconstellationsoftheseoperationalsystems,in
thecaseofGalileoandBeiDouitwasnominalfuture
constellation.
Table1. Global satellite navigation systems, parameters of
spatial segment and time interval of constellation
repeatability (www.beidou.gov.cn; www.glonass‐
ianc.rsa.ru; www.gps.gov; European GNSS, 2016,
www.gsc.europa.eu).
_______________________________________________
System Orbit Orbit Numberof Number
altitudeinclinationorbitalperiods/ ofMEO
[km] [
o
]timeinterval satellites
[min]
_______________________________________________
BeiDou 21,500 5513/10,091.48 27
Galileo 23,222 5617/14,360.75 24
GLONASS 19,100 64.8 17/11,488.44 24
GPS20,183 552/1435.94 31
_______________________________________________
For each system and for each angle (H
min) one
thousand (1000) geographic–time coordinates of the
observerweregeneratedbyrandom–number
generatorwithuniformdistribution:
latitudeinterval0–600minutes(10O),
longitudeinterval0−21600minutes(360O),
timeintervalinminutesequaltimeofconstellation
repeatability(table1).
For each geographic–time coordinates the values
of
all four DOP coefficients were calculated. This
value (v) was divided for H
min = 5
O
into 5 intervals
(1≤v<1.5,1.5≤v<2,2≤v<3,3≤v<4,4≤v<8)andforH
min=
25
O
into8intervals(v<2,2≤v<3,3≤v<4,4≤v<5,5≤v<6,
6≤v<8,8≤v<20,v≥20).
Distributions of Dilution Of Precision (DOP)
coefficients values at latitudes 50 – 60
O
for all four
globalSNSformaskingelevationangleH
min=5
O
and
25
O
arepresentedinthetable2and3respectively.We
cansaythatinthefirstcase:
DOPcoefficientvalueisforallSNSsgreaterthan1
andlessthan3exceptforBeiDou(lessthan4),
thepercentageofthelowestvaluesofHDOP(less
than1.5)
isthegreatestforGPSsystem,
coefficientHDOPvalueisforallSNSsgreaterthan
1andlessthan3,
coefficients PDOP and GDOP are for all SNSs
greaterthan2andlessthan8,
andinthesecondcase:
asfortwoSNSs,GalileoandGPS,
thenumberof
satellitevisibleaboveHmin=25Ocanbelessthan
4 (3D position cannot be determined) No Fix (in
percentage) greater than 0, is equal 0.3 and 1.4,
respectively,
forallfourSNSsHDOPcoefficientcanbelessthan
2, coefficient VDOP is greater than 2
and PDOP
andGDOParegreaterthan3,thevalueofallDOP
coefficientscanbeforallSNSsgreaterthan20but
for GLONASS system this percentage is the
greatest,
thepercentageofHDOPcoefficientvaluelessthan
2 is for GPS system greater than for other three
SNSs,
considerably,
thepercentageofHDOPcoefficientlessthan3and
PDOP and GDOP coefficients less than 4 is the
lowestforGLONASSsystem.
Thenumberofsatellites(ls)usedinGPSorDGPS
positiondeterminationandthechangesofHDOPand
VDOPcoefficientvaluesin different days in Gdynia
arepresentedinthetable4.Alldatawereregistered
each minute. For all series of measurements the
number ls is for GPS system greater than for DGPS
system, in the case of DOP coefficients the both
externalvaluesofintervalareforVDOPgreaterthan
forHDOP.Additionallywecan
saythatiflsisgreater
bothcoefficientscanbeandinthemostcasesareless
andvice versa if ls is lower both coefficients can be
andaregreater(table5).
Table2. Distribution of Dilution Of Precision (DOP)
coefficient values (v) for satellite navigation systems,
BeiDou(BeiD),Galileo(GAL), GLONASS(GLO)andGPS,
elevationmask5
O
,latitude50–60
O
(ownstudy).
_______________________________________________
DOP SystemCoefficientvalue−v
__________________________________
1<v≤1.51.5<v≤22<v≤33<v≤44<v≤8
_______________________________________________
HDOP BeiD 3.287.4 9.4−−
GAL 4.375.6 20.1−−
GLO 4.171.1 24.8−−
GPS 6.989.9 3.2−−
VDOP BeiD 28.1 64.4 6.90.6−
GAL 23.8 67.8 8.4−−
GLO 16.9 68.9 14.2−−
GPS 41.6 48.6 9.8−−
PDOP BeiD−−88.7 11.2 0.1
GAL−−83.2 16.8 −
GLO−−78.8 21.2 −
GPS−−91.2 8.60.2
GDOP BeiD−−70.3 27.8 1.9
GAL−−68.7 30.8 0.5
GLO−−61.3 38.7 −
GPS−−79.5 19.7 0.8
_______________________________________________
ThevaluesofthreeDOPcoefficientsindicatedby
fivedifferentstationaryGPSreceivers(FurunoGP33,
LeicaMX420,MagnavoxMX200,SaabR5Su‐preme
Nav, Simrad MX512)locatedinGdynia for
differentmaskingelevationanglesH
minanddifferent
numbers of satellites visible and used for position
determinationarepresentedinthetable6.Inthecase
ofFuruno33receiverthelowestvalueofH
minis5
O
.
The stationary receivers destined for maritime
userscanindicate one (Furuno and Magnavox), two
(Leicaand Simrad) orthree (Saab) DOP coefficients.
From among five mentioned above receivers HDOP
coefficient is indicated by all except for Furuno,
VDOPbyLeica,Saab,Simrad,PDOPbyFurunoand
Saab.GDOPandTDOP
arenotindicated.
422
All receivers except for Magnavox have 12
channels (MX 200 – 6 only) for GPS pseudorange
mea‐surements but the number of satellites visible
(lv) and used (lu) for position determination is
different and these numbers depend on angle H
min.
Numbers lv and lu are the greatest (12) for Saab if
H
min = 0
O
and 5
O
. Additionally we can recapitulate
that:
HDOPvalueisthegreatestforMagnavoxforeach
H
min. This receiver has 6 channels only and
thereforethenumberofsatellitesusedforposition
fix and for calculate DOP coefficients cannot be
greaterthan6,
VDOP coefficient value is for each H
min greater
thanHDOPvalueinthecaseofallthreereceivers
(Leica, Saab, Simrad) indicating VDOP and the
differencebetweenthesevaluesisthegreatestfor
H
min=25
O
fortwolastreceivers,
some receivers, e.g. Furuno, indicate one DOP
coefficientvalueonly.AsinthecaseofFurunoGP
33allpositionswerecalculatedinmode3D,itwas
consideredthatthiscoefficientmeansPDOP,
ifH
min≤10
O
HDOPvalueisthelowestforSaab,if
H
min=15
O
and20
O
forSaabandLeica,ifHmin=25
O
forLeica,
ifH
min≤5
O
VDOPvalueisthelowestforSaab,if
H
min>5
O
forLeica.
Table5.Thenumberofsatellites lsusedin DGPSposition
determinationandDOPcoefficientsvaluesOctober4,2016
in Gdynia (54
O
31,0849’N, 018
O
33,2738’E), LeicaMX 420
NavigationSystemreceiver,H
min=5
O
(ownstudy).
_______________________________________________
Time[UTC] ls HDOP VDOP
_______________________________________________
11337 1.52.4
11348 1.01.4
11169 0.91.1
11179 1.72.4
_______________________________________________
All values of DOP coefficients presented in the
tables2and3forgivengeographic‐timecoordinates
werecalculatedbyauthorforthisconstellationoffour
satellitesfor which GDOP was the lowest. However
insomereceivers,e.g.MagnavoxMX200,theuserhas
the possibility to choose satellite selection
criterion.
Themostfrequentlyusedandrecommendedwasthe
best(thelowest)DOP.
Currently the receivers, particularly professional
models, use own and at the same time unknown
algorithmfortheDOPcoefficientscalculation.That’s
why in different receivers DOP values for the same
satelliteconstellationareingivenmomentandfor
the
samegeographiccoordinatesdifferent(table6),e.g.if
H
min = 15
O
the number of satellites visible by two
receivers,SaabandSimrad,wasthesame(12)butthe
number of satellites used for position fix was
different, 9 and 7, respectively. Because of this
differenceHDOPandVDOPcoefficientsvalueswere
1.0&2.0and1.2&2.3,respectively.
Table3.DistributionofDilutionOfPrecision(DOP)coefficientsvalues(v)forsatellitenavigationsystems,BeiDou,Galileo,
GLONASSandGPS,andNoFixinpercent,elevationmask25
O
,latitude50–60
O
(ownstudy).
__________________________________________________________________________________________________
DOP SystemNoFix Coefficientvalue−v
______________________________________________________________________________
[%] v≤2 2<v≤33<v≤44<v≤55<v≤66<v≤88<v≤20 v>20
__________________________________________________________________________________________________
VDOP BeiDou−−43.2 32.9 0.70.45.612.0 5.2
Galileo0.3−54.3 10.1 2.01.52.914.9 14.0
GLONASS−−42.4 24.3 3.62.52.73.720.8
GPS1.4−36.5 32.2 15.1 5.93.04.01.9
HDOP BeiDou−6.569.2 11.8 5.31.90.92.43.0
Galileo0.32.562.0 6.15.05.15.07.46.6
GLONASS−2.560.9 11.7 1.90.91.36.814.0
GPS1.417.2 56.1 15.0 3.51.71.82.01.3
PDOP BeiDou−−−42.1 33.2−2.414.9 7.4
Galileo0.3−−50.3 14.0 1.83.315.1 15.2
GLONASS−−−32.4 33.8 3.74.14.721.3
GPS1.4−−37.2 29.6 13.7 9.25.82.3
GDOP BeiDou−−−13.6 39.4 22.3 0.114.7 9.9
Galileo0.3−−15.3 43.3 5.72.712.7 20.0
GLONASS−−−4.447.5 15.4 4.46.222.1
GPS1.4−−18.4 31.0 20.1 16.6 9.43.1
__________________________________________________________________________________________________
Table4.ThenumberofsatelliteslsusedinpositiondeterminationandDOPcoefficientsvaluesindifferentdaysinGdynia
(54
O
31,0849’N,018
O
33,2738’E)forGPSandDGPSsystems,LeicaMX420NavigationSystemreceiver,Hmin=5
O
(ownstudy).
__________________________________________________________________________________________________
System DayUTCLsHDOPVDOP
__________________________________________________________________________________________________
DGPS October4,2017 0824−1224 7–90.9–1.7  1.1–2.4
October5,2017 0709–1108 8–90.9–1.2  1.2–1.8
October6,2017 0710–0910 7–90.9–1.2  1.0–2.1
GPSOctober6,2016 0911−1011 11−12 0.8–0.9  1.1–1.6
__________________________________________________________________________________________________
423
Table5.The numberofGPSsatellites andDOP coefficientsfordifferent receiversand differentmasking elevationangle
H
min,Gdynia(54
O
31,0849’N,018
O
33,2738’E),October17,2016(ownstudy).
__________________________________________________________________________________________________
Hmin ReceiverNumberofsatellitesCoefficient
_______________________________________________________________________
[
o
]visible usedHDOP VDOP PDOP
__________________________________________________________________________________________________
0 LeicaMX42012  91.01.5−
MagnavoxMX20011 61.3−−
SaaBR5SupremeNav 12  120.71.21.4
SimradMX51212 81.02.0−
5 FurunoGP−3312  10−−1.76
LeicaMX42012  91.01.4−
MagnavoxMX20011 61.3−−
SaaBR5SupremeNav 12  120.71.21.4
SimradMX51212 81.02.0−
10 FurunoGP−3312  9−−1.90
LeicaMX42011  81.01.5−
MagnavoxMX20010 61.3−−
SaaBR5SupremeNav 12  100.91.71.9
SimradMX51212 81.02.0−
15 FurunoGP−3311  8−−2.35
LeicaMX42011  81.01.5−
MagnavoxMX20010 61.3−−
SaaBR5SupremeNav 12  91.02.02.2
SimradMX51212 71.22.3−
20 FurunoGP−3311  8−−2.36
LeicaMX42011  71.01.6−
MagnavoxMX2008 51.5−−
SaaBR5SupremeNav 12  91.02.02.2
SimradMX51212 71.22.3−
25 FurunoGP−3311  7−−3.41
LeicaMX42011  71.01.6−
MagnavoxMX2006 51.5−−
SaaBR5SupremeNav 12  71.33.53.7
SimradMX51212 61.43.5−
__________________________________________________________________________________________________
4 CONCLUSIONS
before May 2, 2000 the greatest and at the same
timethemostimportantsourceofthepositioning
errorwasSelectiveAvailability(SA)afterthisday
theionosphere,currently(January2017)itisuser
segment,receiverperformanceparameters,
inthecaseofstand‐aloneGPSreceiverthe
greatest
influencehasits“professionality”, the number of
channelsinparticular,
in restricted area the multipath error must be
takenintoaccountbytheuserinthecaseofeach
SNSreceiver,
as currently the User Range Error (URE) is less
than 1 meter the horizontal position’s accuracy
depends mainly on the user/satellite geometry
expressed by DOP coefficient, ionospheric error
andinrestrictedareamultipatheffects,
DOPcoefficient value depends on the number of
satellites used for position fix, this number
depends on the receiver, its performance
parametersasthenumberofchannels,numberof
signals and frequencies
tracked, year of the
productionandmanufacturer,inparticular,
the tropospheric propagation delay is critical for
precise position and baseline determination, can
significantly degrade the SNS accuracy, in
particularintheheightcomponent.
REFERENCES
Beitz, J. W. 2016. Engineering Satellite‐Based Navigation and
Timing Global Navigation Satellite Systems, Signals, and
Receivers.NewJersey:JohnWiley&Sons,Inc.,Hoboken.
El−Rabbany, A. 2002. Introduction to GPS The Global
PositioningSystem.BostonLondon:ArtechHouse.
European GNSS Service Centre, 2016. Constellation
Information.
Forsell, B.
2008. Radionavigation systems. Boston London:
ArtechHouse.
Hofmann‐Wellenhof,B.etal.2008.GNSS–GlobalNavigation
SatelliteSystems GPS, GLONASS, Galileo, and more.Vien
NewYork:Springer.
Januszewski, J. 2012. How the Ionosophere Affects
Positioning Solution Using Terrestrial and Satellite
Navigation Systems. Communication in Computer and
Information Science,
Telematics in the Transport
Environments.no329.pp.249−257.
Januszewski, J. 2013. How the Troposphere Affects
PositioningSolutionUsingSatelliteNavigationSystems,
Communication in Computer and Information Science,
TelematicsintheTransportEnvironments,Activitiesof
TransportTelematics.no395.pp.275−283.
Misra,P.&Enge,P.2006.Global
PositioningSystem Signals,
Measurements, and Performance. Lincoln: Ganga‐Jamuna
Press.
MunichSatelliteNavigationSummit,Munich,2016.
Prasad, R. &Ruggieri, M. 2005. Applied Satellite Navigation
Using GPS, Galileo, and Augmentation Systems. Boston
London:ArtechHouse.
www.beidou.gov.cn[dataofaccess:16.01.2017]
www.globalnews.ca[dataofaccess:20.01.2017]
www.glonass‐ianc.rsa.ru[dataof
access:16.01.2017]
www.gps.gov[dataofaccess:16.01.2017]
www.gpsworld.com[dataofaccess:20.01.2017]
www.gsc.europa.eu[dataofaccess:16.01.2017]
www.insidegnss.com[dataofaccess:20.01.2017]