Precise Precise GNSS GNSS positioningpositioningThreatsThreats and and opportunitiesopportunities
Department of Mathematical Geodesy and PositioningDepartment of Mathematical Geodesy and Positioning
Delft University of TechnologyDelft University of Technology
The NetherlandsThe Netherlands
KeesKees de Jong de Jong
Simpósio Brasileiro de Geomática
UNESP, Presidente Prudente, July 2002
ContentsContents
Modernized GPS and Galileo
GNSS vulnerabilities
Integrated GPS/Galileo ambiguity resolution
Major error sources
Trends in RTK positioning
No Selective Availability (SA)
Second civil signal on L2
Third civil signal on new L5frequency (1176.45 MHz)
Switched off on 2 May 2000
GPS modernizationGPS modernization
Availability of new GPS signalsAvailability of new GPS signals
New civil signal on L2
Stronger signals on L1 and L2
First launch in 2003
Block IIR
L5 frequency (1176.45 MHz)
New civil signals on L2 en L5
Stronger signals on L1, L2 and L5
First launch 2005
Block IIF
Availability of new GPS signalsAvailability of new GPS signals
02468
1012141618202224262830
Feb
-03
Aug
-03
Feb
-04
Aug
-04
Feb
-05
Aug
-05
Feb
-06
Aug
-06
Feb
-07
Aug
-07
Feb
-08
Aug
-08
Feb
-09
Aug
-09
Feb
-10
Aug
-10
Feb
-11
Aug
-11
Nu
mb
er o
f sa
telli
tes
Block IIR (L1, L2)
Block IIF (L1, L2, L5)
Galileo orbits and frequency bandsGalileo orbits and frequency bands
Orbital planes 3
Satellites/plane 10
Semi-major axis 29900 km
Inclination 56°
Orbits
E1 1587-1591 MHz
E2 1559-1563 MHz
E4 1254-1258 MHz
E5 1164-1214 MHz
E6 1260-1300 MHz
Frequency bands
Satellite constellationSatellite constellation
GPS
Galileo
GNSS GNSS frequency allocationfrequency allocation
Galileo E5/A
GPS L2
Galileo E5/B
GlonassG2
Galileo E6
GPS L1
GlonassG2
ARNS
RNSS* RNSS*RNSS RNSS
ARNS ARNS
Galileo C1
RNSS
Gal
ileo
E3
GPS L5
Gal
ileo
E1
Gal
ileo
E2
Gal
ileo
E4
Upper L-BandLower L-Band C-Band
E5: 11
64-1
214 M
Hz
E4: 12
54-1
258 M
Hz
E6: 12
60-1
300 M
Hz
E2: 15
59-1
563 M
Hz
E1: 15
87-1
591 M
Hz
C1: 5
010-
5030
MHz
Lower L-band Upper L-band C-band
1210 1220 1230 1240 1250 1260 1270 1280 1290 1300 f [MHz]
L2
1227.60
E6
1278.75
1560 1570 1580 1590 f [MHz]
1575.42
L1 E1E2
1176.45
L5 (E5a) E5b
1176.45 1202.25
1160 1170 1180 1190 1200 1210 1220 f [MHz]
GPS civil signal
Galileo civil signal
Galileo commercial signal
GPS precision signal
Galileo precision signal
Non--civil signal
GPS and Galileo signalsGPS and Galileo signals
Galileo scheduleGalileo schedule
Official go-ahead on 26 March 2002
First experimental satellite in 2004(Galileo System Test Bed)
First four operational satellites in 2005-2006
Full operational constellation in 2008
See also http://www.europa.eu.int/comm/energy_transport/en/gal_en.html
Visible satellitesVisible satellites
0
3
6
9
12
15
18
21
24
0:
00:0
0.0
1:
00:0
0.0
2:
00:0
0.0
3:
00:0
0.0
4:
00:0
0.0
5:
00:0
0.0
6:
00:0
0.0
7:
00:0
0.0
8:
00:0
0.0
9:
00:0
0.0
10:
00:0
0.0
11:
00:0
0.0
12:
00:0
0.0
13:
00:0
0.0
14:
00:0
0.0
15:
00:0
0.0
16:
00:0
0.0
17:
00:0
0.0
18:
00:0
0.0
19:
00:0
0.0
20:
00:0
0.0
21:
00:0
0.0
22:
00:0
0.0
23:
00:0
0.0
Nu
mb
er o
f sa
telli
tes
GPSGalileoTotal
Location: Presidente Prudente Date: 11 July 2002 Minimum elevation: 10°
GPS system vulnerabilitiesGPS system vulnerabilities
Radio-frequency interference (RFI)
GPS testing
Ionosphere (solar maximum)
Spectrum congestion
Unintentional interference
See Volpe vulnerability report (august 2001)http://www.navcen.uscg.gov
GPS system vulnerabilitiesGPS system vulnerabilities
Jamming: denial of use of GPS
Spoofing: broadcast wrong GPS-like signals
Meaconing: rebroadcast GPS signals
System damage: satellites or ground control segment
Intentional interference
Errors, over-reliance, lack of knowledge/training
Human factors
JammingJamming
Simple, 1 watt: 10/85 km (loss/no acquisition)
GPS-like, 1 watt: 1000 km (no acquisition)
Characteristics
JammersJammers
Russian, 4 Watt, L1 and L2, GPSand GLONASS, US$ 4,000
USA, 1 Watt
Miniature L1 Miniature L1 jammerjammer
TNO-FEL: Physics research lab of the Netherlandsorganization for applied research
1 mW output
Range > 125 m
Power consumption 9 V, 30 mA
Size of one cubic inch feasible
TNO-FEL prototype jammer
Sources of interferenceSources of interference
Technical enthusiastsStudents or people that cannot withstand the challenge
CriminalsCar or cargo theft, free access to toll-roads
TerroristsBlock airports, rescue and police
MilitaryJam potential enemy to deny use of GPS, Galileo and Glonass signals
GPSGPS
GalileoGalileo
Precise GNSS positioningPrecise GNSS positioning
Carrier ambiguity resolution is the key to precise positioningwith a Global Navigation Satellite System (GNSS) such as
I n t e g r a t e dI n t e g r a t e d
GPS/GalileoGPS/Galileo
GNSS processing stepsGNSS processing steps
LAMBDA method‘Float’ solution ‘Fixed’ solution
Least squares Integer least squares Least squares
Estimateposition andcarrierambiguities
Estimateposition(ambiguitiesfixed)
Estimateintegerambiguities
Success-rateSuccess-rate
Probability (number in interval [0-100%]) of fixing thecarrier ambiguities to their correct integer values
Requires satellite almanac and approximateuser position, but no actual data
Ionosphere modelsIonosphere models
Ionosphere float (long baseline)
New ionosphere parameter for each satellite at each observationepoch (equivalent to eliminating ionosphere)
Ionosphere weighted (medium baseline)
New ionosphere parameter for each satellite at each observationepoch constrained by a priori standard deviation
Ionosphere fixed (short baseline)
Ionosphere assumed absent
Design parameters - frequenciesDesign parameters - frequencies
L1 1575.420 MHz
L2 1227.600 MHz
L5 1176.450 MHz
GPS
E2-L1-E1 1575.420 MHz
E5b 1202.025 MHz
E6 1278.750 MHz
Galileo
Design parametersDesign parameters
Location: Presidente Prudente (22° S, 51° W)
Date: 11 July 2002
Minimum elevation: 10°
Standard deviation carrier: 0.003 m
Ionosphere weight: 0.05 m (medium baseline only)
Computation of instantaneous success-rates(single-epoch ambiguity resolution)
Design parameters - code observationsDesign parameters - code observations
Standard deviation code
L1 0.30 E2-L1-E1 0.15
L2 0.30 E5b 0.10
L5 0.10 E6 0.10
Interpretation of resultsInterpretation of results
0%3%
44%
57%
100% 100%
0%
20%
40%
60%
80%
100%
Dual-frequency Triple-frequency
Su
cces
s ra
te [
%]
Long baseline Medium baseline Short baseline
Succes-rate is 57% orhigher in 95% percent ofall cases
GPS onlyGPS only
0%3%
44%
57%
100% 100%
0%
20%
40%
60%
80%
100%
Dual-frequency Triple-frequency
Su
cces
s ra
te [
%]
Long baseline Medium baseline Short baseline
Integrated GPS and GalileoIntegrated GPS and Galileo
3%
45%
87%92%
100% 100%
0%
20%
40%
60%
80%
100%
Dual-frequency Triple-frequency
Su
cces
s ra
te [
%]
Long baseline Medium baseline Short baseline
Code observations - alternativesCode observations - alternatives
Standard precision
L1 0.30 E2-L1-E1 0.15
L5 0.10 E5b 0.10
High precision
L1 0.30 E2-L1-E1 0.10
L5 0.05 E5b 0.05
Integrated GPS and Galileo (2 Integrated GPS and Galileo (2 freqfreq.).)
27%
59%
91% 94%100% 100%
0%
20%
40%
60%
80%
100%
Standard precision High precision
Su
cces
s ra
te [
%]
Long baseline Medium baseline Short baseline
Integrated GPS and Galileo (2 Integrated GPS and Galileo (2 freqfreq.).)
0
10
20
30
40
50
60
70
80
90
100
0:00
:00.
0
1:
00:0
0.0
2:
00:0
0.0
3:
00:0
0.0
4:
00:0
0.0
5:
00:0
0.0
6:
00:0
0.0
7:
00:0
0.0
8:
00:0
0.0
9:
00:0
0.0
10:
00:0
0.0
11:
00:0
0.0
12:
00:0
0.0
13:
00:0
0.0
14:
00:0
0.0
15:
00:0
0.0
16:
00:0
0.0
17:
00:0
0.0
18:
00:0
0.0
19:
00:0
0.0
20:
00:0
0.0
21:
00:0
0.0
22:
00:0
0.0
23:
00:0
0.0
Nu
mb
er o
f ep
och
s
0
5
10
15
20
25
Nu
mb
er o
f sa
telli
tes
High precision Standard precision Satellites
Success-rate 99%, medium baseline
Integrated GPS and GalileoIntegrated GPS and Galileo
Integrated use of GPS and Galileo significantlyincreases ambiguity succes-rate
Single-epoch ambiguity resolution becomes feasible ifionospheric effects can be constrained
Very long-baseline ambiguity resolution requires morethan one observation epoch and benefits most fromtriple-frequency integrated system
Major GNSS error sourcesMajor GNSS error sources
Ionosphere
Multipath
IonosphereIonosphere
Ionosphere is dispersive: effect is frequency dependent
Effect depends on free electron density in atmosphere
Electron density is related to solar activity
Solar activity has period of 11 years and is characterisedby sunspot numbers
SunspotsSunspots
Solar cycle and sunspot numbersSolar cycle and sunspot numbers
0
50
100
150
200
250
30017
49
1761
1773
1785
1797
1809
1821
1833
1845
1857
1869
1881
1893
1905
1917
1929
1941
1953
1965
1977
1989
2001
Su
nsp
ot
nu
mb
er
Sunspot number predictionSunspot number prediction
Effects of ionosphereEffects of ionosphere
Poor GPS satellite tracking
Disturbance of GPS data links
Positioning biases
Incorrect ambiguity resolution(also results in positioning biases)
Ionosphere and positioningIonosphere and positioning
Two permanent GPS receivers, 4 km apart(short baseline) in The Netherlands (52° N)
Known baseline, continuously monitored
Measurement set-up
Ionosphere and positioningIonosphere and positioning
Compute daily percentage of height errorsgreater than 5 cm for period of two months
Correlate this percentage to daily variationsin Total Electron Content (TEC)
Analysis procedure
Height biases and TECHeight biases and TEC
Hei
ght b
ias
> 5
cm
[%]
Dai
ly v
aria
tion
TE
C [t
ecu]
Effects due Effects due multipathmultipath
Observations
Estimated positions (if same satellites are used)
Biases repeat every 23h56m
Periodic biases in
Multipath Multipath in observationsin observations
GPS SV06, L2-code
Aim
MultipathMultipath detection in positions detection in positions
Set-up
Background
GPS constellation has 23h56m repeatability satellite-reflector-antenna geometry re-occurs every day
Collect data over 2×24 hours and determine presence ofrepeatable effects with period 23h56m
Show multipath explicitly in stationary baselines
Height errorsHeight errors
2nd day shifted 4 minutes
Real-time Real-time kinematic kinematic (RTK) positioning(RTK) positioning
Centimeter accuracy in real-time, providedcarrier ambiguities can be resolved to theircorrect integer values
Ideal world
RTK positioning in the real worldRTK positioning in the real world
Correct height estimates
Incorrect height estimates due to wronginitialization of carrier ambiguities
Precise RTK GNSS positioningPrecise RTK GNSS positioning
Choose reference stations carefully toavoid multipath as much as possible
Integrate RTK GPS with other sensors
Validate estimated carrier ambiguities
Trends in RTK positioningTrends in RTK positioning
From single-reference stationto network-based RTK
Higher reliability
Longer baselines
Less reference stations
Advantages
Fugro StarfixFugro Starfix globalglobal DGPS DGPS networknetwork
Fugro StarfixFugro Starfix HP HP
Regional sub-networks of global Starfix network
Decimeter accuracy for distances up to 500 km
Starfix HP (High Performance)
Jet Propulsion Laboratory’s IGDGJet Propulsion Laboratory’s IGDG
Global network of reference stations
Reference station data collection via the Internet
Data dissemination to users via the Internet
Decimeter positioning accuracy
Internet-based Global Differential GPS (IGSG)
IGDG - demonstration resultsIGDG - demonstration results
See also http://gipsy.jpl.nasa.gov/igdg
Last 30 minutes
Last six hours
ConclusionsConclusions
Modernized GPS and future Galileo offerunprecedented accuracy and availability
Both systems are vulnerable to intentional andunintentional interference
Error sources (ionosphere, multipath) remainmajor concern
Network-based RTK will result in longer baselinelengths and less reference stations