GNSS Data Processing Theory Slides

gAGE
GNSS Data Processing

research group of Astronomy and Geomatics
Theory
Slides
Barcelona, Spain
http://www.gage.upc.edu
gAGE
@ J. Sanz Subirana & J.M. Juan Zornoza

Authorship statement
The authorship of this material and the Intellectual Property Rights are owned
by J. Sanz Subirana and J.M. Juan Zornoza.
These slides can be obtained either from the server http://www.gage.upc.edu,

or jaume.sanz@upc.edu. Any partial reproduction should be previously
authorized by the authors, clearly referring to the slides used.
This authorship statement must be kept intact and unchanged at all times.
Contents
Lecture 0: Introduction.
Lecture 1: GNSS measurements and their combinations.
Lecture 2: Satellite orbits and clocks computation and accuracy.
Lecture 3: Position estimation with pseudorange.
Lecture 4: Introduction to DGNSS.
Lecture 5: Precise positioning with carrier phase (PPP).
Lecture 6: Differential positioning with code pseudoranges.
Lecture 7: Carrier based differential positioning. Ambiguity resolution techniques.
List of Acronyms.
gAGE
Introduction
Contact: jaume.sanz@upc.edu
gAGE/UPC
Web site: http://www.gage.upc.edu
Master of Science in GNSS @ J. Sanz & J.M. Juan

1
gAGE Specific Objectives:

To learn about GNSS observables (code and phase), their
characteristics, properties, combinations and applications.

To learn how to calculate satellites orbits and clocks from
navigation message. To know the achievable precision.
To learn how to model pseudodistance for code and phase
measurements. This includes calculation of: 1) Coordinates at
emission epoch, 2) Ionospheric delay (Klobuchar model), 3)
Tropospheric delay, 4) relativistic correction, 5) clocks offsets and
satellite instrumental delays, 6) phase wind-up, etc.
To learn how to set and solve the navigation equation system
using least-squares or Kalman filter (algorithm level).
To know how to use phase differential positioning: Floating and
fixing ambiguities.
To learn Carrier Phase Ambiguity Fixing techniques.
gAGE/UPC
To get tools and skills to process and analyze GNSS

data. To implement algorithms for satellite navigation.
Master of Science in GNSS @ J.
J Sanz & J.M.
J M Juan
3
gAGE
An intuitive approach to
GNSS positioning
gAGE/UPC

4
A ship determines its location from a set on lighthouses that

gAGE
send an acoustic signal at a known time.
Knowing the emission time

t0 in the lighthouse and
the reception time t1 in
the ship, the traveling time
t1-t0, and the geometric
range U=v(t1-t0) may be
computed.
U=v(t1-t0)
With only one lighthouse there
is a whole circumference of
possible locations
gAGE/UPC
Master
Ma
Master off Science
ter Sciience in GNSS @ J. Sanz & J.M. Juan
5
gAGE
With two lighthouses there

are two possible solutions.
But, one of them is not on
gAGE/UPC
the sea!
Master
Maste
Mas
aster
te off Science
Sciience in GNSS @ J. Sanz & J.M. Juan
6

gAGE
With three lighthouses a

single solution is found U
gAGE/UPC
Master
Ma
Master off Science
ter Sciience in GNSS @ J. Sanz & J.M. Juan
7
Errors in the clocks (lighthouses and ship) synchronism affects the accuracy
gAGE
True range or
Geometric range
The ranges are measured by means the

gAGE/UPC
traveling time of the acoustic signal from

the lighthouses to the ship.
Thence, the synchronism errors between Pseudorange or
the lighthouses and ship clocks will degrade
apparent distance
the positioning
Master
Maste
Mas
aster accuracy.
te off Science
Sciience in GNSS @ J. Sanz & J.M. Juan
due to the error clocks8
SUMMARY: The positioning system is based on:

gAGE
To know the coordinates of the lighthouses
To know the ranges from the ship to the lighthouses
To solve a geometric problem.

eo
True range or
Geometric range
NOTE: the ranges are measured by means

gAGE/UPC
the traveling time of the acoustic signal

from the lighthouses to the ship.
Thence, the synchronism errors between Pseudorange or
the lighthouses and ship clocks will degrade
apparent distance
the positioning
Master
M S i accuracy.
t off Science in
i GNSS @ J. Sanz & J.M. Juan
due to the error clock.9
g
gAGE
How GNSS Works
Satellites broadcast
orbit and clock data
Satellite coordinates
and clock offset
Lighthouses
coordinates P
Receiver measures
traveling time from
satellite
sate
sa
ate
ellite to receiver
Pseudorange (P)
Lighthouses-ship
gAGE/UPC
ranges.
Thence, the receiver coordinates are found solving a

geometrical
Master of Science in GNSS problem: from sat. coordinates and ranges10
@ J. Sanz & J.M. Juan
gAGE
How GNSS Works
One of the solutions is not

gAGE/UPC
on the Earth surface.

11
gAGE
How GNSS Works
Receiver
location
gAGE/UPC

12
gAGE
How GNSS Works
Lesson 1:
GPS measurements
and its combinations
Measurements:
Ranges
Pseudoranges are
computed from the
traveling time sat-rec
Several error sources
affect these
measurements
Receiver
gAGE/UPC
location

13
gAGE
How GNSS Works
Satellite location
and clock offsets are Lesson 2:
computed from the
navigation message: GPS Orbits and
(orbit.f) clocks
Measurements:
Me
Meas
asur
u em
ur emen
ents
en ts::
ts
Raangges
Ranges
Pseudoranges
Pse
P seud
udor
ud oran
ange
g s are
ge are
computed
comp
co mput
mp uted
ut
tedd from
from
from thethhe
traveling
tr
rav
a el ing time
elin
in
ing time sat-rec
e sa
satt-reec
Seve
Se vera
ve
Severalrall error
ra erro
er rorr so
ro sour
urce
urces
ce
sources s
affect
af
ffect
fe
ect these
thes
th e e
es
measurements
meas
me asur
as urem
ur emen
em ents
en ts
Receiver
gAGE/UPC
location

14
gAGE
How GNSS Works
Satellite location
and clock offsets are
computed from the
navigation message:
(orbit.f)
Measurements:
Ranges
Pseudoranges are
Atmospheric propagation: IONO, TROPO computed from the
traveling time sat-rec Lesson 3:
Several error sources
affect GPS
thesemeasurements
modeling (code)
Relativistic effects
measurements
MODEL:
Atmospheric propag.,
relativistic effects,
Receiver clocks and instrum.
gAGE/UPC
delays are modeled

location and removed.
And the navigation
equations are built

15
gAGE
How GNSS Works
Satellite
S
Sa
atte
elllit
ite lo
ite llocation
loca
oc
caati
tion
tion
on
Satellite
S
Sate
Saate
te
ell
llit
llite c
ite co
coordinates
oorordi
dina
dina
dinatette
tes
es
and
annd clock
cloc
cl ock offsets
offfs
o fset
e s a are
arre
computed
co
c omp
mpu utte edd from
ffrro
omm the
tthhe
he
navigation
na
n avviiga
ig
gaa
ati
ttiio
on
n m message:
me es
sssa
agge:
ge
((orbit.f)
(o
orbbit
it.
t.f)
f)
xio x1 yio y1 zio 1

Measurements: z 1
U 1
U 1
Uio1
PP UU dt
11
22
i
11
io
io dt
11
GG
11
kk

x x2
io
Ranges io

'xii
yio y 2are zio z 2 'y
P dt22
Pi UUio2io2 dt GGkk2 Pseudoranges
1
2 io
ii
Uio
computed
c fromUio2 the Uio2
2
'zii
........
........ ttraveling
time sat-rec
..........
P nn
Pi UUioio dtdt GGkk Several nerror sources
cdtii
nn nn nn
xio x
affect yio these
yn zio z n
Un 1
Lesson 3: measurements
io Uion Uion
Solving the
MODEL:
navigation Equations
Atmospheric propag.,
Navigation relativistic effects,
equations clocks and instrum.
The geometric delays are modeled
gAGE/UPC
problem is linearized, and removed.

and Weighted Least And the navigation
Mean Squares or equations are built
Kalman filter are
used to compute the
solution.
16
gAGE
How GNSS Works
Lessons 4, 5, 6:
Code and Carrier phase
Differential positioning.
Floating/fixing
ambiguities
Receiver
gAGE/UPC
location

17
gAGE
References
[RD-1] J. Sanz Subirana, J.M. Juan Zornoza, M. Hernndez-Pajares, GNSS

Data processing. Volume 1: Fundamentals and Algorithms. ESA TM-
23/1. ESA Communications, 2013.
Data processing. Volume 2: Laboratory Exercises. ESA TM-23/2. ESA
Communications, 2013.
[RD-3] Pratap Misra, Per Enge. Global Positioning System. Signals,
Measurements, and Performance. Ganga-Jamuna Press, 2004.
[RD-4] B. Hofmann-Wellenhof et al. GPS, Theory and Practice. Springer-Verlag.
Wien, New York, 1994.
gAGE/UPC

18
gAGE
Thank you!
gAGE/UPC

gAGE
Lecture 1
GNSS measurements
and their combinations
gAGE/UPC

1
gAGE
The authorship of this material and the Intellectual Property Rights are owned by
J. Sanz Subirana and J.M. Juan Zornoza.

gAGE/UPC
22 Jan 2015

2
gAGE
research group of Astronomy and Geomatics Contents
1. Review of GNSS measurements.

2. Linear combinations of measurements.
3. Carrier cycle-slips detection.
4. Carrier smoothing of code pseudorange.
5. Code Multipath.
gAGE/UPC

3
gAGE
Contents

5. Code Multipath.
gAGE/UPC

4
gAGE
GPS SIGNAL STRUCTURE
Two carriers in L-band:
L1=154 fo=1575.42 MHz C/A-code for civilian users [XC(t)]
L2=120 fo=1227.60 MHz P-code only for military and authorized
where fo=10.23 MHz users [XP(t)]
Navigation message with satellite
ephemeris and clock corrections [D(t)]
C/A
P(Y) P(Y)
1227.6
27.6 M
MHz 1575.42
5.42 M
MHz
L2 L1
gAGE/UPC
S L(1k ) (t ) aP X P( k ) (t ) D ( k ) (t ) sin(Z1t IL1 ) aC X C( k ) (t ) D ( k ) (t ) cos(Z1t IL1 )

S L( 2k ) (t ) bP X P( k ) (t ) D ( k ) (t ) sin(Z2t IL2 )

5
gAGE
GPS Code Pseudorange Measurements
S L(1k ) (t ) aP X P( k ) (t ) D ( k ) (t ) sin(Z1t M L1 ) aC X C( k ) (t ) D ( k ) (t ) cos(Z1t M L1 )
S L( 2k ) (t ) bP X P( k ) (t ) D ( k ) (t ) sin(Z2t M L2 )
binary code XP(t)
C1,P
P 1, P2
'T
P(T ) c 'T c trec (T ) t sat (T 'T )

From hereafter we will call:

gAGE/UPC
C1 pseudorange computed from XC(t) binary code (on frequency 1)

P1 pseudorange computed from XP(t) binary code (on frequency 1)
6
gAGE GPS Carrier Phase Measurements
S L(1k ) (t ) aP X P( k ) (t ) D ( k ) (t ) sin(Z1t M L1 ) aC X C( k ) (t ) D ( k ) (t ) cos(Z1t M L1 )
S L( 2k ) (t ) bP X P( k ) (t ) D ( k ) (t ) sin(Z2t M L2 )
Carrier phase
Carrier beat phase:
IL (T ) IL rec (T ) ILsat (T 'T )

C1,,P
P 1, P2
c
'T N Unknown ambiguity L1, L2
O
From hereafter we will call:

L1 =OIL1 measur. computed from the carrier phase on frequency 1
L2 =OIL2 measur. computed from the carrier phase on frequency 2
gAGE/UPC
C1 pseudorange computed from XC(t) binary code (on frequency 1)

7
gAGE
Carrier and Code pseudorange measurements
P1 P1
P1= c 'T= c [trec(T)-tsat(T-'T)]
P1 | U clock offset
| 20.000 Km
P1 is basically the geometric range (U)
between satellite and receiver, plus the
relative clock offset.
The range varies in time due to the
satellite motion relative to the receiver.
gAGE/UPC
P1 is an absolute measurement (unambiguous)

8
gAGE
Phase and Code pseudorange measurements
c 'T O1 N1
L1 (T )
Relative measurement
(shifted by the unknown ambiguity ON )
Each time that the receiver loose the

phase lock, the unknown ambiguity
changes by an integer number of O
gAGE/UPC
L1 | U clock offset O1 N1
9
Code and Carrier Phase measurements

gAGE
Code (unambiguous but noisier)
Ambiguity
Carrier Phase (ambiguous but precise)

gAGE/UPC

10
gAGE GPS measurements: Code and Carrier Phase
my and Geomatics
'
Antispoofing (A/S):
The code P is encrypted to Y. Wavelength
W V noise Main
Only theGPS
code Csignal
at characteristics
((chip-length) (1% of O) [*]
frequency L1 is available.
research group of Astronomy
m
Code measurements
C1 300 m 3m
Unambiguous
P1 (Y1): encrypted 30 m 30 cm
but noisier
Phase measurements
L1 19.05 cm 2 mm Precise
L2 24.45 cm 2 mm but ambiguous
gAGE/UPC
[*] the codes can be smoothed with the phases in order to reduce noise
(i.e, C 1 smoothed with L 1 50 cm noise)

11
gAGE
RINEX FILES
gAGE/UPC

12
GNSS Format Descriptions
GNSS data files follow a well defined set of
standards formats: RINEX, ANTEX, SINEX
Understanding a format description is a tough
task.
These standards are explained in a very easy and
friendly way through a set of html files.
Described formats:
Observation RINEX
Navigation RINEX
RINEX CLOCKS
SP3 Version C
ANTEX
Open GNSS Formats
with Firefox internet browser
More details at: http://www.gage.es/gLAB
gAGE/UPC Tutorial associated to the GNSS Data Processing book 13

Research group of Astronomy & Geomatics J. Sanz Subirana, J.M. Juan Zornoza, M. Hernndez-Pajares
Technical University of Catalonia
gAGE RINEX measurement file

HEADER
MEASUREMENTS
gAGE/UPC

14
gAGE/UPC

15

Measurement time
(receive time tags)
Number of tracked satellites

gAGE/UPC
One satellite per row

Epoch flag 0: OK
16
Synthetic P2
(A/S=on)
gAGE/UPC
Master of Science in GNSS S/N indicatorsLoss of lock indicator

17
gAGE
Pseudorange modeling
P= c 'T= c [trec(T)-tsat(T-'T)]
Psat
sat
rec
rec U sat
rec c (dtrec dt sat
) GG
Geometric range Clock offsets
gAGE/UPC
G Trop sat
rec Ion sat
rec K rec K sat
H
Ionospheric delay noise
Tropospheric delay Instrumental delays 18
gAGE
gAGE/UPC

19
gAGE
gAGE/UPC

20
gAGE
Exercise:
a) Using Exercise 1:
the file 95oct18casa___r0.rnx, generate the txt file
95oct18casa.a (with data ordered in columns).
b) Plot code and phase measurements for satellite PRN28 and
discuss the results.
Resolution:
a) gLAB_linux -input:cfg meas.cfg -input:obs coco0090.97o
gAGE/UPC
b) See next plots:

21
gAGE An example of program to read the RINEX: gLAB

RINEX file gLAB txt file
sta Doy sec PRN L1 L2 C1/ P2
cambiar
gAGE/UPC
The RINEX file is convert to a columnar format to easily plot its

contents and to analyze the measurements (the public domain free
tool gnuplot is used in the book to make the plots). 22
Code measurements The geometry U is the
gAGE
dominant term in the
plot. The pattern in the
figures is due to the

P1 variation of U
P1
gAGE/UPC
sat
P1sta U sta
sat
c (dtsta dt sat ) Tropsta
Master of Science in GNSS
sat
Ion1sat
sta K K sat
H231
1sta 1
Code measurements Similar plot for code

gAGE
measurements at f2.
Notice that
P2 Ionosphere (Ion) and
Instrumental delays (K)
depend on frequency.
gAGE/UPC
P2 sat U sta
sat
sta
sat
Ion2 sat
sta K K sat
24H 2
2 sta 2
Code
Ionosphere delays and
code and Phase measurements
gAGE advances phase measurements
Code measurements: C1,P1,P2

arr group of Astronomy and Geomatics
sat
; U sta c (dtsta dt sat ) Tropsta Ion1sat H1
sta K1sta K1
C
1 sta
sat
P1sta sat sat sat
P2 sat
sta U sta
sat
sat
Ion2 sat
sta K 2 sta K 2
sat
H2
Frequency dependent
Phase measurements: L1,L2
L1sat
sta U sta
sat
sat
Ion1sat
sta b1sta b1
sat
O1 N1 O1w Q 1
L2 sat
sta U sta
sat
sat
Ion2 sat
sta b2 sta b2
sat
O2 N 2 O2 w Q 2
research
rese
phase Ambiguities
gAGE/UPC
N1, N2 are integers

Wind Up

25
Carrier Phase measurements

s The geometry U is the
gAGE

variation of U.
The curves are broken
when the receiver loss
the lock (cycle-slip).
When a cycle-slip happens, the

phase measurement L changes
by un unknown integer number
gAGE/UPC
of cycles (N)
L1sat
sta U sta c (dtsta dt sat ) Tropsta
Master of Science
sat in GNSS sat
Ion1sat
sta b1sta b1
sat
O1 N1 O1w Q 1
Carrier Phase measurements
s The geometry U is the
gAGE

variation of U.
The curves are broken
when the receiver loss
the lock (cycle-slip).
When a cycle-slip happens, the

phase measurement L changes
by un unknown integer number
of cycles (N)
gAGE/UPC
L2 sat
sta U sta c (dtsta dt sat ) Tropsta
Master of Science
sat in GNSS sat
Ion2 sat
sta b2 sta b2
sat
O2 N 2 O2 w Q 2
gAGE
Contents

5. Code Multipath.
gAGE/UPC

28
gAGE
Linear Combinations of measurements:
Geometry-free (or Ionospheric) combination.

Ionosphere-Free combination.
Wide-lane and Narrow-lane combinations.
gAGE/UPC

29
1. Geometry-free (or ionospheric)

ionosph
heriic)) combination
comb
binattion
gAGE
1 of Astronomy and Geomatics
PI= P2 P1=Iono+ctt
LI= L1 L2= Iono+ctt+Ambig
Ambiguity
Code measurements: C1,P1,P2
U sta c (dtsta dt sat ) Tropsta Ion1sat

sta K1sta K1 H1
2 group
sat sat sat sat sat

C1sta ; P1sta
P -P
research
P2 sat
sta U sta
sat
sat
Ion2 sat
sta K 2 sta K 2
sat
H2
s
res
Carrier measurements: L1,L2

gAGE/UPC
A L1-L2
L1sat
sta U sta
sat
sat
Ion1sat
sta b1sta b1
sat
O1 N1 O1w Q 1
L2 sat
sta U sta
sat
sat
Ion2 sat
sta b2 sta b2
sat
O2 N 2 O2 w Q 2
gA

30
Carrier Ambiguities
1. Geometry-free (or ionospheric) combination
gAGE
PGPS observables
I= P2 P1=Iono+ctt The pattern corresponds to the
ionospheric refraction (Ion),
LI= L1 L2= Iono+ctt+Ambig

because the other terms (K) are
constant.
Notice that code measurements
are noisier.
Pij= c 't= c [trec(TR)-tems(TS)]
Ambiguity
gAGE/UPC

31
gAGE Ionospheric effects

The ionospheric refraction depends on:

Geographic location
Time of day
Time with respect to solar cycle (11y)
gAGE/UPC

32
gAGE Ionospheric effects
The ionospheric delay (Ion) is

proportional to the electron density
integrated along the ray path (STEC)
&
40.3 r [ GPSreceiver ]
&
Ion
f2
STEC STEC
&
N (r , t )dr
e
r [ GPStransmitter ]
Ionosphere
Ambiguity
gAGE/UPC

33
2. Ionosphere-free
Ionospheric-Free
Combination
Combination(Pc,Lc)
gAGE
The ionospheric refraction depends on 40.3

the inverse of the squared frequency Ion 2

STEC
and can be removed up to 99.9% f
combining f1 and f2 signals:
f12 P1 f 22 P2 f12 L1 f 22 L2
Pc Lc
f12 f 22 f12 f 22
Note: Ksat cancels in Pc
sat
Pcsta U sta
sat
sat
Hc and Ksta included in dtsta
sat
Lcsta U sta
sat
sat
bc , sta bc sat ON Rc ON w Q c
The ionospheric refraction has been removed in Lc and Pc

gAGE/UPC
ON = 10.7 cm, OW =86.2cm

OW
The Rc ambiguities are NOT integers!! Rc N1 N N2
O2 1
34
gAGE
Comments:
Two-frequency receivers are needed to apply the

ionosphere-free combination.
If a one-frequency receiver is used, a ionospheric

model must be applied to remove the ionospheric
refraction. The GPS navigation message provides the
parameters of the Klobuchar model which accounts
for more than 50% (RMS) of the ionospheric delay.
gAGE/UPC

35
3.- Narrow-lane (PN) and Wide-lane

Ionospheric-Free Combination (LW)
Combination
gAGE
The wide-lane combination LW provides a signal with a large wave-

length (OW=86.2cm ~ 4*O1). This makes it very useful for detecting

cycle-slips through the Melbourne-Wbbena combination: LW PN
f1 P1 f 2 P2 f1 L1 f 2 L2
PN LW
f1 f 2 f1 f 2
The same sign
PN sat
sta U sta
sat
sat
Ionw sat K w K w sat H N
sta sta
LW sat
sta U sta
sat
sat
Ionw sat bw bw sat Ow N w Q w
sta sta
gAGE/UPC
The ambiguities NW are INTEGERS!

Master of Science in GNSS No@ wind-up
J. Sanz & J.M. Juan
36
gAGE
research group of Astronomy and Geomatics Exercises:
1) Consider the wide-lane combination of carrier phase measurements

f1 L1 f 2 L2
LW , where LW is given in length units (i.e. Li= Oi Ii ).
f1 f 2
c
Show that the corresponding wavelength is: OW
f1 f 2
Hint:
LW= OW IW ; IW = I1 I2
2) Assuming L1, L2 uncorrelated measurements with equal noise VL,

show that: 2
J 12 1 f1
VL V L ; J 12
J 12 1 f2
gAGE/UPC
Master of Science in GNSS Slides associated to the GNSS Data Processing book
J. Sanz Subirana, J.M. Juan Zornoza, M. Hernndez-Pajares 37
gAGE
Contents

5. Code Multipath.
gAGE/UPC

38
gAGE
Detecting cycle-slips
This cycle-slip
involves millions
of cycles it is
easy to detect!!
There is a cycle-
slip of only one
gAGE/UPC
cycle (~20cm)
How to detect it?
39
gAGE
Exercise:
a) Using the file 95oct18casa___r0.rnx, generate the txt file

95oct18casa.a (with data ordered in columns).
b) Insert a cycle-slip of one wavelength (19cm) in L1
measurement at t=5000 s (and no cycle-slip in L2).
c) Plot the measurements L1, L1-P1, LC-PC, Lw-PN and L1-L2
and discuss which combination/s should be used to detect the
cycle-slip.
Resolution:
a) gLAB_linux -input:cfg meas.cfg -input:obs 95oct18casa_r0.rnx
b) cat 95oct18casa.a | gawk {if ($4==18)
print $3,$5,$6,$7,$8} > s18.org
gAGE/UPC
cat s18.org | gawk {if ($1>=5000) $2=$2+0.19;

printf %s %f %f %f %f \n, $1,$2,$3,$4,$5} > s18.cl
c) See next plots:
40
The geometry U is the dominant term in the plot. The variation
gAGE
of U in 1 sec may be hundreds of meters, many times greater than
the cycle-slip 19 cm) the variation of U shadows the cycle-slip!
L1 (without the cycle-slip)

L1 (with the cycle-slip)
A jump of O=19 cm (one cycle in L1)

has been introduced in L1 at t=5000s
1 unit = 19 cm
(L1 cycles)
U
gAGE/UPC
L1Master
sta U sta
sat of Science
sat in GNSS sat
Ion1sat
sta b1sta b1
sat
O1 N1 O1w 41Q 1
The geometry and clock offsets have been removed.

gAGE
The trend is due to the Ionosphere. The P1 code noise shadows
the cycle-slip, and without the reference (in blue), the time where
the cycle-slip happens could not be identified.
L1-P1 (with the cycle-slip)

cm
19cm

(L1 cycles)
= 19
unit =
11 unit
L1sat sta ctt ambig H

2 Ion1sat
sat
sta P1sta
gAGE/UPC
P1sat
sta U sta
sat
sat
Ion1sat
sta K1sta K1
sat
H1
L1Master
sta U
sat of Science
sta c
sat in GNSS
( dt sta dt sat
) Trop sat
sta Ion sat
1sta b1sta b1
sat
O1 N1 O1w 42Q 1
gAGE
The trend is due to the Ionosphere. The P1 code noise shadows
the cycle-slip, and without the reference (in blue), the time where
the cycle-slip happens could not be identified.
L1-P1 (without the cycle-slip)

L1-P1 (with the cycle-slip)

cm
19cm

(L1 cycles)
= 19
unit =
11 unit
L1sat sta ctt ambig H

2 Ion1sat
sat
sta P1sta
gAGE/UPC
P1sat
sta U sta
sat
sat
Ion1sat
sta K1sta K1
sat
H1
L1Master
sta U sta
sat of Science
sat in GNSS sat
Ion1sat
sta b1sta b1
sat
O1 N1 O1w 43Q 1
gAGE
The geometry, clock offsets and iono have been removed.
There is a constant pattern plus noise. The P1 code noise also shadows
the cycle-slip, and without the reference (in blue), the time where the
cycle-slip happens could not be identified.

LC-PC (without the cycle-slip)
LC-PC (with the cycle-slip)
1 unit = 10.7 cm
(Lc cycles)
sat
Lcsta Pcsta
sat
ctt ambig H
gAGE/UPC
sat
Pcsta U sta
sat
sat
Hc
sat
Lcsta U sta
sat
sat
bC , sta bC sat ON Rc ON w Q c
44
gAGE
The geometry, clock offsets and iono have been removed.
There is a constant pattern plus noise. The P1 code noise also shadows
the cycle-slip, and without the reference (in blue), the time where the
cycle-slip happens could not be identified.

LI-PI (without the cycle-slip)
LI-PI (with the cycle-slip)
1 unit = 5.4 cm
sat
LI sta PI sta
sat
ctt ambig H
gAGE/UPC
sat
PI sta IonI K I sta K I sat H I
sat
LI sta IonI bI sta bIsat O1 N1 O2 N 2 (O1 O2 ) w Q I45
The geometry , clock offsets and iono have been removed.

gAGE
There is a constant pattern plus noise. The PN code noise is under one
cycle of Lw. Thence, the cycle-slip is clearly detected
Lw-PN (without the cycle-slip)

Lw-PN (with the cycle-slip)
86.2 cm
= 86.2cm
(Lw cycles)
unit =
1 unit
1
sta PN sta
LW sat sat
ctt ambig H
gAGE/UPC
sat
PN sta U sta
sat
sat
Ionw sat K w K w sat H N
sta sta
LWMaster
sta U sta
sat of Science
sat in GNSS sat
Ionw sat
sta
bw sta
bw
sat
Ow N w Q46w
gAGE
The trend is due to the Iono. The L1 code noise is few mm, and
the variation of the ionosphere in 1 second is lower than O1 =19 cm
Thence, the cycle-slip is detected.

LI (without the cycle-slip)
LI (with the cycle-slip)
1 unit = 5.4 cm
H mm
(O1 O2 )w
L1,satsta Lsat
2, sta Ion1,satsta Ion2,satsta ctt ambig H
gAGE/UPC
L1sat
sta U sta
sat
sat
Ion1sat
sta b1sta b1
sat
O1 N1 O1w Q 1
L2Master
sta U sta
sat of Science
sat in GNSS sat
Ion2 sat
sta b2 sta b2
sat
O2 N 2 O2 w 47Q 2
Summary
gAGE
L1
L1-P1
LI-PI LC-PC
gAGE/UPC

48
The cycle-slips are detected by Combination
Ionospheric-Free the Ionospheric combination
gAGE (LI=L1-L2) and the Melbourne Wbbena (W=Lw-PN)
LI Lw-PN
Two independent combinations, LI and Lw, allow to detect

two independent cycle-slips (in L1 and L2 phase measur.).
L1 L2
gAGE/UPC
Notice that, from L1, L2 is not possible to detect short cycle-slips

49
gAGE
Contents

3.1 Cycle-slip Detection Algorithms
5. Code Multipath.
gAGE/UPC

50
Cycle-slip detector based on carrier phase data:
gAGE
research group of Astronomy and Geomatics The Geometry-free combination
LI L1 ( s; k ) L2 ( s; k ) The detection is based on fitting a

second order polynomial over a
sliding window of NI samples.
The predicted value is compared with
the observed one to detect cycle-slip.
p ( s; k )
> LI ( s; k N I ), , LI ( s; k 1)@
LI ( s; k )
p ( s; k )
LI ( s; k ) p ( s; k ) ! threshold
gAGE/UPC
LI ( s; k )
gAGE
LI ( s; k )
Carrier
measurements
p ( s; k )
Fitted Polynomial
Under not disturbed ionospheric conditions,

the geometry-free combination performs as a
very precise and smooth test signal,
driven by the ionospheric refraction.
Although, for instance, the jump produced by
a simultaneous one-cycle slip in both signals
gAGE/UPC
is smaller in this combination than in the

original signals (O2 -O1 =5.4cm), it can provide
reliable detection even for small jumps
Cycle-slip detector based on code and carrier phase
gAGE
data: The Melbourne-Wbbena combination
LW PN OW NW H
BW
The detection is based on real-time

computation of mean (mBW) and sigma (SBW)
values of the measurement test data Bw.
A cycle-slip is
declared when
the measurement
differs form the
mean value by a
gAGE/UPC
predefined
number of
standard
deviations (SBW)
gAGE The Melbourne-Wbbena

combination has a double
benefit:
Threshold
The enlargement of the
Moving sigma
ambiguity spacing, thanks to
Moving average the larger wavelength
OW =80.4cm.
The noise is reduced by
the narrow-lane combination
of code measurement
Cycle-slip detection
Nevertheless, in spite of these

benefits, the performance is
worse than in the previous
gAGE/UPC
carrier-phase-only based
detector and it is used as a
secondary test.
Master of Science in GNSS @ J. San

Sanz
a z & J.M. Juan
n
gAGE Exercises:
1) Show that 'N1 9 and 'N 2 7

produces jumps of few millimetres in the geometry-free combination.
2) Show that no jump happens in the geometry-free combination when

'N1 / 'N 2 77 / 60 . In particular when 'N1=77 and 'N2=60 the
jump in the wide-lane combination is: 17 Ow 15 m
Hint: Consider the following relationships (from [RD-1]):

gAGE/UPC

55
Example of Single frequency Cycle-slip detector

gAGE
P1 L1
The detection is based on

real-time computation of
mean and sigma values of
the differences (d=L1-P1) of
the code pseudorange and
carrier over a sliding window
of N samples (e.g. N=100).
d ( s; k ) L1 ( s; k ) P1 ( s; k )
A cycle-slip is declared when
a measurement differs from
the mean bias value over a
predefined threshold.
gAGE/UPC
Missed detection
This detector is affected by the
gAGE code pseudorange noise and
multipath as well as the
divergence of the ionosphere.
Higher sampling rate improves

detection performance, but
shortest jumps can still escape
from this detector.
On the other hand, a minimum
number of samples is needed
for filter initialization in order to
ensure a reliable value of sigma
for the detection threshold
More details, exercises and examples

of software code implementation of
gAGE/UPC
these detectors can be found in

[RD-1] and [RD-2].
Master of Science in GNSS @ J. San

Sanz
nz & J.M. Juan
n
gAGE
Contents

5. Code Multipath.
gAGE/UPC

58
gAGE Carrier smoothing of code pseudorange
The noisy (but unambiguous) code pseudorange can be smoothed with
the precise (but ambiguous) carrier. A simple algorithm is given next:

Hatch filter:
n 1
1
P ( k )
n
P(k )
n

P ( k 1) L( k ) L(k 1)
where P (1) P (1) and
n k; k N
n N; k t N
This algorithm can be interpreted as a real-time alignment of the

carrier phase with the code measurement:
n 1
P (k )
1

gAGE/UPC
P(k ) P(k 1) L(k ) L(k 1) L(k ) P L (k )

n n
gAGE
L PL
This algorithm can be interpreted as a real-time alignment of the

carrier phase with the code measurement:
gAGE/UPC
n 1
P (k )
1
n
P(k )
n

P (k 1) L(k ) L(k 1) L(k ) P L (k )

60
gAGE
P
P
PL
L
gAGE/UPC
n 1
P (k )
1
n
P(k )
n

P (k 1) L(k ) L(k 1) L(k ) P L (k )

61
gAGE
P
P
PL
L
gAGE/UPC
n 1
P (k )
1
n
P(k )
n

P (k 1) L(k ) L(k 1) L(k ) P L (k )

62
gAGE Code-carrier divergence: SF smoother
Time varying ionosphere induces a bias in the single frequency (SF)
smoothed code when it is averaged in the smoothing filter (Hatch filter).

Let: Where U includes all non dispersive terms (geometric range,
P1 U I1 H1 clock offsets, troposphere) and I1 represents the frequency
dependent terms (ionosphere and DCBs). B1 Is the carrier
L1 U I1 B1 9 1 ambiguity, which is constant along continuous carrier phase
arcs and H1 , 9 1 account for code and carrier multipath and
thermal noise.
thence,
P1 L1 2 I1 B H1 2 I1 : Code-carrier divergence
Substituting P1 L1 in Hatch filter equation
P (k ) L(k ) P L (k )
U (k ) I1 (k ) B1 2 I1 B1 (k )

U (k ) I1 (k ) 2 I1 ( k ) I1 (k ) P1 U I1 biasI X1
gAGE/UPC

biasI
where X1 is the noise term
where, being the ambiguity term B1 a constant bias, after smoothing.
thence B1 k B1 , and cancels in the previous expression.
63
gAGE
N=3600 s
Iono
gAGE/UPC

gAGE
gAGE/UPC

65
gAGE
Halloween storm
Data File: amc23030.03o_1Hz
STEC
gAGE/UPC
N=100 (i.e. filter smoothing time constant W=100 sec ).

66
gAGE Carrier-smoothed pseudorange: DFree
Divergence-Free (Dfree) smoother:

With two frequency carrier measurements a combination of carriers
with the same ionospheric delay (the same sign) as the code can be
generated: 2
f2 1
D1 1.545
L1 2D1 L1 L2 U I1 B1, DF 9 1, DF f f 22 J 1
2
L1, DF 1
2
77
J
60
With this new combination we have:
This smoothed code is immune to temporal
P1 U I1 H1 gradients (unlike the SF smoother), being
L1, DF U I1 B1, DF 9 1, DF the same ionospheric delay as in the
original raw code (i.e. I1). Nevertheless, as
it is still affected by the ionosphere, its
Thence, spatial decorrelation must be taken into
gAGE/UPC
account in differential positioning.

P1 L1, DF B1, DF H1
No Code-carrier divergence! P1, DF U I1 X12
gAGE
gAGE/UPC

68
gAGE Carrier-smoothed pseudorange: IFree
Ionosphere-Free (Ifree) smoother:

Using both code and carrier dual-frequency measurements, it is possible
to remove the frequency dependent effects using the ionosphere-free
combination of code and carriers (PC and LC). Thence:
PC U H PC J P1 P2 J L1 L2 77
2
PIFree { PC ; LIFree { LC J
LC U BLC X LC J 1 J 1 60
Thence, J 2 1
V Pc V 3V P1
J 1 P 1
PC LC BC H PC PIFree { PC U X IFree
This smoothed is based on the ionosphere-free combination of

gAGE/UPC
measurements, and therefore it is unaffected by either the spatial and

temporal inospheric gradients, but has the disadvantage that the noise
in amplified by a factor 3 (using the legacy GPS signals).
gAGE
Vertical range: [-5 : 5] Vertical range: [-15:15]

gAGE/UPC

70
C1, L1 PC, LC
gAGE
N=100 N=100
Exercise:
STEC
Justify that the ionosphere-free
combination (PC) is (obviously) not
affected by the code-carrier
gAGE/UPC
divergence, but it is 3 times noisier.
C1, L1 PC, LC
gAGE
N=360 N=360
N=100 N=100
gAGE/UPC

C1, L1 PC, LC
gAGE
N=3600 N=3600
N=360 N=360
gAGE/UPC
gAGE
Halloween storm
Data File: amc23030.03o_1Hz
STEC
gAGE/UPC
N=100 (i.e. filter smoothing time constant W=100 sec ).

74
gAGE

5. Code Multipath.
gAGE/UPC

75
gAGE
Multipath
One or more reflected signals reach the antenna in addition
to the direct signal. Reflective objects can be earth surface
(ground and water), buildings, trees, hills, etc.

It affects both code and carrier phase measurements, and it
is more important at low elevation angles.
Butterfly shape
Code: up to 1.5 chip-length up to 450m for C1 [theoretically]

Typically: less than 2-3 m.
gAGE/UPC
Phase: up to O/4 up to 5 cm for L1 and L2 [theoretically]

Typically: less than 1 cm
76
Exercise
gAGE
Plot code and phase geometry-free combination for satellite PRN 15
of file 97jan09coco_r0.rnx and discuss the results.
gAGE/UPC
Butterfly shape:
High multipath for low elevation rays (when satellite rises and sets)
77
gAGE
M Pc Pc Lc
gAGE/UPC

gAGE
M Pc Pc Lc
gAGE/UPC
gAGE
M MW PN LW
gAGE/UPC

80
gAGE
M MW PN LW
gAGE/UPC

81
gAGE
After one year, the directions of the Sun and Aries coincide again, but
the number of laps relative to the Sun (solar days) is one less than
those relative to Aries (sidereal days).
24h
3m56s
gAGE/UPC
365.2422
Thus, a sidereal day is shorter than a solar day for about 3m 56s
82
Receiver and multipath noise
gAGE
gAGE/UPC
GPS standalone (C1 code) 10,000

83
Receiver and multipath noise

gAGE
Same environment!
gAGE/UPC
GPS standalone (C1 code) 100

84
gAGE
References

gAGE/UPC

85
gAGE
Thank you!
gAGE/UPC

gAGE
Lecture 2
Satellite orbits and clocks
computation and accuracy
gAGE/UPC

1
gAGE

gAGE/UPC
24 April 2014

2
gAGE
1. Elliptic orbit: Keplerian elements.
2. Perturbed Keplerian orbits: Osculating orbit.
3. GPS satellite coordinates computation and accuracy
3.1. From Broadcast Navigation Message.
3.2. From precise products.
4. GPS Satellite clock computation and accuracy
5. Geographic decorrelation of ephemeris errors.
gAGE/UPC

3
gAGE
Contents

gAGE/UPC

4
gAGE The GPS navigation
message provides
pseudo-Keplerian
elements to compute
satellite coordinates
gAGE/UPC

5
gAGE
(X, Y, Z, Vx, Vy, Vz) (a, e, i, :, Z, V)
6 values are needed (x,y,z,vx,vy,vz) to provide the

position and velocity of a body. They can be map into
the six Keplerian elements (a, e, i, :, Z, V ), which
provides the natural representation of the orbit!
gAGE/UPC

6
(a, e, i, :, Z, V)
gAGE V
perigee
x
a ae
focus
orbit orbit position in

shape orientation the orbit
gAGE/UPC

7
gAGE True anomaly V(t)
Fictitious body moving at

velocity n=2S/P=constant
M Mean anomaly M(t)
Perigee
T0 : time of passage by
satellites perigee
2S P
T0 M (t ) n(t T0 ) ; n
P a3
t n
2S V(t) E (t ) M (t ) e sin E (t )
P
gAGE/UPC
1 e E (t )
P , a, e V (t ) 2 arctan tan
1 e 2

8
gAGE
gAGE/UPC

9
Due to the non-spherical nature of gravitational potential, the attraction

gAGE
of the Sun and Moon, the solar radiation pressure, etc., the true
satellite path deviates from the elliptic orbit.
At any time an elliptical orbit

tangent to the true path can be
defined. This is the osculating
orbit, whose Keplerian elements
vary with time t:
a(t),e(t),i(t),:(t),Z(t),V(t)
True path
gAGE/UPC
Instantaneous
elliptic tangent
(osculating) orbit.

10
gAGE
Different magnitudes of perturbation and

their effects on GPS orbits

gAGE/UPC

11
gAGE
Central body
Central body + J2
Central body + J2 + SM
gAGE/UPC

12
gAGE
Central body
Central body + J2
Central body + J2 + SM
gAGE/UPC

13
gAGE
Central body
Central body + J2 Central body + J2 + SM

gAGE/UPC

14
gAGE Calculation of osculating orbital elements from position
gAGE/UPC and velocity (rv2osc.f)

15
gAGE Calculation of position and velocity from osculating orbital

elements (osc2rv.f)
Where:
gAGE/UPC

16
gAGE Exercise: Orbital elements variation:
File 1995-10-18.eci contains the precise position and
velocities of GPS satellites every 5 minutes for October

18th, 1995 in a Earth-Centred Inertial system (ECI)
[from JPL/NASA server:
ftp://sideshow.jpl.nasa.gov/pub/gipsy_products]
a) Use program rv2osc to compute the instantaneous
orbital elements
(X,for
Y, each epoch
Z, Vx, (a,
in the
Vy, Vz) file.e,That
i, :, is:
Z, V )
b) Plot the orbital elements in function of time to show
their variation: a(t),e(t),i(t),:(t),Z(t),V(t)
c) Compare with the broadcast orbital elements
Solution:
gAGE/UPC
a) cat 1995-10-18.eci|rv2osc> orb.dat

b) See the following plots
17
gAGE Semi-major axis Eccentricity

Inclination Ascending node

gAGE/UPC

18
gAGE
Argument of Perigee Mean Anomaly

gAGE/UPC

19
gAGE
Contents

gAGE/UPC

20
gAGE
research group of Astronomy and Geomatics GPS navigation message
25 pages
Subframes #4 and #5
have 25 pages
One Master Frame includes All 25 pages of

gAGE/UPC
Subframes #4 and #5 25 x 30s = 12.5 min

21
gAGE Subframe 1 contains information about the parameters to be applied

to satellite clock status for its correction. These values are
polynomial coefficients that allow time onboard to be converted to
GPS time. The subframe also contains information on satellite

health condition.
Subframes 2 and 3 contain satellite ephemerides.
Subframe 4 provides ionospheric model parameters (in order to

adjust for ionospheric refraction), UTC information, part of the
almanac, and indications whether the A/S is activated or not
(which transforms the P code into encrypted Y code).
Subframe 5 contains data from the almanac and on constellation

status. It allows rapid identification of the satellite from which the
gAGE/UPC
signal comes. A total of 25 frames are needed to complete the

almanac.

22
gAGE Ephemeris in navigation message
gAGE/UPC
In order to calculate WGS84 satellite coordinates, you should apply

de following algorithm [GPS/SPS-SS, table 2-15] (see in the book
FORTRAN subroutine orbit.f)
23
gAGE RINEX ephemeris file

gAGE/UPC

24
3.1. Computation of satellite coordinates from navigation
gAGE
message (orbit.f)
Computation of tk time since ephemerids reference epoch toe (t and toe
are given in GPS seconds of week):

tk t toe
Computation of mean anomaly Mk for tk,
P
Mk M0 'n tk
a3

Iterative resolution of Keplers equation in order to compute eccentric
anomaly Ek :
Mk Ek e sin Ek
Calculation of true anomaly vk :
1 e2 sin E
vk arctan k

cos Ek e

gAGE/UPC
Computation of latitude argument uk from perigee argument W, true

anomaly vk and corrections cuc and cus:
uk Z vk cuc cos 2 Z vk cus sin 2 Z vk

25
Computation of radial distance rk, taking into consideration

gAGE
corrections crc and crs:
a 1 2cos Ek crc cos 2 Z vk crs sin 2 Z vk

rk
Calculation of orbital plane inclination ik from inclination io at
reference epoch toe and corrections cic and cis :
ik i0 itk cic cos 2 Z vk cis sin 2 Z vk

Computation of ascending node longitude :k (Greenwich), from
longitude :0 at start of GPS week, corrected from apparent variation
of sidereal time at Greenwich between start of week and and
reference time tk=t-toe, and also corrected from change of ascending
node longitude since reference epoch toe.
:k :0 : ZE tk ZE toe
Calculation of coordinates in CTS system, applying three rotations
(around uk, ik, :k) : X r
gAGE/UPC
k k
Y R 3 (:k )R1 (ik )R 3 (uk ) 0
k
Z k 0
26
gAGE
t Orbit.f (x,y,z)[CTS]
z
Nav. message
(ephemeris)
Conventional Terrestrial System

(CTS):
y
Earth Centred, Earth-Fixed
(ECEF) System X
the reference system rotates Greenwich
with Earth.
gAGE/UPC

27
gAGE
gAGE/UPC

28
gAGE
SA=on/off
gAGE/UPC

29
gAGE Selective Availability (S/A): Intentional degradation of satellite

clocks and broadcast ephemeris. (from 25 March, 1990)
GPS Before and After S/A was switched off

Colorado Springs, Colorado 2 May 2000
160
140 Horizontal Error (meters)
120 Vertical Error (meters)
100
Instantaneous Error (meters)
80
60
40
20
0
-20
-40
-60
-80 ANALYSIS NOTES
-100
-120 - Data taken from Overlook PAN Monitor Station,
equipped with Trimble SVeeSix Receiver
-140 - Single Frequency Civil Receiver
gAGE/UPC
-160 - Four Satellite Position Solution at Surveyed Benchmark

-180 - Data presented is raw, no smoothing or editing
-200
0 1 2 3 4 5 6 7 8 9 10
Time of Day (Hours UTC)

30
gAGE
SA=off
gAGE/UPC

31
gAGE
SA=off
gAGE/UPC

32
gAGE
Zoom
gAGE/UPC

Ephemeris Updates
33
gAGE
Contents

gAGE/UPC

34
gAGE
3.2 Computation of satellite coordinates from
research group of Astronomy and Geomatics precise products.
Precise orbits for GPS satellites can be found on the International

GNSS Service (IGS) server http://igscb.jpl.nasa.gov
Orbits are given by (x,y,z) coordinates with a sampling rate of 15

minutes. The satellite coordinates between epochs can be computed
by polynomial interpolation. A 10th-order polynomial is enough for a
centimetre level of accuracy with 15 min data.
gAGE/UPC

35
gAGE
IGS orbit and clock products (for PPP):
Discrepancy between the different centres
gAGE/UPC

36
gAGE
gAGE/UPC

37
gAGE
GPS Satellite Clock computation: Broadcast message
dtsat=a0 + a1(t-t0) + a2(t-t0)2

t0 a0 a1 a2
YY MM DD H M S
PRN
2 NAVIGATION DATA GPS RINEX VERSION / TYPE
srx/v1.8.1.4 BAI 95/10/19 03:18:35 PGM / RUN BY / DATE
CASA COMMENT
-2444431.2031 -4428688.6270 3875750.1442 COMMENT
END OF HEADER
14 95 10 18 00 51 44.0 1.129414886236D-05 1.136868377216D-13 0.000000000000D+00
1.730000000000D+02-5.175000000000D+01 4.375182243902D-09-5.836427291652D-01
-2.712011337280D-06 2.427505562082D-03 8.568167686462D-06 5.153718931198D+03
2.623040000000D+05 4.470348358154D-08 1.698435481558D+00 1.676380634308D-08
gAGE/UPC
9.636381916043D-01 2.153437500000D+02 3.056960010495D+00-8.030691653399D-09

-5.178787145843D-11 1.000000000000D+00 8.230000000000D+02 0.000000000000D+00
3.200000000000D+01 0.000000000000D+00 1.396983861923D-09 1.730000000000D+02
2.592180000000D+05 0.000000000000D+00 0.000000000000D+00 0.000000000000D+00

38
gAGE
Computation of satellite clocks from precise products
Precise clocks for GPS satellites can be found on the International

They are providing precise orbits and clock files with a sampling rate
of 15 min, as well as precise clock files with a sample rate of 5 min
and 30 s in SP3 format.
Some centres also provide GPS satellite clocks with a 5 s sampling
rate, like the les obtained from the Crustal Dynamics Data Information
System (CDDIS) site.
Stable clocks with a sampling rate of 30 s or higher can be

interpolated with a first-order polynomial to a few centimetres of
accuracy. Clocks with a lower sampling rate should not be
gAGE/UPC
interpolated, because clocks evolve as random walk processes.

39
gAGE
Exercise:
Show that a common error
on all satellites does not
affect user positioning.
gAGE/UPC

40
gAGE
gAGE/UPC

41
gAGE
gAGE/UPC

42
gAGE
gAGE/UPC

43
gAGE
IGS Precise orbit and clock products:
RMS accuracy, latency and sampling
gAGE/UPC
http://igscb.jpl.nasa.gov/components/prods.html

44
gAGE
gAGE/UPC

45
gAGE
Ephemeris Errors and Geographic decorrelation
Position
from
Satellite
broadcast
location error
ephemeris
True position H
user
Uuser
Differential range error due
to satellite obit error
ref

U ref
Uuser user ref
GU
Uuser Uref
Uuser A conservative bound:
b
GU H
U
User
gAGE/UPC
with a baseline b 20km

Reference Station 20 1
GU H H
20000 1000

46
gAGE
Satellite Range
location H error from
user
error
CREU and Uuser
EBRE
Position
from
Satellite
location error
broadcast Differential range error from
ephemeris
True position H
user
Uuser
between CREU and EBRE
ref

U ref
U user user reff
GU
Uuser Uref
U user
gAGE/UPC
User
288 km of baseline
Reference Station

47
gAGE Differential range error from

Range between CREU and EBRE
error from user

CREU and Uuser
user reff
EBRE GU
Uuser Uref
CREU-EBRE
CREU
Differential positioning
Absolute positioning
gAGE/UPC
288 km of baseline

48
By Miguel Juan Zornoza
gAGE
Errors over the hyperboloid (i.e.
.U B U A ctt ) will not produce
u differential range errors.
The highest error is given by
the vector u , orthogonal to

the hyperboloid and over the
plain containing the baseline
vector b and the LoS vector .
Note:
Being the baseline b much smaller than
the distance to the satellite, we can
assume that the LoS vectors from A and
B receives are essentially identical to U.
a ( U B U A ) / 2 : hyperboloid semiaxis That is, U # U # U
B A
b / 2 : focal length
where a b / 2 cos(I ) u
u b u b - b
T T
Note: in this 3D problem I is NOT the elevation of ray. I b T b I b

T
Note: u sin I u
Differential range error GU produced by an

orbit error H parallel to vector u Note: being u a vector orthogonal
al
Let H {H to the LoS , thence, H T u
Thence:
gAGE/UPC
GU { G ( U B U A ) 2G a b sin I b
wa wa wI wI GU T u T sin I u
2 H 2
wH wI wH
H b sin I
wH
H U U
T I T
1 b Where: b = b b
b sin I H
U
Note: A H U GI U is the baseline
J. Sanz Subirana, JM. Juan Zornoza, M. Hernndez-Pajares, vector
gAGE
ORBIT TEST :
Broadcast orbits
Along-track Error (PRN17)
PRN17:
Doy=077, Transm. time: 64818 sec
17 10 3 18 20 0 0.0 1.379540190101E-04 2.842170943040E-12 0.000000000000E+00

7.800000000000E+01-5.059375000000E+01 4.506973447820E-09-2.983492318682E+00
-9.257976353169E-05 5.277505260892E-03 8.186325430870E-06 5.153578153610E+03
4.176000000000E+05-5.401670932770E-08-4.040348681654E-01-7.636845111847E-08
9.603630515702E-01 2.215312500000E+02-2.547856603060E+00-7.964974630307E-09
-3.771585673111E-10 1.000000000000E+00 1.575000000000E+03 0.000000000000E+00
2.000000000000E+00 0.000000000000E+00-1.024454832077E-08 7.800000000000E+01
4.104180000000E+05 4.000000000000E+00
gAGE/UPC
diff EPH.dat.org EPHcuc_x0.dat --------------------------------------------------

< -2.579763531685E-06 5.277505260892E-03 8.186325430870E-06 5.153578153610E+03
> -9.257976353169E-05 5.277505260892E-03 8.186325430870E-06 5.153578153610E+03
---------------------------------------------------------------------------------
Master of Science in GNSS J. Sanz Subirana, JM. Juan Zornoza, M. Hernndez-Pajares,

50
gAGE
Orbit error Range error

rov2
Differential range error

39.3 km Baseline: b=31.3km
rov1 31.3 km
11 km
15.2 km
gAGE/UPC
T I T
b
19.7 km
GU
rov3 U

51
gAGE
Exercise:
Justify that clock errors completely

cancel in differential positioning.
gAGE/UPC

52
gAGE
ERRORS on the Signal
Space Segment Errors:

Clock errors Common
Ephemeris errors
Strong spatial
Propagation Errors correlation
Ionospheric delay
Tropospheric delay Weak spatial
correlation
Local Errors
Multipath No spatial
gAGE/UPC
Receiver noise correlation

53
gAGE References

Measurements, and Performance. Ganga Jamuna Press, 2004.
gAGE/UPC

54
gAGE
Thank you
gAGE/UPC

55
gAGE

Lecture 3
Position estimation with
pseudoranges
gAGE/UPC

1
gAGE

gAGE/UPC
24 April 2014

2
gAGE
Contents
1. Measurements modelling and error sources

1.1. Introduction: Linear model and Prefit-residual
1.2. Code measurements modelling
1.3. Example of computation of modelled pseudorange
2. Linear observation model and parameter estimation
2.1. Least Squares solution (conceptual view)
2.2. Weighted Least Squares and Minimum Variance estimator
Example of solution computation
2.3. Kalman Filter (conceptual view)
Examples of static and kinematic positioning
gAGE/UPC

3
gAGE
Contents

gAGE/UPC

4
gAGE
Introduction: Linear model and Prefit-residual
Input:
- Pseudoranges (receiver-satellite j): P j
- Navigation message. In particular:
satellite position when transmitting signal: r j = (x j , y j , z j )
offsets of satellite clocks: dt j
(j = 1,2,n) (n>=4)
gAGE/UPC
Unknowns:
- receiver position : r = (x , y, z)
- receiver clock offset: DT

5
gAGE
For each satellite in view Iono+Tropo+TGD
c ( dtrec dt sat ) G k H
sat
C1rec U rec
sat
Linearising Uaround an a priori receiver position ( x0,rec , y0,rec , z0,rec )
x0,rec x sat y0,rec y sat z0,rec z sat

U sat
'xrec 'yrec 'zrec c dtrec dt sat G k
U U U
0, rec sat sat sat
0, rec 0, rec 0, rec
where:
'xrec =xrec x0,rec ; 'yrec =yrec y0, rec ; 'zrec zrec z0, rec
Prefit-residuals (Prefit)
C1 U
sat sat
c dt G k
sat
'xrec 'yrec 'zrec c dtrec
U0,satrec U0,satrec U0,satrec
rec 0, rec
gAGE/UPC
measurement computed unknown

6
For all satellites
gAGE x0,rec x sat1 y0,rec y sat1 z0,rec z sat1
in view
U0,satrec1 U sat 1
U sat1
1

Prefit
Prefit11
0, rec 0, rec
x ''xxrec
rec

0,rec
x sat 2 y0,rec y sat 2 z0,rec z sat 2 'y
Prefit 22
Prefit 1 'yrec
rec
U 0,satrec2 U0,satrec2 U0,satrec2
........
........ ''zzrec

.......... rec
Prefit
Prefit nn
x ccdt
dtrec
rec
x sat n y0,rec y sat n z0,rec z sat n
0,rec 1
U 0,satrecn U0,satrecn U0,satrecn
Observations Unknowns
(measured/computed)
Measurements modelling:
Prefit residual is the difference between measured and modeled
pseudorange: sat sat
Prefit recsat
C1rec [measured] C1rec [modelled]
gAGE/UPC
where:
,0 c dt
sat
C1rec [modelled] U rec
sat
sat 'rel sat Troprec
sat
Ion1rec
sat
TGD sat
7
gAGE
Contents

gAGE/UPC

8
gAGE Code Pseudorange modeling
C1
The pseudorange modeling is based in the GPS Standard

Positioning Service Signal Specification (GPS/SPS-SS).
gAGE/UPC
,0 c dt
sat
sat
sat
Ion1rec
sat
TGD sat
9
gAGE
gAGE/UPC
,0 c dt
sat
sat
sat
Ion1rec
sat
TGD sat

10
gAGE Geometric range
U rec
sat
Euclidean distance between satellite coordinates at emission time

and receiver coordinates at reception time.
x x0,rec y sat y0,rec z sat z0,rec

2 2 2
U0,satrec sat
Of course, receiver coordinates are not known (is the target of this
gAGE/UPC
problem). But ....
,0 c dt
sat
sat
sat
Ion1rec
sat
TGD sat
11
x sat x0,rec y sat y0,rec z sat z0,rec

2 2 2
U0,satrec
gAGE
Of course, receiver coordinates xrec , yrec , zrec are not known (they are
the target of this problem). But, we can always assume that an
approximate position x0,rec , y0,rec , z0,rec is known.
Thence, as it will be shown in next lesson, the navigation problem will

consist on:
1.- To start from an approximate value for receiver position
x0,rec , y0,rec , z0,rec (e.g. the Earths centre ) to linearise the equations.
2.- With the pseudorange measurements and the navigation
equations, compute the correction 'xrec , 'yrec , 'zrec to have
improved estimates: xrec , yrec , zrec x0,rec , y0,rec , z0,rec 'xrec , 'yrec , 'zrec
gAGE/UPC
3.- Linearise the equations again, about the new improved estimates,
and iterate until the change in the solution estimates is sufficiently
small. The estimates converges quickly. Generally in two to four
Master of Science in GNSS iterations, even if starting from the Earths @Centre.
J. Sanz & J.M. Juan
12
gAGE Satellite coordinates at emission time (rec2ems.f)

The GPS signal travels from
satellite coordinates at
emission time (tems) to
receiver coordinates at
reception time (trec).
The satellite can move

several hundreds of meters
from tems to trec.
The receiver time-tags are

g
given at reception time and
in the receiver clock time.
gAGE/UPC
An algorithm is needed to compute the satellite

coordinates at emission time in the GPS system time
from reception
Master of Science in GNSS time in the receiver time tags. @ J. Sanz & J.M. Juan
13
T[em s]
gAGE
The satellite offset

Geomatics
clock dt S can
C1 be
computed from the
time
navigation message
systemand
C1= c 't = c [trec(TR)-tems(TS)]

of Astronomy
As it is known, the pseudorange measurements link the emission

GPS
time (tems) in satellite clock (TS) with reception time (trec) in

group
receiver clock (TR) (receiver time tags).

in the
Thence, the emission time in the satellite clock is:

research
tems(TS) = trec(TR)-C1/c
Emission time
gAGE/UPC
Finally, since dt S = t S T is the time offset between satellite clock (tS)

and GPS system time (T), thence:
T[ems]= tems(TS)-dtS = trec(TR)-(C1/c+dtS)

14
gAGE Distance:'r
Distance 'r
gAGE/UPC

15
gAGE Variation in range:'U UemissionUreception

gAGE/UPC

16
gAGE Vertical error comparison
gAGE/UPC

17
gAGE Horizontal error comparison

gAGE/UPC

18
gAGE Satellite coordinates computation at emission time
The algorithm provided by the GPS/SPS-SS (orbit.f) supplies satellite

coordinates in an Earth-Fixed reference frame. To compute the
satellite coordinates
U See rec2ems.f
At the emission time, the following algorithm can be applied:
1. From receiver time-tags, compute emission time in GPS system
time: S
T[ems]= trec(TR)-(C1/c+d
dt )
2. Compute satellite coordinates at emission time T[ems]
T[ems] [orbit] (Xsat,Ysat,Zsat)CTS[emission]

3. Account for Earth rotation during traveling time from emission to
't (CTS reference system at reception time is used to
reception '
gAGE/UPC
build the navigation equations).
(Xsat,Ysat,Zsat)CTS[reception] =R3(ZE 't).(Xsat,Ysat,Zsat)CTS[emission]

19
gAGE Variation in range:'U UUemission

U
gAGE/UPC

20
gAGE/UPC

21

gAGE/UPC

22
gAGE Satellite and receiver clock offsets
They are time-offsets between satellite/receiver time and GPS
system time (provided by the ground control segment):
- The receiver clock offset (dtrec) is estimated together with receiver

coordinates.
- Satellite clock offset (dtsat) may be computed from navigation

message plus a Relativistic clock correction
dtsat=a0 + a1(t-t0) + a2(t-t0)2 + 'rel

gAGE/UPC
,0 c dt
sat
sat
sat
Ion1rec
sat
TGD sat
23
gAGE
a0 + a1(t-t0) + a2(t-t0)2
t0 a0 a1 a2
YY MM DD H M S
PRN

gAGE/UPC

24
gAGE Range variation: satellite clocks
gAGE/UPC

25

gAGE/UPC

26
gAGE/UPC

27
gAGE Relativistic clock correction ('rel)

A constant component depending only on nominal value of satellites

orbit major semi-axis, being corrected modifying satellites clock
oscillator frequency*: 2
f 0' f 0 1 v 'U 10
2 4.464 10
f0 2c c
A periodic component due to orbit eccentricity (to be corrected by user

receiver):
Pa rv
' rel 2 2 e sin( E ) 2 2 ( seconds )
c c
Being P=3.986005 1014 (m3/s2) universal gravity constant, c =299792458

(m/s) light speed in vacuum, a is orbits major semi-axis, e is its eccentricity,
E is satellites eccentric anomaly, and r and v are satellites geocentric position
and speed in an inertial system.
gAGE/UPC
*being f0 = 10.23 MHz, we have 'f=4.464 10-10 f0= 4.57 10-3 Hz

so satellite should use fo=10.22999999543 MHz.

28
gAGE Range variation: relativistic correction
gAGE/UPC

29

gAGE/UPC

30
gAGE/UPC

31
sat
gAGE Ionospheric Delay Ion f rec
The ionosphere extends from about 60 km in height until more than

2000 km, with a sharp electron density maximum at around 350 km.
The ionosphere delays code and advances carrier by the same amount
The ionospheric delay depends on sat 40.3
Ion1 rec I
signal frequency as given by: f12
Where I is number er of electrons per area unit

in the direction of observation, or STEC (Slant
Total Electron Content) sat
I N e ds
rec
For two-frequency receivers, it may be cancelled (99.9%) using

ionosphere-free combination f12 L1 f 22 L 2
LC
f12 f 22
For one-frequency receivers, it may be corrected (about 60%)
gAGE/UPC
using Klobuchar model (defined in GPS/SPS-SS), whose

parameters are sent in navigation message. (See program klob.f)
sat
CMaster
1rec [modelled]
of Science in GNSS
U0,satrec c dt sat 'rel sat Troprec
sat
Ion1rec
sat
TGD sat
32
gAGE Klobuchar model (klob.f)
Vertical delay
research group of Astronomy and Geomatics It was designed to minimize user IPP
computational complexity.
Slant delay
Minimum user computer storage
Minimum number of coefficients
transmitted on satellite-user link
At least 50% overall RMS ionospheric
error reduction worldwide.
Obliquity factor
It is assumed that the electron IonSLANT IonVERT m (elev)
content is concentrated in a thin R
1/ 2

2
layer at 350 Km in height. m (elev) 1 E

cos (elev)
R h
E
The slant delay is computed
from the vertical delay at the
gAGE/UPC
ionospheric Pierce Point (IPP),

multiplying by the obliquity factor.

33
gAGE
IONOSPHERIC PIERCE
POINTS (IPP)
IPPs trajectories Vertical Delay

for a receiver in Slant Delay
Barcelona, Spain
IPP
IonSLANT IonVERT m (elev)

1/ 2
R
2
m (elev) 1 E
cos (elev)
RE h
gAGE/UPC
Ionospheric Layer
(350 Km in height)

34
gAGE Klobuchar model
research group of Astronomy and Geomatics 30
Klobuchar
coefficients
25
)Time Delay (ns at 1.6 GHz
20 2S (t ) )
DC A cos ) (day
P
Amplitude IonVERT
15 ) DC ; if 2S (t ) ) ! S (night
P 2
10
Being :
3 3
5
*Period
A D
n 0
n
n
; P E
n 0
n
n
Dc=5ns

4 8 12 16 18 24
)Local Time (hours Where:
gAGE/UPC
IonSLANT )IonVERT m (elev DC= 5ns

1/ 2
R
2
))= 14 (ctt. phase offset
)m (elev 1 E
)cos (elev
RE h t = Local Time

35
gAGE
(time, rsta, rsat,DDDDEEEE) [Klob] Iono

elev, I

gAGE/UPC

36
gAGE Range variation: Ionospheric correction
Klobuchar model
STEC: I ON SLANT
TEC: I on VER T
gAGE/UPC
IonSLANT IonVERT m (elev)

1/ 2
R
2
m (elev) 1 E
cos (elev)
Master of Science in GNSS R h
E @ J. Sanz & J.M. Juan
37

gAGE/UPC

38
gAGE/UPC

39
gAGE
gAGE/UPC

Backup 40
gAGE
Tropospheric Delay
Troposphere is the atmospheric layer placed between Earths surface
research group of Astronomy and Geomatics and an altitude of about 60km.
The tropospheric delay does not depend on frequency and affects
both the code and carrier phases in the same way. It can be
modeled (about 90%) as:
- ddry corresponds to the vertical delay of the dry atmosphere
(basically oxygen and nitrogen in hydrostatical equilibrium)
It can be modeled as an ideal gas.
- dwet corresponds to the vertical delay of the wet component
(water vapor) difficult to model.
A simple model is:
sat
Troprec (d dry d wet ) m(elev) d dry 2.3exp(0.116 103 H ) meters
d wet 0.1m ; > H : heigh @ over the sea level
gAGE/UPC
1.001
m(elev)
0.002001 sin 2 (elev)
sat
C1Master
rec [modelled]
of Science in GNSS U0,satrec c dt sat 'rel sat Troprec
sat
Ion1rec
sat
TGD41
sat
gAGE
1.001
m(elev)
0.002001 sin 2 (elev)
gAGE/UPC

42
gAGE Range variation: Tropospheric correction
gAGE/UPC Trop SLANT
Trop VER TiCAL

RTCA-Do229C model

43

gAGE/UPC

44
gAGE/UPC

45
gAGE Instrumental Delays

Some sources for these delays are antennas, cables, as well as
several filters used in both satellites and receivers.

They are composed by a delay corresponding to satellite and
other to receiver, depending on frequency:
K1,sat
rec K1,rec TGD sat
f12
K sat
2, rec K 2,rec 2 TGD sat
f2
K1rec may be assumed as zero (including it in receiver clock offset).

TGDsat is transmitted in satellites navigation message (Total Group Delay).
According to ICD GPS-2000, control segment monitors satellite

gAGE/UPC
timing, so TGD cancels out when using free-ionosphere

combination. That is why we have that particular equation for K2.
sat
CMaster
1rec [modelled]
sat
Ion1rec
sat
TGD sat
gAGE Range variation: Instrumental delays (TGD)
gAGE/UPC

47

gAGE/UPC

48
gAGE/UPC

49
gAGE
Measurement noise (thermal noise)
y and Geomatics
Antispoofing (A/S):
The code P is encrypted to Y. Wavelength
W V noise Main
Only theGPS
code Csignal
at ((chip-length) (1% of O) [*] characteristics
frequency L1 is available.
research group of Astronomy
Code measurements
C1 300 m 3m
Unambiguous
but noisier
P2 (Y2): encrypted
P 30 m 30 cm
Phase measurements
L1 19.05 cm 2 mm Precise
L2 24.45 cm 2 mm but ambiguous
[*] codes may be smoothed with the phases in order to reduce noise
gAGE/UPC
(i.e., C1 smoothed with L1 50 cm noise)

gAGE
Multipath
One or more reflected signals reach the antenna in addition to the
research group of Astronomy and Geomatics direct signal. Reflective objects can be earth surface (ground and
water), buildings, trees, hills, etc.
It affects both code and carrier phase measurements, and it is more

important at low elevation angles.
Butterfly shape
Code: up to 1.5 chip-length up to 450m for C1 [theoretically]

gAGE/UPC
Typically: less than 2-3 m.

Phase: up to O/4 up to 5 cm for L1 and L2 [theoretically]
Typically: less than 1 cm
51
gAGE Receiver and multipath noise

gAGE/UPC

52
Same environment!
gAGE/UPC

53
gAGE
Contents

gAGE/UPC

54
gAGE
Example of Computation of modeled pseudorange
Using data of files gage2860.98o and brdc2860.98n,

compute by hand the modeled pseudorange for
satellite PRN 14 at t=38230 sec (10h37m10s).
sat
C1rec [modelled] U0,satrec c dt sat 'rel sat Troprec
sat
Ion1rec
sat
TGD sat
Follow these steps:
See also exercise 5, Session 5.2 in [RD-2]

gAGE/UPC

55
gAGE
1. Select orbital elements closer to 38230
2. Compute satellite clock offset

3. Compute satellite-receiver aprox. geometric range
3.1 Compute emission time from receiver (reception) time-tags
and code pseudorange.
3.2 Compute satellite coordinates at emission time
3.3 Compute approximate geometric range.
4. Compute satellite Instrumental delay (TGD):
5. Compute relativistic satllite clock correction
6. Compute tropospheric delay
7. Compute ionospheric delay
8. Compute modeled pseudorange from previous values:
gAGE/UPC
sat
sat
Ion1rec
sat
TGD sat
56
gAGE
1. Selection of orbital elements: From file brdc2860.98n,

research group of Astronomy and Geomatics select the last transmitted navigation message block before instant
t=38230 s (10h37m10s).
Transmission time:
979 208818 10h 0m 18s
PRN

GPS week

GPS sec of week

gAGE/UPC

57
gAGE
2. Satellite clock offset computation: From file
brdc2860.98n, compute satellite clock offset at time t=3830
s for PRN14:
PRN t0 a0 a1 a2

t = 38230 sec
t0= 12h 0m 0s= 43200 s
dt sat a0 a1 (t t0 ) a2 (t t0 ) 2 5.65 106 s

gAGE/UPC
sat
sat
Ion1rec
sat
TGD sat
58
gAGE
3. Satellite-receiver geometric range computation:
Use the following values (4789031, 176612, 4195008) as
research group of Astronomy and Geomatics approximate coordinates.
3.1: Emission time computation from receiver time-tag and code

pseudorange: T[ems]= t (T )-(C1/c + dtsat)
rec R
Measurement Pseudorange C1 at receiver time-tag

file gage2860.98o t=38230: C1= 23585247.70 m
Ephemeris file Satellite clock offset at t=38230 sec

brdc2860.98n dtsat= 5.65 10-6 sec (see previous
results)
Thence, the emission time in GPS satellite clock is:

gAGE/UPC
T[em s] = 38230 (23585247.70/c + 5.65 10-6) =

= 38229.9213224 (where c=299792458 )

59
gAGE Note:
From RINEX measurement file gage2860.98o, select the C1
pseudorange measurement at receiver time-tag for PRN14:
PRN 14
t = 38230 sec= 10h 37m 10s
4 L1 L2 C1 P2 # / TYPES OF OBSERV
98 10 13 10 37 10.0000000 0 5G18G14G16G 4G19

5007753.999 0.000 20143892.105 0.000
-220595.001 0.000 23585247.703 0.000
1305085.999 0.000 23146887.826 0.000
6246118.999 0.000 20798091.711 0.000
-19853878.999 0.000 22235319.057 0.000
Thence:
gAGE/UPC
Measurement Pseudorange C1 at receiver time-tag

file gage2860.98o t=38230: C1= 23585247.70 m

60
gAGE 3.2: Satellite coordinates at emission time pseudorange:
research group of Astronomy and Geomatics X= 11453479.346

T[emission]=
38229.921 sec
Orbit.f Y= 122468524.004
Z= 8245076.145

CTS [emission]
Use the selected
ephemeris for PRN14
(from file brdc2860.98n)
The previous coordinates are given in an Earth-fixed

reference frame (CTS) at t=T[emission]= 38229.921 s.
This reference frame rotates by un amount ZE 't during
traveling time 't=T[reception]-T[emission].
gAGE/UPC
(Xsat,Ysat,Zsat)CTS[reception] =R3(ZE 't).(Xsat,Ysat,Zsat)CTS[emission]

61
gAGE (Xsat,Ysat,Zsat)CTS[reception] =R3(ZE 't).(Xsat,Ysat,Zsat)CTS[emission]

11453350.377 cos(ZE 't ) sin(ZE 't ) 0 11453479.346

122468589.797 sin(Z E 't ) cos(Z E 't ) 0 122468524.004
8245076.145 0 0 1 8245076.145

CTS [ reception ] CTS [ emission ]
ZE 't 5.74 106 rad . ( where ZE 7.2921151467 105 rad / sec)

U0,satrec
't 0.079sec
c
x x0,rec y sat y0,rec z sat z0,rec 23616673.3m

2 2 2
U0,satrec sat
( x, y, z ) satellite | (11453479, 22468524, 8245076) An approximate value

( x0 , y0 , z0 ) receiver | (4789031, 176612, 4195008) is enough to compute
't.
gAGE/UPC
Note: Both satellite and receiver coordinates must be given in the

same reference system!
the CTS[reception]
Master of Science in GNSS will be used to build navigation equations.
662
2
gAGE 3.2: Geometric range computation
The geometric range between satellite coordinates at
emission time and the approximate position of the
receiver at reception time (both coordinates given in
the same reference system [for instance the CTS
system at reception time]) is computed by:
x x0,rec y sat y0,rec z sat z0,rec

2 2 2
U0, receiver
satellite sat
23616699.124m
( x, y, z ) satellite (11453350.2771, 22468589.7975, 8245076.1448)CTS [ reception ]

( x0 , y0 , z0 ) receiver (4789031, 176612, 4195008)CTS [ reception ]
gAGE/UPC
Approximate receiver coordinates

at reception time.
sat
CMaster
1rec [modelled]
sat
Ion
@
1rec TGD
sat sat
J. Sanz & J.M. Juan
63
gAGE 4. Satellite Instrumental delay (TGD): From file

brdc2860.98n, compute the Total Group Delay for PRN14:
PRN TGD (in sec)

TGD= -2.32830643654E-09 * c= -0.69801 m

gAGE/UPC
sat
sat
Ion1rec
sat
TGD sat
64
gAGE5. Relativistic clock correction:
PRN e )sqrt(a

= ]T[em ission E = 0.095 rad.

38229.921 s
Orbi t .f )(eccentric anom aly
Pa 10 P 3.986005 1014 m3 s 2

'rel sat
2 ) e sin( E 2.3 10 s
c2 c 299792458 m s 1
gAGE/UPC
sat
sat
Ion1rec
sat
TGD sat
65
gAGE
6. Tropospheric correction
d d wet m(elev) 6.76m

sat
Troprec dry
3
d dry 2.3e 0.11610 H
2.3m
See klob.f
d wet 0.1m
1.001 S
)m(elev ele
elev 20.57 0.359rad
)0.002001 sin 2 (elev 180
) H 160m (heigh over the ellipsoid
(x,y,z)rec [car2geo] (Lon, Lat, H)rec

gAGE/UPC
sat
sat
Ion1rec
sat
TGD sat
66
gAGE
7. Ionospheric correction

D
(time, rsta, rsat,DDDDEEEE) [Klob] Iono=10.26m
2 NAVIGATION DATA GPS RINEX VERSION/ TYPE

XPRINT v1.1 gAGE 00/06/04 17:36:23 PGM / RUN BY / DATE
gAGE BROADCAST EPHEMERIS FILE COMMENT
+1.9558E-08 +0.0000E+00 -1.1921E-07 +0.0000E+00 ION ALPHA
+0 +1.9661E+05
+1.2288E+05 -1.6384E+04 -2.6214E+05 ION BETA
20
-8.381903171539E-09-1.421085471520E-14 405504 979 DELTA_UTC: A0,A1,T,W
12 LEAP SECONDS
END OF HEADER
t 38230sec
( x, y, z ) satellite (11453350.2771, 22468589.7975, 8245076.1448)CTS [ reception ]
( x0 , y0 , z0 ) receiver (4789031, 176612, 4195008)CTS [ reception ]
Approximate values for
receiver or satellite
gAGE/UPC
coordinates are enough
sat
sat
Ion1rec
sat
TGD sat
67
gAGE
7. Compute the modeled pseudorange.
,0 c dt
sat
sat
sat
Ion1rec
sat
TGD sat
U0,satrec 23616699.124 m
c dt sat 5.65 10-6 c 1693.828 m
c 'rel sat 2.33 10-10 c 0.071m
sat
sat
Troprec 6.760 m C1rec [modelled] 23615021.689m
sat
Ion1rec 10.260 m
TGD sat 0.698 m
gAGE/UPC

68
gAGE
Prefit residual:
Is the difference between measured and modeled pseudorange
sat
Pref rec sat
C1rec C1[mod]rec
sat
U rec
sat
U0,satrec c dtrec K1rec H
In the previous example (PRN14 at t = 38230 s):

2
Pref= 23585247.703 - 23615021.689 = -29773.986 m
gAGE/UPC
Previously calculated
From measurement file
69
gAGE
Contents

gAGE/UPC

70
gAGE Solving navigation equations
Input:
- Pseudoranges (receiver-satellite j): P j
- Navigation message. In particular:
satellite position when transmitting signal: r j = (x j , y j , z j )
offsets of satellite clocks: dt j
(j = 1,2,n) (n>=4)
gAGE/UPC
Unknowns:
- receiver position : r = (x , y, z)
- receiver clock offset: DT

71
gAGE
For each satellite in view Iono+Tropo+TGD
c ( dtrec dt sat ) G k H
sat
C1rec U rec
sat
Linearising Uaround an a priori receiver position ( x0,rec , y0,rec , z0,rec )

U sat
'xrec 'yrec 'zrec c dtrec dt sat G k
U U U
0, rec sat sat sat
0, rec 0, rec 0, rec
where:
'xrec =xrec x0,rec ; 'yrec =yrec y0, rec ; 'zrec zrec z0, rec
C1 U
sat sat
c dt G k
sat
'xrec 'yrec 'zrec c dtrec
U0,satrec U0,satrec U0,satrec
rec 0, rec
gAGE/UPC

72
gAGE For all sat. x0,rec x sat1 y0,rec y sat1 z0,rec z sat1
in view 1
U0,satrec1 U sat 1
U sat1

Prefit
Prefit11
0, rec 0, rec
x ''xxrec
rec

0,rec
Prefit 22
Prefit 1 'yrec
rec
........
........ ''zzrec

.......... rec
Prefit
Prefit nn
x ccdt
dtrec
rec
0,rec 1
Observations Geometry of rays

(measured/computed)
G T sat n
0 rec

U0 sat
rec
n

0 sat
gAGE/UPC
rec
G sat n
T0 rec
T sat n
{ Unitary Line-Of-Sight
0 rec
U0 sat
rec
1
vector from receiver
to satellite
73
(x,y,z) coordinates
gAGE x0,rec x sat1 y0,rec y sat1 z0,rec z sat1
1
U0,satrec1 U sat 1
U sat1

Prefit
Prefit11
0, rec 0, rec
x ''xxrec
rec


0,rec
Prefit 22
Prefit 1 'yrec
rec
........
........ ''zzrec

.......... rec
Prefit
Prefit nn
x ccdt
dtrec
rec
0,rec 1

(measured/computed)
Prefit1 T0 rec
sat 1
1
T sat 2
2
T0 rec
sat n Prefit
........
0 rec

1

'rrec
c dt
.... ... rec
Prefit
n T sat n 1
(e,n,u) coordinates 0 rec
Prefit 1 cos el 1 sin az1 cos el1 cos az1 sin el1 1 'erec
gAGE/UPC
2
cos el sin az cos el cos az sin el 1 'nrec
2 2 2 2 2
Prefit
........ .......... 'urec

Prefit cos el sin az cos el cos az sin el 1 c dtrec
n n n n n n

74
gAGE From ECEF (x,y,z) to Local (e,n,u) coordinates

gAGE/UPC

Backup 75
x0,rec x sat1 y0,rec y sat1 z0,rec z sat1

1
gAGE COMMENTS: Prefit1

x
U0,satrec1 U sat1
0, rec U0,satrec1
'xrec
2
0,rec 1
Prefit U 0,rec
sat 2
U sat 2
U sat 2
rec
........ 0, rec 0, rec
'zrec

..........
Prefit
n
x c dtrec
0,rec 1
Of course, receiver coordinates xrec , yrec , zrec are not known (they are
the target of this problem). But, we can always assume that an
approximate position x0,rec , y0,rec , z0,rec is known.
Thence, as it will be shown in next lesson, the navigation problem will
consist on:
1.- To start from an approximate value for receiver position
x0,rec , y0,rec , z0,rec ( e.g. the Earths centre ) to linearise the equations
2.- With the pseudorange measurements and the navigation
equations, compute the correction 'xrec , 'yrec , 'zrec to have
gAGE/UPC
improved estimates: x , y , z x , y , z 'x , 'y , 'z

rec rec rec 0, rec 0, rec 0, rec rec rec rec
3.- Linearise the equations again, about the new improved estimates,
and iterate until the change in the solution estimates is sufficiently
small. 76
gAGE For all sat. x0,rec x sat1 y0,rec y sat1 z0,rec z sat1
in view 1
U0,satrec1 U sat 1
U sat1

Prefit
Prefit11
0, rec 0, rec
x ''xxrec
rec

0,rec
Prefit 22
Prefit 1 'yrec
rec
........
........ ''zzrec

.......... rec
Prefit
Prefit nn
x ccdt
dtrec
rec
0,rec 1

(measured/computed)
Thence, the basic linearized GPS measurement
yy
equation can be written as:
Gx
gAGE/UPC
This is a linear system with, in general, n t 4 equations which we can

solve using LS, WLS, Kalman filter,

77
gAGE
Contents

gAGE/UPC

78
gAGE Least Squares solution (conceptual review)
As a driving problem, let us consider the problem of fitting a set of
research group of Astronomy and Geomatics points (noisy measurements) to a straight line y=m x+n.
y
x y
x1 y1
x2 y2
yi
# #
xN yN
xi x
y1 m x1 n y1 x1 1
y mx n y x 1 m
2
2 2 y Gp
gAGE/UPC
2
# # n
#
N u1 N u 2 2 u1
yN m xN n yN xN 1
79
gAGE y1 m x1 n y1 x1 1
y mx n y x 1 m
2
2 2 y Gp
2
# # n
#
N u1 N u 2 2 u1
yN m xN n yN xN 1
This is an over-determined (incompatible) system of equations
(due to the measurement noise H).
It is evident that there is no straight line passing over all the data
points (red points). Thence, we have to look for a solution that
fits the measurements best in some sense.
y Note that, as G is not an squared
matrix (N>2), we cannot try:
y G p p G 1y
But, GTG is a squared (N x N)
gAGE/UPC
yi
matrix, thence, we can try:
xi x GT y G T G p p (G T G ) 1 G T y
80
gAGE
Results from Linear Algebra:

1) (G T G ) 1 The columns of matrix G are linearly independents.
N 2
2) p (G G ) G y min y y min yi yi Least Squares
T 1 T
where i1 solution
y G p
But, what is the physical meaning

of the least square solution?
yi What does it mean the condition
yi
N 2
gAGE/UPC
min y y min yi yi ?
xi x i1

81
What is the physical meaning of the least square solution?

gAGE
y What does it mean to minimize
N 2
min y y min yi yi ?
i1
yi
yi yi
yi
gAGE/UPC
xi x
82
The Least Squares solution gives the solution of equilibrium
gAGE
of the following physical system, in which the red dots (i.e. data
y points) are attached to the straight line by springs that can only
research group of Astronomy and Geomatics move in the direction of the y axis.
yi
yi
N 2
min yi yi
i1
gAGE/UPC
xi x
83
Indeed, the equilibrium solution is reached when the Total

gAGE
Potential Energy of the system is the minimum. That is,
y assuming the same spring constant k: V 1 k 'y 2 1 k y y
2
i i i i
2 2
1 N k
k yi yi
2
VTOTAL
2 i1 k
k
k
k
yi k k
k
yi
k
N 2
min VTOTAL min yi yi
i1
gAGE/UPC
xN i 2
x
min y y min yi yi p (G G ) G y
T 1 T
i1 @ J. Sanz & J.M. Juan
84
gAGE Let be the basic linearized GPS measurement equation:

y Gx
Least Squares solution:
2 2
min Y Y min yi y i
G G
1
x
t
Gt y i
Y AX
The same error is assumed in all measurements
Weighted Least Squares solution

wy1 0
If the measurements have different errors,
the equations can be weighted by matrix W : W %
0 wyn

And the weighted least squares solution is: Uncorrelated errors are assumed
gAGE/UPC
G W G
1 2
min wi yi yi
2
min y y
x
t
G Wy
t W
i
y G x
85
Weighted Least Squares solution:

gAGE
y The same, but assuming different spring constants wi:
1 1
wi yi yi
2
Vi wi 'yi2
2 2 wN
1 N
i i i

2
VTOTAL w y
y
2i1
yi w1 wj
wi
yi
w2
N 2
min VTOTAL min wi yi yi
i1
gAGE/UPC
N x 2 x
min y y W
min wi yi i yi p (G WG ) G Wy
T 1 T
i1 @ J. Sanz & J.M. Juan
86
gAGE
Contents

gAGE/UPC

87
gAGE Let be the basic linearized GPS measurement equation:

y Gx
Least Squares solution:

2 2
min Y Y min yi y i
G G
1
x
t
Gt y i
Y AX
The same error is assumed in all measurements
Weighted Least Squares solution

wy1 0
If the measurements have different errors,
the equations can be weighted by matrix W : W %
0 wyn

And the weighted least squares solution is: Uncorrelated errors are assumed
gAGE/UPC
G W G
1 2
min wi yi yi
2
min y y
x
t
G Wy
t W
i
y G x
88
Assuming that measurements Y have
gAGE random errors with zero mean and 1/ V y21 0
variance V2 , and assuming that error sources
research group of Astronomy and Geomatics for each satellite are uncorrelated with W %
error sources for any other satellite, the 0 1/ V y2n
following weighted matrix may be used:
wi
1
V y2 n wi p
V y2 i
i
greater error less weight V -V
V
V
VDt= 300Km= 1msec
V
Best Linear Unbiased Minimum Variance Estimator (BLUE):

Let be Py the error covariance matrix for measurements y.
1
If the weighting matrix is taken as W
W =PPy -1
y , thence the
Minimum Variance Solution is found:
G t Py1 G G t Py1 y
1
x
gAGE/UPC
G P G
And the error covariance matrix 1 1
for the estimation X is: Px t
y

89
gAGE
Contents

gAGE/UPC

90
gAGE

gAGE/UPC

See exercises 6 and 7, Session 5.2 in [RD-2] 91
gAGE
gAGE/UPC

92
gAGE
Contents

gAGE/UPC

93
gAGE Kalman filtering:

It is based on computing the weighted average between:
y ( n) (i.e., at t = t )
the measurement Y(n) n
( n) from previous estimation x

x
the prediction of the state X(n), ( n 1)
1. Weighted average:
y ( n) G ( n) x( n)
Lets assume, that we have the
prediction x (n) , with Pthence,
x
(n)
it can be used to add an

x ( n) x( n) additional set of equations
to the measurement equation
y= G x
gAGE/UPC
1
y ( n) G ( n) Py ( n ) 0
x( n) W
0

Px ( n )
x
( n )
Master of Science in GNSS I
94
1
y ( n) G ( n) Py ( n )
gAGE
0
x( n) W
0

Px ( n )
x
( n ) I
And the following solution of the previous equation system can be

found with some elemental algebraic manipulations:
G P G G 1 1
x
t
y
t
Py1 y
G P G 1 1
Px t
y
x
( n) Px ( n ) G t ( n) Py(1n ) y ( n) Px1( n ) x
( n)

gAGE/UPC
1
Px ( n ) G t ( n) P 1 G ( n) P 1
y(n) x ( n )
95
gAGE
2.- Prediction
Scalar case:
Lets xn 1 be the state at epoch n-1 with variance V x2 n 1
The simplest prediction model is to assume that the prediction

at epoch n is proportional to the state at epoch n-1. That is:
xn I xn 1
Thence, existing a linear relation between xn 1 and xn , the
variance of the prediction will be: 2
V x I 2 V x2 q 2
n 1
n
gAGE/UPC
An additional term is added to

account for modeling error!

96
gAGE
Generalization to the vector case:
xn I xn 1
V x2
n
I 2 V x2 n 1
q2
xn o x( n)
I o ( n) ( n) : transition matrix
V x2n
o Px ( n ) Q( n) : process noise matrix
q 2
o Q( n)
( n)
x ( n 1)
( n 1)
gAGE/UPC
Px ( n ) ( n 1) x ( n 1) t ( n 1) ( n 1)

97
gAGE Kalman filter (see kalman.f)

Measurements
( n)
x y ( n)
Px ( n ) Py ( n )
Estimation
Prediction
x (n) (n 1) x (n) Px ( n ) G t Py(1n ) y (n) Px1( n ) x (n)

1
Px ( n ) x ( n 1)t Px ( n ) G t Py(1n ) G P 1
x ( n )
x
(0)
Px (0) x
( n)
Px ( n )
gAGE/UPC
Initialization

98
gAGE Kalman filter (classical version)
Measurements
( n)
x y ( n)
Px ( n ) Py ( n )
Estimation
Prediction 1

x (n) x (n) K (n) y (n) G (n) x (n)
x (n) (n 1) 1
K ( n) Px ( n ) G t (n) G (n) Px ( n ) G t (n) Py ( n )
Px ( n ) x ( n 1)t
Px ( n ) >I K (n) G (n)@ Px ( n )
x
(0)
Px (0) x
( n)
Px ( n )
gAGE/UPC
Initialization

99
gAGE
Contents

gAGE/UPC

100
gAGE Some simple examples to define matrices )and Q
research group of Astronomy and Geomatics ( n)

x ( n 1)
( n 1)
Px ( n ) ) ( n 1) Px ( n 1) t ( n 1) ( n 1)
a) Static positioning:
State vector to be determined is X = (xrec , yrec , zrec , dtrec) where coordinates
(xrec , yrec , zrec) are considered constant (because receiver is fixed) and clock
offset dTrec is treated as white noise with zero mean and variance VDt . In
these conditions, matrices have the form:
1 0

1 0
( n) Q( n)
1 0

0 V DT
2

gAGE/UPC
Being V DT process noise associated to clock offset (in some way, the
2
uncertainty in clock value).
Master of Science in GNSS -V@ J. Sanz & J.M.VJuan

VDt= 300Km= 1msec 101
Constant
gAGE
The coordinates are always the same!

* * *
)=1 x(n)=x(n-1)
Q=(no prediction error)
gAGE/UPC
We can assure that, the next x(n) will be

the same as x(n-1).
( n)
x
x
( n 1)
Px ( n )
Px ( n 1) t
- 1) Q(n - 1) 102
gAGE
White Noise process N(0,V)
From a given epoch is not possible

research group of Astronomy and Geomatics to predict the following one
-2V
-V
+V
* * *
V
-V
V
)=0 x(n)=0 (zero mean)
Q=V2 (prediction error noise)
gAGE/UPC
We only can assume that, the next x(n)

can be x(n)=0 with a confidence V.
( n)
x
x
( n 1)
Px ( n )
Px ( n 1) t
- 1) Q(n - 1) 103
gAGE
gAGE/UPC

104
Static positioning: constant coordinates and white noise clock
gAGE

gAGE/UPC
S/A=on
105
gAGE
b) Kinematic positioning
1) In case of a fast moving vehicle, coordinates will be modeled as white

noise with zero mean, and the same rationale applies for clock offset:
0 V dx
2

0 V 2

( n) Q( n) dy
0 V dz2

0 V DT
2

2) In case of a slow moving vehicle, coordinates2 may be modeled as
random walk, process spectral density q dV , and the clock as a
dt
white noise:
1 qdx 't

1 qdy 't
( n) Q( n)
gAGE/UPC
1 qdz 't

0 V DT
2


106
gAGE Random Walk process: it varies slowly
The uncertainty increases with time

dV 2
't
dt
** *
( n)
x
x
( n 1)
Px ( n )
Px ( n 1) t
- 1) Q(n - 1)
gAGE/UPC
)=1 x(n)=x(n-1) (the same value is assumed)

Q=(dV
Master of Science
2
in GNSS/dt)*'t (but, with prediction error noise increasing
@ J. with time)
Sanz & J.M. Juan
107
gAGE
gAGE/UPC

108
Pure Kinematic positioning: white noise coordinates and clock
gAGE

gAGE/UPC
S/A=on
109
Kinematic positioning: random walk noise coordinates and white noise clock
gAGE
gAGE/UPC
S/A=on
110
gAGE

gAGE/UPC

111
Positioning: Random walk coordinates with Q=0 and white noise clock
gAGE
gAGE/UPC
S/A=on
112
gAGE
Solving with the kalman filter (by hand):

gAGE/UPC
See exercise 8, Session 5.2 in [RD-2]

113
gAGE
gAGE/UPC

114
gAGE

gAGE/UPC

115
gAGE
gAGE/UPC

116
gAGE

gAGE/UPC

117
gAGE
gAGE/UPC

118
gAGE References
gAGE/UPC

119
gAGE
Thank you
gAGE/UPC

120
gAGE
Lecture 4
Introduction to DGNSS
gAGE/UPC
gAGE

gAGE/UPC
24 April 2014

2
gAGE
1. Introduction: GNSS positioning and measurement

errors.
2. Differential positioning concept and differential
corrections.
3. Error mitigation in differential positioning.
4. DGNSS implementations: RTK, LADGNSS, WADGNSS.
5. DGNSS commercial services.
gAGE/UPC

3
gAGE
Contents

errors.
corrections.
gAGE/UPC

4
gAGE
GNSS Positioning
Standalone Positioning: GNSS receiver autonomous positioning
using broadcast orbits and clocks (SPS, PPS).

gAGE/UPC

5
gAGE
GNSS Positioning
Differential Positioning: GNSS augmented with data (differential
corrections or measurements) from a single

reference station or a reference station network.
Errors are similar for users separated tens,

gAGE/UPC
even hundred of kilometres, and these errors

are removed/mitigated in differential mode,
improving positioning.

6
gAGE ERRORS on the Signal

Clock errors Common
Ephemeris errors
Strong spatial
Propagation Errors correlation
Ionospheric delay
correlation
Local Errors
correlation
gAGE/UPC
Receiver noise

7
gAGE
Selective Availability (S/A)
was an intentional degradation of
public GPS signals implemented for
US national security reasons.
S/A was turned off at May 2nd 2000

(Day-Of-Year 123).
It was permanently removed in

2008, and not included in the next
generations of GPS satellites.
In the 1990s, the S/A motivated the

development of DGPS.
-These systems typically computed
PseudoRange Corrections (PRC) and
Range-Rate Corrections (RRC) every
5-10 seconds.
- With S/A=off the life of the
corrections was increased to more
gAGE/UPC
than one minute.

8
gAGE

errors.
corrections.
gAGE/UPC

9
gAGE
BELL
~100 Km baseline
S/A=on
EBRE
Error
gAGE/UPC
S/A=on
10
gAGE
research group of Astronomy and Geomatics BELL
<100 Km baseline
S/A=on
EBRE
BELL- EBRE
gAGE/UPC
Most of the errors cancel out when

computing the difference between BELL
and EBRE solutions.
(the same satellites are used in both solutions) S/A=on
11
gAGE
BELL
Error
b
The determination of the vector between the receivers S/A=on

APCs (i.e. the baseline b) is more accurate than the
single receiver solution, because common errors cancel
EBRE
BELL- EBRE
gAGE/UPC
Most of the errors cancel out when

computing the difference between BELL
and EBRE solutions.
(the same satellites are used in both solutions) S/A=on
12
gAGE Differential GNSS (DGNSS): Relative positioning
Error
b
User
Reference receiver
receiver
gAGE/UPC
The determination of the vector between the receivers APCs (i.e. the baseline b)
is more accurate than the single receiver solution, because common errors cancel.

13
gAGE Differential GNSS (DGNSS): absolute position

Computed Computed
position position
True
position
(known)
Error More
accurate
position
Reference
receiver User receiver
gAGE/UPC
If the coordinates of the reference receiver are known, thence the reference
receiver can estimate its positioning error, which can be transmitted to the user.
Then, the user can apply these corrections to improve the positioning
Note: Actually the corrections are computed in range domain (i.e. for each
satellite) instead of in the position domain. 14
In the previous example, the differential error has been cancelled in
gAGE
the position domain (i.e. solution domain approach).
But:
It requires to use the same satellites in both stations.

Thence, is much better to solve the problem in the range domain
than in the position domain. That is, to provide corrections for each
satellite in view (i.e. measurement domain approach)
Two implementations can be considered:
1.- The reference station, with known coordinates , computes range

corrections for each satellite in view. These corrections are broadcasted
to the user. The user applies these corrections to compute its absolute
position.
2.- The reference receiver (not necessarily at rest) broadcast its time-
tagged measurements to the user. The user applies these
gAGE/UPC
measurements to compute its relative position to the reference

station. Note: if the reference station coordinates are known, the user
can estimate its absolute position, as well.
15
gAGE
1.- Range Differential Correction Calculation
Broadcast SV Actual SV
Position Position
Calculated Measured Puser

Range Uref
Pref Pseudoranges
Differential Message Broadcast

PRC, RRC
Reference station
(known Location) User
The reference station with known coordinates, computes pseudorange

and range-rate corrections: PRC= Uref, Pref , RRC= 'PRC/'t .
The user receiver applies the PRC and RRC to correct its own
gAGE/UPC
measurements, Puser + (PRC + RRC (t-t0)), removing SIS errors and

improving the positioning accuracy.
DGNSS with code ranges: users within a hundred of kilometres can obtain
one-meter-level positioning accuracy using such pseudorange
corrections.
16
gAGE
GODN
USN3 76 m
GODS
Ref. station
ftp://cddis.gsfc.nasa.gov/highrate/2013/
gAGE/UPC
1130752.3120 -4831349.1180 3994098.9450 gods

1130760.8760 -4831298.6880 3994155.1860 godn
1112162.1400 -4842853.6280 3985496.0840 usn3

17
Differential Positioning Performance

gAGE
GPS Standalone GPS Standalone

S/A=off
DGPS DGPS
gAGE/UPC

18
Differential Corrections
gAGE
PRC
PRC= Uref, Pref GPS Standalone

S/A=off
RRC= 'PRC/'t DGPS

gAGE/UPC

19
In the previous example, the differential error has been cancelled in

gAGE
the position domain (i.e. solution domain approach).
But:
It requires to use the same satellites in both stations.

Thence, is much better to solve the problem in the range domain
than in the position domain. That is, to provide corrections for each
satellite in view (i.e. measurement domain approach)
Two implementations can be considered:
1.- The reference station, with known coordinates , computes range

corrections for each satellite in view. These corrections are broadcasted
to the user. The user applies these corrections to compute its absolute
position.
2.- The reference receiver (not necessarily at rest) broadcast its time-
tagged measurements to the user. The user applies these
gAGE/UPC
measurements to compute its relative position to the reference

station. Note: if the reference station coordinates are known, the user
can estimate its absolute position, as well.
20
gAGE 2.- Differential GNSS (DGNSS): Relative position
This concept of DGNSS can be applied even if the position of the reference
station is not known accurately or is moving, as well. In this case, the user
estimates its relative position vector with the reference receiver.
In this implementation of DGPS, the reference station broadcast its time-tagged

measurements rather than the computed differential corrections. The user receiver
form differences of its own measurements with those at the reference receiver,
(satellite by satellite) and estimate its position relative to the reference receiver.
gAGE/UPC
Real-Time Kinematics (RTK) is and example of this DGNSS.

Users within some ten of kilometres can obtain centimetre level positioning. The
baseline is limited by the differential ionospheric error that can reach up to
10cm, or more, in 10km, depending of the ionospheric activity.
21
gAGE
gAGE/UPC

22
COMMENTS
gAGE RRC= 'PRC/'t
Real-Time implementation entails delays in data
transmission, which can reach up to 1 or 2 s. RRC
Differential corrections vary slowly and its

useful life is of several minutes (S/A=off)
But, the measurements change much faster:
The range rate dU/dt can be up to 800m/s RRC ~ 1 cm/s
and, therefore, the range can change by
more than half a meter in 1 millisecond.
Moreover the receiver clock offset can be
up to 1 millisecond (depending on the dL1 | d U dTrec
receiver configuration).
Thence, the reference station dL1 / dt
measurements must be :
Synchronized to reduce station clock
mismatch: station clock can be estimated
to within 1Ps H dtsta 1mm
gAGE/UPC
d GW up to ~ 800 m/s
Extrapolated to reduce error due to dTrec / dt ~ 670 m/s
latency: carrier can be extrapolated with
error < 1cm.
23
gAGE
PRC L1
RRC dL1 / dt
gAGE/UPC
RRC ~ 2cm/s
dTrec / dt ~ 670 m/s
Receiver: JAVAD TRE_G3TH DELTA3.3.12

24
gAGE
L1 L1 ZOOM
1 ms jump
of receiver
clock adjust
dTrec / dt | 300 km/450 s 667 m/s
dL1 / dt ZOOM
dL1 / dt
300 km=1 ms
dL1 | d U dTrec
dL1 | d U dTrec
gAGE/UPC
dTrec / dt ~ 670 m/s

25
gAGE
Contents

errors.
corrections.
gAGE/UPC

26
gAGE Error mitigation: DGNSS residual error
Errors are similar for users separated tens, even hundred of kilometres, and
these errors vary slowly with time. That is, the errors are correlated on
space and time.
The spatial decorrelation depends on the error component (e.g. clocks are
common, ionosphere ~100km...). Thence, a reference stations network is
needed to cover a wide-area.
Error
gAGE/UPC
Short-baselines Long-baselines

27
gAGE Space Segment Errors

Satellite clock error:
Clock error
Clock modelling error is small (~2m RMS) and
changes slowly over hours.
Does not depend on user location, thence, it can
be eliminated in differential mode.
Satellite ephemeris:
Only the Line-Of-Sight (LOS) of error affects
positioning. This error is small (~2m RMS) and
Orbit error
changes slowly over minutes.
The residual error, after applying the differential
corrections depends upon the separation
between the LOS from user and reference
station.
A conservative bound is given by:
gAGE/UPC
b with a baseline b 20km

GU GH
U GU
20km
20000km
GH
1
1000
GH

28
gAGE Ephemeris Errors and Geographic decorrelation
Position from
broadcast
Satellite
ephemeris
location error
&
True position
H &
user
Uuser
due to satellite obit error
& &
&
& ref & user & ref
GU
U ref Uuser U ref
Uuser
b
T , T
&
U
Uuser
A conservative bound:
b
GU H
User U
gAGE/UPC

Reference Station
20 1
GU H H
20000 1000

29
gAGE
Range error
Satellite H from &
& user
location error CREU and

EBRE Uuser
Position from
broadcast
Satellite ephemeris
location error
&
True position
H &
user
Uuser
&
& ref

U ref
U user
MATA
Barcelona U user
gAGE/UPC
User
Reference Station

30
gAGE
Range error
Satellite H from &
& user
location error CREU and

EBRE Uuser
Differential range error from

288 km of baseline
& &
& user & ref
GU
MATA Uuser U ref
Barcelona
gAGE/UPC
b
T , T
&
GU
U

31

Range error
from &
CREU and & user

EBRE Uuser & &
& user & ref
GU
Uuser U ref
CREU-EBRE
CREU Differential positioning
gAGE/UPC
288 km of baseline

32
gAGE
Satellite clock anomaly

gAGE/UPC
Reference stations can detect and remove clock failures and other anomalies (e.g. GBAS)

33
gAGE
Atmosphere Propagation Errors
Ionospheric propagation delay:
Ionospheric delay depends on the STEC

(integrated electron density along ray path).
Reference and user receiver locations (i.e.
Baselines) are mapped to Ionospheric Pierce
Points (IPPs) associated to each satellite.
IPP
Typical spatial gradients of ionosphere are 1-2

mm/km (1V) 0.1-0.2 m in 100km. This value
can reach up to 300 mm/km (6-7 April 2000)
gAGE/UPC

34
gAGE
76 km
140 km
97 km
317 km
gAGE/UPC

35
gAGE
138 km
93 km
gAGE/UPC

36
gAGE Atmosphere Propagation Errors
Tropospheric propagation delay:
Tropospheric delay depends upon the air density profile along the signal
path.
Most of the tropospheric delay (~90%) comes from the predictable
hydrostatic component.
Wet component delay can vary considerably, both spatially and
temporally.
With 10km separation between
receivers, the residual range error Nominal
can be 0.1-0.2m
For long distance or significant
altitude difference it is preferable to
correct for the tropospheric delay at
both reference and user receivers. Actual
For a low elevation satellite, the
gAGE/UPC
Variation of Zenith
residual range error can be 2-7 mm
Total Tropospheric Delay (ZTD)
per meter of altitude difference.
gAGE
Nominal
USN3
GODZ
Actual
GODE
Actual
Zoom
Nominal
gAGE/UPC
Actual Zenith Tropospheric Delay

geographical decorrelation

38
gAGE Troposphere slant factor
Nominal
1.001
m(elev)
0.002001 sin 2 (elev)
Actual
Actual
Zoom
Nominal
gAGE/UPC
Actual Zenith Tropospheric Delay

geographical decorrelation

39
gAGE
72 km
98 km 93 km
Zenith Wet Tropospheric Geographical decorrelation

gAGE/UPC
Vertical delay Vertical delay

40
gAGE Local Errors
Receiver noise and multipath:
These errors are uncorrelated at the

reference and user receivers an cannot be
corrected by DGPS.
In fact any error incurred in the reference
station affects the user. Thence, it is
important to minimize errors at the
reference station.
Code noise can be reduced by smoothing
with carrier (at the level of 0.25-0.50m ).
But single frequency smoothing is affected
by code-carrier ionosphere divergence.
High accuracy applications use carrier Carrier
measurements, about two orders of ambiguity
gAGE/UPC
magnitude more precise than code

measurements, but the unknown
ambiguities must be fixed.

41
gAGE
Ionospheric delay (STEC)

Halloween storm
Halloween storm
gAGE/UPC

42
gAGE
GNSS Positioning: Local errors
Receiver and multipath noise Receiver and multipath noise
Same environment!
Measur. Noise & Multipath

GPS standalone (C1 code) 12,000 $ GPS standalone (C1 code) 300 $
Multipath
Error in carrier measurement due to multipath

gAGE/UPC
(cm level) or thermal noise (mm level) is

typically 2 orders of magnitude lower than in
code, but carrier has an unknown ambiguity.
43

gAGE/UPC

44
Same environment!
gAGE/UPC

45
gAGE ERRORS on the Signal


Clock errors Common
Ephemeris errors
Strong spatial
Propagation Errors
correlation
Ionospheric delay
correlation
Local Errors
Receiver noise correlation
gAGE/UPC

46
Source Potential Error size Error mitigation &
Residual error
and Geomatics and
gAGE
This table is from the book: Pratap Misra, Per Enge.
Satellite clock Clock modelling error: 2 m (RMS) DGPS: 0.0m
Measurements,
Performance. Ganga Jamuna Press, 2004.
Ephemeris Line-Of-Sight error: 2 m (RMS) DGPS: 0.1m (RMS)
prediction
Ionospheric Vertical delay: ~ 2-10 m Single-freq. using
Delay (depending upon user location, time of day & Klobuchar: 1-5m.
RTCM format
solar activity)
Astronomy
DGPS: 0.2m (RMS)

Obliquity factor: 1 at zenith, 1.8 at 30 , 3 at 5.
Signals,
Tropospheric Vertical delay ~ 2.3-2.5 m at sea level. Model based on

Delay (lower at a higher altitudes) average meteorolog.
group of
Conditions: 0.1 -1 m
System.
Obliquity factor: 1 at zenith, 2 at 30 , 4 at 15

and 10 at 5. DGPS: 0.2m (RMS)
research
plus altitude effect.

Multipath In clean environment: Uncorrelated
Global Positioning
Code : 0.5 1 m between antennas.

Carrier: 0.5 -1 cm Mitigation trough antenna
gAGE/UPC
design and sitting and

carrier smoothing of code.
Receiver Code : 0.25 0.50m (RMS) Uncorrelated

noise Carrier: 1-2 mm (RMS) between receivers
Master of Science in GNSS DGPS is based assuming baselines of Juan
@ J. Sanz & J.M. tens of
km and signal latency of tens of seconds. 47
gAGE
Contents

errors.
corrections.
gAGE/UPC

48
gAGE Differential GNSS (DGNSS) IMPLEMENTATIONS

The differential corrections can be broadcast as an:
Scalar correction (Local Area), where all corrections are lumped

together.
Examples: GBAS, LAAS, RASAN, RTK, VRS.
Vector correction (Wide Area), where the corrections are given for
each error source separately (state-space approach)
Examples: SBAS (WAAS, EGNOS...), DGPS (JPL), Fast-PPP.
gAGE/UPC

49
gAGE
Real Time Kinematics (RTK)
In this implementation of DGPS, the
reference station broadcast its time-tagged
measurements rather than the computed

differential corrections. The user receiver
form differences of its own measurements
with those at the reference receiver, satellite
by satellite and estimate its position relative
to the reference receiver.
Real-Time-Kinematics (RTK) is a technique for relative precise positioning

based on carrier phase data. The key feature of RTK is the ability to fix the
carrier ambiguities On-The-Flight (OTF), i.e. while on the move. Major
receivers manufacturers offer RTK solution packages consisting on a pair or
receivers, a radio link, and software.
The performance of RTK is measured by (i) initialization time, and (ii)
reliability (or, correctness) of the ambiguity fixing. There is an obvious trade-
gAGE/UPC
off between getting the answer quickly and getting it right.

For typical baselines of several kilometres, integer ambiguity resolution in
thirty or sixty seconds is common, achieving centimetre error level of accuracy
50
gAGE
(This picture is from http://water.usgs.gov/osw/gps/index.html)
Note: In Carrier phase Differential positioning, with a single reference station, the
accuracy decreases as a function of the distance from the reference station by a
rate of about 0.5-1cm per km.
Mode Horizontal accuracy

gAGE/UPC
Typical baselines up to
static 5 mm + 0.5 ppm 10-15km, depending on
Kinematic 5 cm + 5 ppm the ionosphere conditions

51
gAGE Message Type Title

1 Differential GPS corrections
2 Delta Differential corrections

3 GPS reference station parameters
9 GPS partial satellite set
RTCM format
10 P-code differential corrections

11 GPS C/A code L1, L2 delta corrections
15 Ionospheric delay message
17 GPS ephemeris
18 RTK uncorrected carrier phases
19 RTK uncorrected code pseudoranges
20 RTK carrier phase corrections
21 RTK code pseudorange corrections
gAGE/UPC
59 Proprietary message
The NTRIP protocol has been defined For transmission RTCM data over internet

52
gAGE
gAGE/UPC

53
gAGE Local Area DGNSS (LADGNSS): VRS

LADGNSS includes a Master station and several monitor stations. The master
station collects the range measurements of the monitor stations and process
the data to generate the range corrections, which are broadcasted to users
An example for High Accuracy Positioning, is the Virtual Reference
Station (VRS) technique, which is based on generating measurements
of a virtual (non existing) reference station, close to the user, from real
measurements of a multiple reference station network. These virtual
measurements are transmitted to the user to compute the RTK solution.
The NRTK yields accuracies at the level of 5cm for baselines up to 40km.
gAGE/UPC
Image courtesy of Trimble

54
gAGE
Virtual Reference Station
The basic scenario for VRS surveying is as
follows:
The user goes to a work site. The work

site preferably is, as with any other style
of GPS surveying, physically surrounded
by the reference station network.
The user sets up a surveying job and
logs into the Real-Time Network (RTN)
system using a cell phone (or other
communication method). As part of the
login process, the rover sends its position
(via a NMEA sting) to the RTN system.
The RTN system computes a virtual reference station, in close
proximity to the rover based on the position sent. Using input from
the closest surrounding reference stations, the RTN system then
computes and sends corrections as if a real base station were
gAGE/UPC
broadcasting from the location of the virtual reference station.

Using the cell phone, the receiver then obtains and applies the
corrections in real time.
After initialization, the survey proceeds in exactly the same manner as
an RTK survey. http://water.usgs.gov/osw/gps/real-time_network.html 55
Benefits of VRS surveying over traditional RTK surveying include:

gAGE
No need for a user to establish a permanent/semi-permanent base station.
This eliminates the time for initial site selection and (daily) set-up, any issues
of security and power supply, and the possibility of set-up errors.

The RTN can monitor its own integrity and can detect if there is a problem
with a particular reference station. With a single-base RTK setup it can be
difficult to tell if a problem exists or occurs with a base station while
conducting a survey.
Since the reference stations are part of a network, a loss of one station does
not result in failure of the entire network or the resulting survey. Whereas the
loss of a reference station with single-base RTK setup results in the end of
data collection, with RTN surveying the system accuracy degrades gradually.
A sufficiently dense reference station network can result in shorter baselines.
As with any other style of GPS surveying, shorter baselines result in improved
accuracy because of reduced effects of atmospheric interference.
The RTN reference stations allow for network atmospheric modeling resulting
gAGE/UPC
in improved accuracy. With RTK, atmospheric effects are computed using

(usually) one location.
All users of the system are using a common, established ref. coordinate frame.
56
gAGE Limitations of VRS surveying include:
There is a high cost of setting up and maintaining the RTN and to use
an RTN setup by other organizations there is typically a yearly

subscription fee that must be paid for network access.
Use of the RTN can be limited by cell phone coverage and system
down times.
Availability is dependent on network extent and accuracy can be

affected by the network density.
http://water.usgs.gov/osw/gps/real-time_network.html
Comment:
VRS is the most widely used implementation method of Network RTK
(NRTK). Other possible implementations are briefly described in:
gAGE/UPC
http://www.wasoft.de/e/iagwg451/intro/introduction.html

57
gAGE Local Area DGNSS (LADGNSS): GBAS

LADGNSS includes a Master station and several monitor stations. The master
station collects the range measurements of the monitor stations and process
the data to generate the range corrections, which are broadcasted to users
Examples using L1 carrier smoothed code are the Local Area Augmentation
System (LAAS) or the Ground Based Augmentation System (GBAS), where a
ground facility computes differential corrections and integrity data from
measurements collected by
several redundant receivers.
This system is used to
support aircraft operations
during approach and
landing. The differential
corrections are transmitted
on a VHF channel, up to
about 40km. Meter level
gAGE/UPC
accuracies with integrity

fulfilling the stringent
requirements of Civil
Aviation are met.
58
gAGE
gAGE/UPC

59
gAGE Differential GNSS (DGNSS) IMPLEMENTATIONS


The differential corrections can be broadcast as an:
Scalar correction (Local Area), where all corrections are lumped

together.
Examples: GBAS, LAAS, RASAN, RTK, VRS.
Vector correction (Wide Area), where the corrections are given for
each error source separately (state-space approach)
Examples: SBAS (WAAS, EGNOS...), DGPS (JPL), Fast-PPP.
gAGE/UPC

60
gAGE Wide Area DGNSS (WADGNSS)
Differential Corrections to cover Continent-wide (or world-wide) users must be
broadcast for each error source separately: Satellite clocks, ephemeris and
ionosphere.
These corrections are computed by a Central processing Facility (CPF) from the
range measurements of the monitor stations network with baselines of several
hundreds up to thousand of kilometres.
Examples using L1 carrier

smoothed code are the Satellite
Based Augmentation Systems
(SBAS), e.g. WAAS, EGNOS,
MSASS, for Civil Aviation, where
differential corrections and integrity
data fulfilling the Civil aviation
requirements are broadcast over
continental areas by a GEO
satellite. Meter level accuracies
gAGE/UPC
with integrity are met.

Evolution to a dual frequency
(L1,L5) system.
61
Error Mitigation
gAGE
Error Local Area Wide Area

component (GBAS) (SBAS)
Satellite clock Estimation and
Ephemeris Common Mode Removal each
error
Ionosphere Differencing component
Troposphere Fixed Model
Multipath and Carrier Smoothing by user

Receiver Noise
gAGE/UPC

62
gAGE
EGNOS
gAGE/UPC

63
gAGE
Wide Area DGNSS (WADGNSS)
For dual frequency users, the JPL-NASA provides

Real-Time Global Differential GPS (GDGPS) to world
wide users to achieve decimetre/centimetre level of
accuracy, after the best part of one hour, using the
Precise Point Positioning (PPP) technique.
Indeed, precise orbits and clocks are computed from a global sparse reference
stations network. The ionospheric error is eliminated from the combination the
two frequencies.
gAGE/UPC
http://www.gdgps.net/monitoring/index.html

64
gAGE Current research: The Fast-PPP
Fast-PPP have prove that the high accuracy positioning of PPP users, can be
reached very quickly (almost instantaneously) over continental areas with the
3-frequency Galileo signals. Moreover:

- World-wide users can perform undifferenced ambiguity fixing, thanks
to the broadcasting of the fractional part of carrier ambiguities.
- Continent-wide area users (e.g. Europe) can achieve the accuracy quickly
(few minutes, with 2-frequency or single epoch with 3-frequency, instead
the 1/2-1 hour of PPP), thanks to the broadcasting of highly accurate
ionospheric corrections.
- Single frequency users can take also benefit of the highly accurate
ionospheric corrections achieving submetre positioning since the beginning
and decimetre level positioning after the best pat of one hour.
As in any WADGNSS, the differential corrections (satellite clocks, ephemeris,

ionosphere, fractional part of ambiguities and DCBs) are computed by CPF
from the measurements collected by an sparse network (few tens world wide
gAGE/UPC
distributed plus 30-40 for the continental enhancement of ionosphere model)
The Fast-PPP technique has been developed gAGE/UPC and is protected by

several ESA-funded patents.
65
gAGE
CPF Corrections
Service
Time +
Correction Coverage Content

Update Added
capability
Fast Global ~5s Satellite Clocks

Classic PPP
Orbit Corrections
Slow Global 300s Satellite DCBs +

Fractional part of Ambiguity
ambiguities (B1, BW) Fixing
+
Fast PPP
Ionospheric Continental 300s Iono. corrections +
Single
Frequency
Global/ +
Integrity ~5/300s Confidence bounds
gAGE/UPC
Continental Integrity
The additional required bandwidth of F-PPP is about 10% of the

classical PPP bandwidth
66
gAGE
Horizontal Vertical
24h of data: The
full user state
has been reset
No iono
every two hours
to better
characterize the
convergence
10 cm process.
10 cm
gAGE/UPC
With iono
M LVL (violet), EUSK (light blue) and EI JS (brown), at 252,@ J.170

& 94 km
Sanz & J.M. Juan
respectively far from the nearest reference receiver BRUS.
gAGE
Contents

errors.
corrections.
gAGE/UPC

68
gAGE DGNSS Commercial services
Commercial WADGNSS services are already operational and in world-wide use
for different applications: agriculture (e.g. OmniSTAR or CenterPoint RTX from

Trimble), operations at sea (e.g. Starfix and Skyfix from Fugro), among others
http://www.fugro.com/ http://www.trimble.com
OmniSTAR provides four levels of service: http://www.omnistar.com/

Virtual base Station (VBS) offering sub-meter positioning,
World-wide service XP delivering better than 20 centimeter accuracy,
High performance (HP) service delivering greater than10 centimeter
accuracy
OmniSTAR G2 service combines GPS plus GLONASS-based corrections to
provide decimeter level positioning.
OmniSTAR services were initially introduced by Fugro company and in
2011 was acquired by Trimble company .
gAGE/UPC
Similar levels of services are provided by Starfix: http://www.starfix.com

Starfix.L1 , Starfix.XP, Starfix.HP , Starfix.G2
69
gAGE
OmniSTAR HP
10 cm High
Performance
http://www.omnistar.com/
SubscriptionServices/
OmniSTARHP.aspx
OmniSTAR HP (10cm) service is the most accurate solution available

in the OmniSTAR portfolio of correction solutions. It is a L1/L2 solution
requiring a dual frequency receiver.
OmniSTAR HP corrections are modeled on a network of reference sites
using carrier phase measurement to maximize accuracy.
The expected 2-sigma (95%) accuracy of OmniSTAR HP is 10cm. It is
particularly useful for Agricultural Machine guidance and many surveying
gAGE/UPC
tasks. It operates in real time and without the need for local Base Stations
or telemetry links. OmniSTAR HP is a true advance in the use of GPS for
on-the-go precise positioning.
70
gAGE
OmniSTAR XP:
15 cm Worldwide
Service
OmniSTAR XP (15cm) is a worldwide dual frequency high accuracy

solution. It is a L1/L2 solution requiring a dual frequency receiver.
Orbit and Clock correction data is used together with atmospheric
corrections derived from the dual frequency data.
By utilizing carrier phase measurement, very high accuracy can be achieved.
OmniSTAR XP service provides short term accuracy of 1-2 inches and long
term repeatability of better than 10 centimeters, 95%CEP.
gAGE/UPC
It is especially suited for Agricultural automatic steering systems. While it is

slightly less accurate than OmniSTAR HP, it is available worldwide and its
accuracy is a significant improvement over regional DGNSS such as WAAS.
71
gAGE
OmniSTAR VBS
Global Reliable
Sub-Meter Accuracy
OmniSTAR VBS is the foundational "sub-meter" level of service. It is an

L1 only, code phase pseudo-range solution.
Pseudo-range correction data from OmniSTARs regional reference sites is
broadcast via satellite link to the user receiver.
These data are used, together with atmospheric modeling and knowledge of
the receivers location, to generate an internal RTCM SC104 correction
specific to that location. This correction is then applied to the R-T solution.
gAGE/UPC
A typical 24-hour sample of OmniSTAR VBS will show a 2-sigma (95%) of

significantly less than 1 meter horizontal position error and the 3-sigma
(99%) horizontal error will be close to 1 meter.
72
gAGE
OmniSTAR G2
GPS + GLONASS
OmniSTAR G2 is a worldwide dual frequency high-accuracy solution

which uses Orbit and Clock correction data.
OmniSTAR G2 includes GLONASS satellites and GLONASS correction data in
the solution. The addition of GLONASS to the solution significantly increases
the number of satellites available which is useful when faced with
conditions that limit satellite visibility, such as terrain, vegetation or buildings.
OmniSTAR G2 service provides short-term accuracy of 1-2 inches and long
gAGE/UPC
term repeatability of better than 10 cm, 95%CEP. It is especially suited for

operations in areas where trees or buildings may block the view of the sky and
in areas affected by scintillation during times of high sunspot activity.
73
Observa Baseline Broadcast accuracy Init. Examples of

gAGE
ble message time Products
DGPS Smoothed <100km PRC, RRC ~ metre Single
(code) code (L1) epoch RASANT

RTK Carrier <10-15km Carrier ~ cm ~10 s Several
L1/L2, L1 measurements packages
GBAS/ Smoothed < 40 km PRC, RRC ~ metre Single Honeywell
LA- LAAS code (L1) + Integrity epoch GBAS station
DGNSS VRS Carrier < 50 km Virtual ~10 - Omnistar HP,
L1/L2, L1 Carrier few cm 30 s StarFix HP
measurements
SBAS Smoothed Continental Orbits+ ~ metre Single WAAS,
code (L1) Clocks+ + Integrity epoch EGNOS,
Ionosphere MSAS
GDGNSS Iono-free Worldwide Orbits + 1/2h- Omnistar XP,
WA- code and Clocks ~ dm 1h. StarFix XP,
DGNSS carrier JPL-GDGPS
F-PPP L1, L2 Worldwide Orbits+ ~ dm/cm
gAGE/UPC
code and with Clocks+ (ambiguity Feasibility

carrier Continental Ionosphere+ fixing ~5 m studies
enhancement DCBs + capability)
Frac. Ambig.

74
gAGE
References

23/1. ESA Communications, May 2013.
Communications, May 2013.
[RD-5] B. W. Parkinson and J.J. Spilker. Global Positioning System: Theory and
Applications, Vol1 and Vol2. Progress in Astronautics and Aeronautics,
Volume 164, Cambridge, Massachusetts, US.
gAGE/UPC

75
gAGE
Thank you
gAGE/UPC

76
gAGE
Backup slides
gAGE/UPC

77
Example of
gAGE
Differential Atmospheric propagation effects analysis
GODN
USN3
76 m
GODS
DGPS DGPS
gAGE/UPC

78
gAGE
Single and double differences of receivers/satellites
x x x
x { k
x R
x
' { rov ref ' { ' k
' R
Receiver errors affecting SIS errors affecting both

both satellites are removed receivers are removed rov ref
(e.g. Receiver clock) (e.g. Satellite clocks,...)
Receiver errors common for all satellites do not affect positioning

(as they are assimilated in the receiver clock estimate). Thence:
gAGE/UPC
- Only residual errors in single differences between sat. affect absolute posit.
- Only residual errors in double differences between sat. and receivers affect
relative positioning.
Exercise,
Master of Science in GNSS Discuss the previous @sentences.
J. Sanz & J.M. Juan
79
gAGE
Depicting atmosphere propagation errors affecting DGNSS:
Double-differences between satellites and receivers
U rec c (dtrec dt sat ) Trec I1,satrec O1Zrec rec mL H L

b1, rec b1sat O1 N1,sat
sat sat sat sat
L1rec 1 1
U recR c (dtrecR dt sat ) TrecR I1,satrecR O1ZrecR recR mL H L

b1, recR b1sat O1 N1,sat
sat sat sat sat
L1recR 1 1
Differencing between receivers cancels satellite-only-dependent terms
'L1sat 'U sat c'dt 'T sat 'I1sat O1'Z sat 'b1 O1'N1sat m'L1 H 'L1
'L1sat 'U sat c'dt 'T sat 'I1sat O1'Z sat 'b1 O1'N1sat m'L1 H 'L1
'L1satR 'U satR c'dt 'T satR 'I1satR O1'Z satR 'b1 O1'N1satR m'L1 H 'L1
Differencing between satellites cancels receiver-only dependent terms

gAGE/UPC
'L1 'U 'T 'I1 O1'Z O1'N1 m'L1 H 'L1
Master of Science in GNSS Only residual errors in double differences between

sat. and receivers affect relative positioning. 80
gAGE
Double-differences between satellites and receivers
'U 'T 'I1 O1'Z O1'N1 m'L1 H 'L1

'L1
Satellite and receiver clocks and fractional part of ambiguities cancel.
Comments:
The wind-up term 'Z can be neglected, except over long baselines.
Double-differenced ambiguities are integer numbers of wavelengths.
Exercise,
f12 L1 f 22 L2
Show that, neglecting the wind-up, Lc
f12 f 22
the following expressions are met over
a continuous carrier phase arch: 2
f1
J
'L1 'U 'T 'I1 bias1 f2
'L2 'U 'T J 'I1 bias2 Ionosphere-free:
gAGE/UPC
'LC 'U 'T biasC Only Troposphere

Geometry-free:
'L1 'L2 1 J 'I1 biasI
Only Ionosphere
81
gAGE
GODN
USN3
76 m
GODS
DGPS DGPS
gAGE/UPC

82
gAGE
GODN
USN3 76 m
GODS
' L1 U ' Tropo Iono ' L1 U ' Tropo Iono

gAGE/UPC
PRN06-PRN13 PRN05-PRN15

83
gAGE
GODN
USN3
76 m
GODS
DGPS DGPS
gAGE/UPC

84
gAGE
research group of Astronomy and Geomatics GODN
USN3
76 m
GODS
gAGE/UPC

85
gAGE
L1 L2
' L1 U ' Tropo Iono f22

' L2 U u ' Tropo Iono
f12
PRN05-PRN15
Ionosphere-free Geometry-free
' L1 L2 'Iono
gAGE/UPC
' LC U 'Tropo

86
gAGE
research group of Astronomy and Geomatics L1 L2
f22
' L1 U ' Tropo Iono ' L2 U u ' Tropo Iono
f12
PRN06-PRN13
Ionosphere-free Geometry-free
gAGE/UPC
' L1 L2 'Iono
' LC U 'Tropo

87
gAGE
Other
Backup slides
gAGE/UPC

88
gAGE
research group of Astronomy and Geomatics ' L1 U ' Tropo Iono
' L1 U ' Tropo Iono

' L2 U ' Tropo Iono
PRN05-PRN15
' L1 L2 'Iono
gAGE/UPC
' LC U 'Tropo

89
gAGE
' L1 U ' Tropo Iono
' L1 U ' Tropo Iono ' L2 U ' Tropo Iono
PRN06-PRN13
gAGE/UPC
' L1 L2 'Iono
' LC U 'Tropo

90
gAGE
GNSS Positioning
Standalone Positioning: GNSS receiver autonomous positioning
using broadcast orbits and clocks (SPS, PPS).

gAGE/UPC

91
gAGE
GNSS Positioning: Space Segment errors
Orbit error
Clock error
gAGE/UPC

92
gAGE
GNSS Positioning: Propagation errors
Ionosphere
Troposphere
gAGE/UPC

93
gAGE
gAGE/UPC

94
gAGE
GNSS Positioning: Propagation errors
Ionosphere
Troposphere
gAGE/UPC

95
gAGE
Lecture 5
Precise Positioning
with carrier phase (PPP)
gAGE/UPC

1
gAGE

gAGE/UPC
22 Jan 2015

2
gAGE
Precise Point Positioning (PPP)
1.1. Precise Orbits and Clocks

1.2. Code and carrier measurements and modelling errors
1.3. Linear observation model for PPP
1.4. Parameter estimation: Floating Ambiguities
1.5. Carrier Ambiguity fixing concept: DD and undifferenced
gAGE/UPC

3
gAGE The PPP technique allows centimetre-level accuracy to be

achieved for static positioning and decimetre level, or
better, for kinematic positioning, after the best part of
one hour.
24h Static positioning Kinematic positioning

gAGE/UPC
This high accuracy requires the use of code and carrier measurements
and an accurate measurement modelling up to centimetre level or
better.
4
gAGE

gAGE/UPC

5
gAGE Satellite Orbits and Clocks

Broadcast versus precise
Geomatics
a ic
mat
With S/A=on, clocks were

eom
degraded several tens of

Ge
meters.
Astronomy
sttrronom a dG
and
o y an
S/A=on off
As
of A
S/A=off
a ch
research gro po
oup
h group
Under S/A=off , the broadcast

sear
orbits and clocks are accurate

ese
at few meters level (see plots

re
at left).
UPC
gAGE/UPC
IGS precise orbits & clocks

/U
are accurate at few cm level

6
gAGE
Zoom
gAGE/UPC

Broadcast Ephemeris Updates
7
gAGE
IGS Precise orbit and clock products:
RMS accuracy, latency and sampling
gAGE/UPC
http://igscb.jpl.nasa.gov/components/prods.html

8
gAGE
Computation of satellite coordinates from precise
research group of Astronomy and Geomatics products.
Precise orbits for GPS satellites can be found on the International

Orbits are given by (x,y,z) coordinates with a sampling rate of 15

minutes. The satellite coordinates between epochs can be computed
by polynomial interpolation. A 10th-order polynomial is enough for a
centimetre level of accuracy with 15 min data.
gAGE/UPC

9
gAGE
gAGE/UPC

10
gAGE
Computation of satellite clocks from precise products
Precise clocks for GPS satellites can be found on the International

They are providing precise orbits and clock files with a sampling rate
of 15 min, as well as precise clock files with a sample rate of 5 min
and 30 s in SP3 format.
Some centres also provide GPS satellite clocks with a 5 s sampling
rate, like the les obtained from the Crustal Dynamics Data Information
System (CDDIS) site.
Stable clocks with a sampling rate of 30 s or higher can be

interpolated with a first-order polynomial to a few centimetres of
accuracy. Clocks with a lower sampling rate should not be
gAGE/UPC
interpolated, because clocks evolve as random walk processes.

11
gAGE
gAGE/UPC

12
gAGE
gAGE/UPC

13
gAGE
gAGE/UPC

14
gAGE
Exercise:
Show that a common error
on all satellites does not
affect user positioning.
gAGE/UPC

15
gAGE
Contents

gAGE/UPC

16
gAGE
Measurements: Code and carrier
For high-accuracy positioning, the carrier phase must
be used, besides the code pseudorange.
As commented before, the carrier measurements are
very precise, typically at the level of a few
millimetres, but contain unknown ambiguities which
change every time the receiver locks the signal after
a cycle slip.
Nevertheless, such ambiguities can be estimated in
the navigation solution, together with the coordinates
and other parameters.
gAGE/UPC

17
gAGE
G
Carrier is ambiguous, but
precise
and
my an Geomatics
Geo
nd Ge cs
mattiic
om
Zoom of carrier noise

Astronomy
nom
Code is (LC prefit-residuals)

tron
Cycle-slip unambiguous,
str
but noisy
off As
up o
research group
Code measurements are unambiguous but noisy (meter level noise).

Carrier measurements are precise (few millimetres of noise) but ambiguous
(the unknown biases can reach thousands of km).
Carrier phase biases are estimated in the navigation filter along with
gAGE/UPC
the other parameters (coordinates, clock offsets, etc.). If these biases were
fixed, measurements accurate to the level of few millimetres would be
available for positioning. However, some time is needed to decorrelate such
biases from the other parameters in the filter, and the estimated values are
notoffully
Master Science unbiased.
in GNSS @ J. Sanz & J.M. Juan
18
Precise Point Positioning: Static
From default configuration of [PPP Template],
Select Static in the [Filter] panel. Run gLAB and plot results
Centimetre error
level positioning after
24h data
Receiver positioned as a
permanent station (static mode)

Precise Point Positioning: Kinematic

From default configuration of [PPP Template],
Select kinematics in the [Filter] panel. Run gLAB and plot results
Decimetre error level

navigation after the
best part of an hour
Receiver navigated as a rover
in a pure kinematic mode.

gAGE
research group of Astronomy and Geomatics PPP Model components
In the laboratory session (Tutorial 1)

we review the measurements
modelling for the Standard Point
Positioning (SPP). A brief summary is
given next.
After this summary, we will focus on

the additional modelling need for
Precise Point Positioning.
Remember that the error component most difficult to model is the ionosphere.
gAGE/UPC
But, in the PPP technique the ionosphere error is removed (more than 99.9%)
using dual-frequency measurements in the ionosphere-free combination (Lc,Pc).
This combination also removes the Differential Code Bias (or TGD).

21
gAGE
U U
U'
U
Astronomy
p of Astr ono
n and Geomatics
U
U'
up
group
Coord. at emission
research
res
es
U
/UPC
gAGE/UPC
Earth rotation

Satellite coordinates computation at signal emission time
22
gAGE
Troposphere
gAGE/UPC
Ionosphere

Signal propagation errors on the Atmosphere
23
gAGE
Satellite clock
Relativity sat clock

gAGE/UPC
TGD (satellite)

Satellite clocks and Total Group Delay (TGD)
gAGE
research group of Astronomy and Geomatics Additional Modelling for PPP
The PPP technique allows centimetre-
level of accuracy to be achieved for
static positioning and decimetre level,
or better, for kinematic navigation.
This high accuracy requires an
accurate modelling up to the
centimetre level or better ,
where all previous model terms must
be taken into account (except
ionosphere and TGD [*]), plus some
additional terms given next:
[*] Remember that in the PPP technique the ionosphere error is removed
gAGE/UPC
(more than 99.9%) using dual-frequency measurements in the ionosphere-

free combination (Lc,Pc). This combination also removes the Differential
Code Bias (or TGD).

25
Additional Modelling for PPP

Satellite Mass Center
to Antenna Phase gLAB1.out gLAB1.out
Center Satellite
Antenna
Phase
Center
(APC)
Satellite
Mass gLAB.out gLAB.out
Center
(MC) gLAB.out Satellite MC to APC:
The satellite MC to APC
Broadcast orbits are eccentricity vector depends
on the satellite. The APC
referred to the antenna values used in the IGS orbits
phase center, but IGS and clocks products are
precise orbits are referred referred to the iono-free
to the satellite mass combination (LC, PC) . They
are given in the IGS ANTEX
center. files (e.g., igs05.atx).
gAGE/UPC Tutorial associated to the GNSS Data Processing book

Research group of Astronomy & Geomatics J. Sanz Subirana, J.M. Juan Zornoza, M. Hernndez-Pajares 26
Receiver Antenna
gLAB1.out gLAB1.out
Phase center (APC)
L2 Antenna Phase
APC center
L1 Antenna Phase
center
Antenna Reference
ARP Point (ARP)
(ARP)
gLAB.out gLAB.out
GNSS measurements are

referred to the APC. This is Receiver APC:
not necessarily the geometric The antenna used for this
center of the antenna, and it experiment, has the APC
depends on the signal position vertically shifted
frequency and the incoming regarding ARP.
radio signal direction. Thence, neglecting this
correction, an error on the
For geodetic positioning a gLAB.out vertical component occurs,
reference tied to the antenna but not in the horizontal one.
(ARP) or to monument is used
gAGE/UPC Tutorial associated to the GNSS Data Processing book

gAGE
Antenna biases and orientation:
The satellite and receiver antenna phase centres can be
found in the IGS ANTEX files, after GPS week 1400 (Nov. 2006)
The carrier phase wind-up effect due to the satellite's motion
must be taken into account.
gAGE/UPC

28
Wind-up affects only
carrier phase. It is due to the gLAB1.out gLAB1.out
electromagnetic nature of
circularly polarized waves of
GNSS signals.
As the satellite moves along
its orbital path, it performs a
rotation to keep its solar
panels pointing to the Sun gLAB.out gLAB.out
direction. This rotation
causes a carrier variation,
gLAB.out Wind-Up
and thence, a range
measurement variation.
Wind-up changes smoothly
Satellite rotation along continuous carrier
Phase phase arcs.
variation
In the position domain, wind-
up affects both vertical and
horizontal components.

gAGE Additional Modelling for PPP: Eclipse condition

High-accuracy GNSS positioning degrades during the GNSS satellites'

eclipse seasons.
Once the satellite goes into shadow, the radiation pressure vanishes.
This effect introduces errors in the satellite dynamics due to the
difficulty of properly modelling the solar radiation pressure.
On the other hand, the satellite's solar sensors lose sight of the Sun
and the attitude control (trying to keep the panels facing the Sun).
gAGE/UPC
As a consequence, the orbit during shadow and eclipse periods may be

considerably degraded and the removal of satellites under such conditions
can improve the high-precision positioning results.

30
gAGE Additional Modelling for PPP: Atmospheric Effects
The ionospheric refraction and TGDs are removed using the
ionosphere-free combination of code and carrier measurements

(PC, LC).
The tropospheric refraction can be modelled by Dry and
Wet components (and different mappings are usually used for
both components, e.g. mapping of Niell).
A residual tropospheric
delay is estimated (as
wet ZTD delay) in the
Kalman filter, together with
the coordinates, clock and
carrier phase biases.
The troposphere is estimated as
a Random Walk process in the
gAGE/UPC
Kalman Filter. A process noise of

1cm/sqrt(h) has been taken.

31
gAGE Additional Modelling for PPP: Earth Deformation

Earth deformation effects:
Solid tides must be modelled by equations

Ocean loading and pole tides are second-order effects and
can be neglected for PPP accuracies at the centimetre level
Solid Tides concern the movement of

Earth's crust (and thus the variation in the
receiver's location coordinates) due to
Sun
gravitational attractive forces produced by
external bodies, mainly the Sun and Moon.
Solid tides produce vertical and horizontal
displacements that can be expressed by
the spherical harmonics expansion (m, n),
gAGE/UPC
Moon characterised by the Love and Shida

numbers hmn and lmn.

32
Solid Tides gLAB1.out gLAB1.out
It comprises the Earths
crust movement (and
thence receiver
coordinates variations)
due to the gravitational
attraction forces
produced by external
V
gLAB.out gLAB.out
bodies, mainly the Sun
and the Moon. Solid Tides:
These effects do not affect the
Sun GNSS signals, but if they were
not considered, the station
coordinates would oscillate with
relation to a mean value.
They produce vertical (mainly)
gLAB.out and horizontal displacements.
Moon

gAGE
Contents

gAGE/UPC

34
gAGE Linear observation model for PPP
It is based code and carrier measurements in the
ionosphere-free combination (Pc, Lc), which are

modelled as follows:
sat
PC rec U rec
sat
c (dtrec dt sat ) Troprec
sat
M Pc H P C
sat
LC rec U rec
sat
c (dtrec dt sat ) Troprec
sat
ON Zrec
sat
BC rec
sat
mL H L C C
where
f12 P1 f 22 P2 f12 L1 f 22 L2 O
PC ; LC BC ON B1 W BW
f12 f 22 f12 f 22 O2
BW B1 B2
Ionosphere is removed
gAGE/UPC
Remark, U is referred to the Antenna Phase Centres (APC) of satellite

and receiver antennas in the ionosphere free combination.

35
gAGE
Linear model: For each satellite in view
sat
PC rec U rec
sat
c ( dtrec dt sat ) Trop H
Linearising Uaround an a priori receiver position ( xrec ,0 , yrec ,0 , zrec ,0 )

xrec ,o x sat yrec ,o y sat zrec ,o z sat
U sat
'xrec 'yrec 'zrec c dtrec dt sat Trop
U U U
rec ,o sat sat sat
rec ,o rec ,o rec ,o
where:
'xrec =xrec xrec ,o ; 'yrec =yrec yrec ,o ; 'zrec zrec zrec ,o
sat
and taking: Troprec
sat
Trop0 rec M wet
sat
, rec 'TrZ , wet
xrec ,o x sat yrec ,o y sat zrec ,o z sat
P sat
U sat
c dt Trop
sat sat
'xrec 'yrec 'zrec c dtrec M wet
sat
'TrZ ,wet
U U U
C rec rec ,o 0 rec sat sat sat
gAGE/UPC
rec , o rec , o rec , o

The same for carrier, but adding the ambiguity as an unknown
36
Following the same procedure as with the code based
gAGE
positioning (SPP), the linear observation model y = G x
for the code and carrier measurements can be written as
follows: y Gx
xo ,rec x1 yo ,rec y1 zo ,rec z1 1

1 M wet 0 0 'xrec
U0,rec U U
1 1 1
0, rec 0, rec
Prefit ( Pc)1 x x1 yo ,rec y1 zo ,rec z1 'yrec
o ,rec 1
1 M wet 1 0 'zrec
1
Prefit ( Lc) U0,1 rec U0,1 rec U0,1 rec
........ cdtrec

'TrZ , wet
Prefit ( Pc) n x xn yo ,rec y n zo ,rec z n
o ,recn 1 M wet
n
0 0 1

n
BC
Prefit ( Lc) U0,rec U n
0, rec U n
0, rec

xo ,rec x yo ,rec y n zo ,rec z n
n
n
Un 1 M wet
n
0 1 BC
0, rec U n
0, rec U n
0, rec
gAGE/UPC
Prefit ( PC ) k PC k U0k cdt k Trop0k Carrier

ambiguities
Prefit ( LC ) k LC k U0k cdt k Trop0k ON Z k
37
Equivalent notation:
gAGE
0
Using U U0 T U , where
U0
The previous system, can be written as:
T0 rec
1 1
1 M wet 0 0 'rrec
Prefit ( Pc)1 cdt

1 rec
0 rec
T 1 1
Prefit ( Lc) 1 M wet 1 0
'Trz , wet
........ 1

T recn B
Prefit ( Pc) n 1 M wet
n
0 0
n
0

T recn
Prefit ( Lc ) 0
1 M wet
n
0 1 B n
gAGE/UPC
Prefit ( PC ) k PC k U0k cdt k Trop0k Carrier

ambiguities
Prefit ( LC ) k LC k U0k cdt k Trop0k ON Z k
38
gAGE

gAGE/UPC

39
gAGE Parameter estimation PPP:

Floating Ambiguities
The linear observation model y = G x can be solved using

the Kalman filter. The next stochastic model can be used
Carrier phase biases (BC ) are taken as `constant along
continuous phase arcs, and as `white noise when a cycle slip
happens (V= 104 m can be taken, for instance) FLOATED AMB.
Wet tropospheric delay (Trz;wet) is taken as a random walk
process (with dV/dt= 1 cm2/h, for instance).
Receiver clock (cdt) is taken as a white-noise process
(with V= 3.105 m , i.e. 1 ms for instance).
Receiver coordinates ('x, 'y, 'z)
gAGE/UPC
o For static positioning the coordinates are taken as constants.

o For kinematic positioning the coordinates are taken as white
noise or a random walk process.
40
gAGE
Comment: Floating Ambiguities
This solution procedure is called floating ambiguities.
`Floating in the sense that the ambiguities are estimated
by the filter `as real numbers.
The bias estimations BC will converge to a solution after a
transition time that depends on the observation geometry,
model quality and data noise.
In general, one must expect errors at the decimetre level,
or better, in pure kinematic positioning (after the best
part of one hour) and at the centimetre level in static PPP
over 24h data.
See exercises in the laboratory session (Tutorial 1).
gAGE/UPC

41
gAGE
Contents

gAGE/UPC

42
gAGE
Carrier Ambiguity Fixing concept
In the previous formulation, the ambiguities have been estimated as
real numbers in the Kalman filter, without exploiting its integer
nature.
That is, the ambiguities in L1, L2 or LW signals are an integer number
(N) of its associated wavelength (O) plus a fractional part associated
to the satellite and to the receiver.
sat
B1, rec O1 N1, rec
sat
b1, rec b1sat
BC
O
ON N1 w NW bC ,rec bC sat
O2
sat
B2, rec O2 N 2, rec
sat
b2, rec b2 sat
Bc is not an integer number of ON but
sat
BW , rec OW NW , rec
sat
bW , rec bW sat can be related with the integers N1, LW
gAGE/UPC
The Ambiguity Fixing techniques take benefit of this INTEGER

NATURE of N1, N2 and Nw ambiguities to properly fix them, reducing
convergence time, and thence, achieving high accuracy quickly.
43
gAGE Carrier Ambiguity Fixing concept

Two different approaches can be considered:
Double differenced Ambiguity Fixing:

This is the classical approach which relies in the fact that the
fractional part of carrier ambiguities cancels when forming the
double differences between receivers and satellites:
That is, given: Brec sat
O N rec
sat
brec b sat
The double differences, regarding a reference receiver and
satellite, yield:
rec rec
'B sat B sat B sat B sat , R B sat , R O 'N sat
rec , R rec rec , R rec
where the satellite and receiver ambiguity terms (brec, bsat)

cancel out.
Absolute (or undifferenced) ambiguity fixing:
This is a new approach, where the fractional part of the ambiguities
gAGE/UPC
are estimated from a global network of permanent stations and

provided to the users. Thus, the user can remove this fractional part
and fix the remaining ambiguity as an integer number.
44
gAGE
Example of Ambiguity fixing layout
Kalman filter: User solution (x,y,z)
'LW 'PN
'NW
Differential Ionosphere OW rundoff
Error
Ambiguity FIXED
'L1 'L2 'I O2 'NW

'N1
O2 O1 rundoff
OW
'BC ON 'N1 'NW Quick convergence time: Few
O2
gAGE/UPC
minutes to fix 'NW

Baseline limited by the Ionosphere

45
gAGE Example of Ambiguity fixing layout

Kalman filter: User solution (x,y,z)

Floated estimate 'BC
'LW 'PN
'NW
OW rundoff
Ambiguity FIXED
1 O
'N1
ON

'BC W 'NW
O2 rundoff
OW
'BC ON 'N1 'NW
Large convergence time, due to the
O2
gAGE/UPC

time-span need for 'BC to converge
Immune to the ionosphere

46
gAGE
References

[RD-5] Juan, J.M., Hernandez-Pajares, M., Sanz, J., Ramos- Bosch, P., Aragon-
Angel, A., Orus, R., Ochieng, W., Feng, S., Coutinho, P., Samson, J.
and Tossaint, M., 2012a. Enhanced Precise Point Positioning for GNSS
Users. IEEE Transactions on Geoscience and Remote Sensing PP, issue
gAGE/UPC
99, pp. 1-10.

47
gAGE
Thank you
gAGE/UPC

48
gAGE
Lecture 6
Differential positioning with
Code pseudoranges
gAGE/UPC

1
gAGE

gAGE/UPC
24 April 2014

2
gAGE
Contents
1. Linear model for DGNSS: Single Differences

1.1. Linear model
1.2. Geographic decorrelation of ephemeris errors
1.3. Error mitigation and `short baseline concept
1.4. Differential code based positioning
2. Augmentation Systems
2.1. Introduction
2.2. Ground-Based Augmentation system (GBAS)
2.3. Satellite based Augmentation System (SBAS)
gAGE/UPC

3
gAGE Error mitigation: DGNSS residual error

Errors are similar for users separated tens, even hundred of kilometres,
and these errors vary slowly with time. That is, the errors are
correlated on space and time.
The spatial decorrelation depends on the error component (e.g. Clocks

not decorrelate, ionosphere ~100km...). Thence, long baselines need a
reference stations network.
Error
gAGE/UPC
Short-baselines
Long-baselines

4
gAGE
Linear model for Differential Positioning
Code and carrier measurements
Pi j Ui j c G ti G t j Ti j I i j Ki K j Mi j Q j
Pi
Lij Ui j c G ti G t j Ti j Ii j O Zi j O Ni j bi b j mij Q j
Li
Satellite coordinates 'r j

deviation 'r j r j r0j
When approximate values of both
receiver and satellite APC positions
are taken, a linearization around
them yields:
Ui j U0ij
Ui j U0ij 0ij 'ri 0ij 'r j
j
0 i r0j roi
j
'ri j
r0j roi
0i
User coordinates
i
deviation 'ri ri r0i

gAGE/UPC
r0i 'ri : Receiver coordinates error

ri 'r j : Satellite coordinates error

To Earth Centre 5
gAGE
Pi j Ui j c G ti G t j Ti j I i j Ki K j Mi j Q Pi
j
Lij Ui j c G ti G t j Ti j Ii j O Zi j O Ni j bi b j mij Q Li
j
Single difference (x)ruj { '(x)ruj (x)uj (x)rj 'r j
Pu j Uuj c G tu G t j Tu j Iuj Ku K j Q j
Pu
Pr j Urj c G tr G t j Tr j I rj Kr K j Q Pr
j
Pruj Uruj c G tru Truj I ruj Kru Q j uj rj

P ru rru
User Reference
receiver Station
Baseline
The same for the carrier :
gAGE/UPC
rru ru rr
Lruj Uruj c G tru Truj I ruj O Zruj O Nruj bru Q j

L ru
ru rr

To Earth Centre 6
gAGE
Pi j Ui j c G ti G t j Ti j I i j Ki K j Mi j Q Pi
j
Lij Ui j c G ti G t j Ti j Ii j O Zi j O Ni j bi b j mij Q j
Li
Single difference (x)ru { '(x)ru (x)uj (x)rj 'r j

j j
Pruj Uruj c G tru Truj I ruj Kru Q j

P ru
Lruj Uruj c G tru Truj I ruj O Zruj O Nruj bru Q L ru

j
Single difference cancels:

Satellite clock (G t )
j
j
Satellite code instrumental delays ( K )
Satellite carrier instrumental delays (b )
j uj rj
User
rru
Referen
Single differences mitigate/remove errors due receiver Station
Baseline
Satellite Ephemeris ('r j )
gAGE/UPC
rru ru rr
Ionosphere ( I i )
j
Troposphere (Ti )
j
ru rr
Wind-up (Zi )
j
The residual errors will depend upon the

baseline
Master of Sciencelength.
in GNSS @ J. Sanz & J.M. Juan
To Earth Centre 7
gAGE
Single-Difference of measurements
(corrected by geometric range!!)
PRN06
PRN06
PLAN-GARR: 15km
PRN30
'( L1 U ) { L1ru
sat
Urusat PRN30 '( P1 U ) { L1ru
sat
Urusat
Dif. Wind-up: Very small
'( L1 U ) { Lruj Uruj c G tru Truj I ruj O Zruj O Nruj bru Q L ruj
'( P1 U ) { Pruj Uruj c G tru Truj I ruj Kru Q P ruj

gAGE/UPC
Dif. Instrumental
Dif. Receiver clock: delays and carrier
Main variations Common Dif. Tropo. and Iono. : ambiguities:
for all satellites Small variations constant
8
gAGE
research group of Astronomy and Geomatics (corrected by geometric range!!)
PRN06
PRN06
PLAN-GARR: 15km
PRN30 PRN30
'( L1 U ) { L1ru
sat
Urusat '( P1 U ) { L1ru
sat
Urusat
PRN06 PRN06
IND1-IND2: 7m IND1-IND2: 7m
gAGE/UPC
PRN30 PRN30
'( L1 U ) { L1ru
sat
sat
Urusat
9
gAGE
PRN06 PRN06
IND1-IND2: 7m IND1-IND2: 7m
PRN30 PRN30
'( L1 U ) { L1ru
sat
sat
Urusat
'( L1 U ) { Lruj Uruj c G tru Truj I ruj O Zruj O Nruj bru Q L ruj
'( P1 U ) { Pruj Uruj c G tru Truj I ruj Kru Q P ruj

gAGE/UPC
Dif. Instrumental
10
gAGE
Pi j Ui j c G ti G t j Ti j Ii j Ki K j Q P ij
Lij Ui j c G ti G t j Ti j Ii j O Zi j O Ni j bi b j Q Li
j
where: U j 'r j
i U0ij 0ij 'ri 0ij 'r j
Single difference (x)ruj { '(x)ruj (x)uj (x)rj
Pruj Uruj c G tru Truj I ruj Kru H ruj

Lruj Uruj c G tru Truj I ruj O Zruj O Nruj bru Q j
L ru
0j u 0j r
where: U
j
U U j j uj j
'rr
ru u r
User
'ru rru r
Uruj U0 ruj 0uj 'ru 0 rj 'rr 0uj 'r j 0j r 'r j receiver Ref.
Station
gAGE/UPC
Baseline
U0 ruj 0uj 'rru 0 ruj 'rr 0 ruj 'r j rru ru rr
being: U0ruj { U0uj U0rj ; 'rru { 'ru 'rr ru rr

To Earth Centre 11
gAGE
Exercise:
Let be: Uru Uuj Urj

j
where U j U0ij 0ij 'ri 0ij 'r j (from Taylor expansion)

i
Show that the Single Differences are given by:
Uruj U0ruj 0uj 'rru 0ruj 'rr 0ruj 'r j
being: U0ruj { U0uj U0rj ; 'rru { 'ru 'rr

gAGE/UPC

12
gAGE
Uruj U0ruj 0uj 'rru 0 ruj 'rr 0ruj 'r j with 'rru { 'ru 'rr
Lets assume that:

eph
j
The satellite coordinates are known with an uncertainty 'r { eph

j j
The reference station coordinates are known with an

uncertainty 'rr { site (i.e. rr r0r site )
Thence, the user position can be computed from the 'rru

estimate, with an error site , as: r r 'r u 0u ru site
and where the 'rru estimate will be affected in site

turn by the ephemeris and sitting site errors as: User rru ru rr
receiver Ref.
Sta.
Uruj U0ruj 0uj 'rru 0 ruj site 0 ruj eph

j
ru rr
gAGE/UPC
Range error due to Range error due to

reference station Sat. coordinates
coordinates uncertainty uncertainty To Earth Centre

ru r0u 'ru r0u 'rru 'rr r0u@J. Sanz
'rru& J.M.Juan
site 13
eph
j
gAGE
and where the 'rru estimate will be affected in
turn by the ephemeris and sitting site errors as:
Uruj U0ruj 0uj 'rru 0 ruj site 0 ruj eph

j
site
Range error due to Range error due to User rru ru rr
reference station Sat. coordinates receiver Ref.
Sta.
coordinates uncertainty uncertainty
Thence, taking into account the relationships: ru rr

b rru
b b
0 ru
j
site d site ; 0 ru
j
eph
j
d eph
j
U u
j
U u
j To Earth Centre
and being Uuj 20000 km it follows that for a baseline b 20 km
b 1 The effect of 5 metres error in orbits or in site coordinates

|
gAGE/UPC
U u
j
1000 is less than 5 mm in range for the estimation of 'rru .
But, the user position estimate will be shifted by the error

in the site coordinates ru r0u 'rru site
14
gAGE
Exercise:
Demonstrate the following relationship: Satellite-j
b
0 ru
j
site d site
U
Hint:
0 ru
j
site j
0u 0 rj site rj
0j u 0j r b
site d
U U site uj
b rrurru rj
User Ref
receiver Sta
Note: the following approaches have been taken: rru ru rr
j j
j
0 r
0r 0r
U j
U j
U0jr U0ju U 0r
ru rr
j j
0uj
gAGE/UPC
0u
0u U j
0u Uj
b rru r0 ru u v d u v To Earth Centre

15
gAGE
Contents

1.1. Linear model
2.1. Introduction
gAGE/UPC

16
gAGE
Geographic decorrelation of ephemeris errors
Position from
broadcast
Satellite
ephemeris
location error
True position H
user
Uuser
Differential range error due to
satellite obit error
ref

U ref
Uuser user ref
GU
Uuser Uref
Uuser
A conservative bound:
b
GU H
U
User
gAGE/UPC

Reference Station 20 1
GU H H
20000 1000

17
gAGE
Satellite Range error
location error H from
user
CREU and
EBRE Uuser
Position
from
Satellite
broadcast
Differential range error from
location error
ephemeris between CREU and EBRE
True position H
user
Uuser
ref

U ref
U user user reff
GU
Uuser Uref
U user
gAGE/UPC
288 km of baseline
User
Reference Station

18
Range error
from
CREU and user

EBRE Uuser reff
user
GU
Uuser Uref
CREU-EBRE
CREU Differential positioning
gAGE/UPC
288 km of baseline

19
gAGE
H eph
U1 U2
U2
U1
Orbit Errors over the

hyperboloid will not
produce differential
range errors.
U1 U1 'UH
gAGE/UPC
U1 U2 U1 U2 constant GU
eph
0
U2 U 2 'UH eph

20
gAGE 'UH eph
H eph
U1 U2
U2
U1
Orbit Errors over the

hyperboloid will not
produce differential
range errors.
U1 U1 'UH
gAGE/UPC
U1 U2 U1 U2 constant GU
eph
0
U2 U 2 'UH eph

21
By Miguel Juan Zornoza

gAGE
Errors over the hyperboloid (i.e.
U B U A ctt ) will not produce
u differential range errors.
The highest error is given by
the vector u , orthogonal to

the hyperboloid and over the
plain containing the baseline
vector b and the LoS vector .
Note:
Being the baseline b much smaller than
the distance to the satellite, we can
assume that the LoS vectors from A and
a ( U B U A ) / 2 : hyperboloid semiaxis B receives are essentially identical to U.
That is, U # U # U
b / 2 : focal length B A
where a
1
b cos I
u
u b u b - b
T T
I b T b I b
2 T
Note: in this 3D problem I is NOT the elevation of ray.
Note: u sin I u
Differential range error GU produced by an

orbit error H parallel to vector u Note: being u a vector orthogonal
al
Let GH { H to the LoS , thence, H T u
Thence:
gAGE/UPC
GU { G ( U B U A ) 2G a b sin I b
wa wa wI wI GU T u T sin I u
2 GH 2
wH wI wH
GH b sin I
wH
GH U U
T I T
1 b Where: b = b b
b sin I GH
U
Note: A G
GH U GI U @ J. Sanz
is the baseline
& J.M. Juan vector
gAGE
ORBIT TEST :
Broadcast orbits
Along-track Error (PRN17)
PRN17:
Doy=077, Transm. time: 64818 sec
17 10 3 18 20 0 0.0 1.379540190101E-04 2.842170943040E-12 0.000000000000E+00

7.800000000000E+01-5.059375000000E+01 4.506973447820E-09-2.983492318682E+00
-9.257976353169E-05 5.277505260892E-03 8.186325430870E-06 5.153578153610E+03
4.176000000000E+05-5.401670932770E-08-4.040348681654E-01-7.636845111847E-08
9.603630515702E-01 2.215312500000E+02-2.547856603060E+00-7.964974630307E-09
-3.771585673111E-10 1.000000000000E+00 1.575000000000E+03 0.000000000000E+00
2.000000000000E+00 0.000000000000E+00-1.024454832077E-08 7.800000000000E+01
4.104180000000E+05 4.000000000000E+00
gAGE/UPC
diff EPH.dat.org EPHcuc_x0.dat --------------------------------------------------

< -2.579763531685E-06 5.277505260892E-03 8.186325430870E-06 5.153578153610E+03
> -9.257976353169E-05 5.277505260892E-03 8.186325430870E-06 5.153578153610E+03
---------------------------------------------------------------------------------

23
gAGE
Orbit error Range error

rov2

39.3 km Baseline: b=31.3km
rov1 31.3 km
11 km
15.2 km
gAGE/UPC
T I T
b
19.7 km
GU
rov3 U

24
gAGE
Exercise:
Justify that clock errors completely

cancel in differential positioning.
gAGE/UPC

25
gAGE
Contents

1.1. Linear model
2.1. Introduction
gAGE/UPC

26
gAGE
Error mitigation and short baseline concept
If the distance between the user and the reference station is "short enough",
so that the residual error ionospheric, tropospheric and ephemeris are small
compared to the typical errors due to receiver noise and multipath, it can be
assumed: j j
Tru j j
I ru 0; 0ru eph 0
Note that the previous definition of shortness is quite fussy
Working with smoothed code, a residual error of about 0.5 metres could be
tolerable, but for carrier based positioning it should be less than 1 cm to
allow the carrier ambiguity fixing.
The differential ephemeris error is at the level of few centimetres for

or
H j
baselines up to 100 Km (i.e. 5 cm assuming a large bound of eph 10 m ).
The typical spatial gradient of the ionosphere (STEC) is 1-2 mm/km (i.e.
gAGE/UPC
0.1-0.2 m in 100km), but it can be more than one order of magnitude higher
when the ionosphere is active.

27
gAGE
Error mitigation and short baseline concept
Note that the previous definition of shortness is quite fussy

The correlation radio of the troposphere is lower than for the ionosphere. At
10km of separation the residual error can be up to 0.1-0.2 m. Nevertheless,
90% of the tropospheric delay can be modelled and the remaining 10% can
be estimated together with the coordinates (for high precision applications).
For distances beyond a ten of kilometres or significant altitude difference it
would be preferable to correct for the tropospheric delay at the reference
station and user receiver.
Carrier-smoothed code:
Pseudorange code measurement errors due to receiver noise and multipath
can be reduced smoothing the code with carrier measurements.
Smoothed codes of 0.5m (RMS) can be obtained with 100 seconds smoothing.
On the other hand, the ionospheric error is substantially eliminated in
gAGE/UPC
differential mode and the filter can be allowed for lager time smoothing
windows.

28
gAGE
Contents

1.1. Linear model
2.1. Introduction
gAGE/UPC

29
gAGE Differential code based positioning

If the reference station coordinates are known at the Truj 0 ; I ruj 0
centimetre level and the distance between reference

station and user are not too large, we can assume 0 ru
j
eph
j
0
site 0 'rru '
'rru
Thence,
Pruj Uruj c G tru Truj I ruj Kru Q j
P ru Pruj Uruj c G tru Kru Q j
P ru
Uruj U0ruj 0uj 'rru 0ruj site, r 0ruj eph

j
Uruj U j
0 ru
0 uj 'ru
Note : for baselines up to 100 km Or, what is the same:

the range error of broadcast Pruj U 0 ruj 0 uj 'ru c G tru Kru Q P ruj
orbits is less than 10 cm
(assuming H ephj
10 m ).
The left hand side of previous equation can be spitted in two terms:
one associated to the reference station and the other to the user:
gAGE/UPC
Pruj U 0 ruj
Pu j U 0 uj Pr j U 0 rj
'rru 'ru 'rr 'ru site @ J. Sanz & J.M. Juan
Pruj U 0 ruj 0 uj 'ru c G tru Kru Q P ruj
The left hand side of previous equation can be spitted in two terms:
one associated to the reference station and the other to the user:
Pruj U 0 ruj
Pu j U 0 uj Pr j U 0 rj
The term Pr U 0 r is the error in range measured by the reference
j j
station, which can be broadcasted to the user as a differential correction:

PRC j U rj Pr j
0
The user applies this differential correction to remove/mitigate common

errors: P j U j PRC j j 'r c G t Q j
u 0u 0u u r
ru P ru
Where the receivers instrumental

me delay term K ru is included
in the differential clock c G tru
gAGE/UPC
For distances beyond a ten of kilometres, or significant altitude difference, it

would be preferable to correct for the tropospheric delay at the reference
station and user receiver. 31
gAGE
Range Differential Correction Calculation
Position Position

Range Uref
Pref Pseudoranges

PRC, RRC
Reference station
The reference station with known coordinates, computes pseudorange

and range-rate corrections: PRC= Uref, Pref , RRC= 'PRC/'t .
gAGE/UPC
measurements, Puser + (PRC + RRC (t-t0)), removing SIS errors and

DGNSS with code ranges: users within a hundred of kilometres can obtain
corrections.
32
errors: P j U j PRC j j 'r c G t Q j

u 0u 0u u r
ru P ru
rum
where the receivers instrumental delay term K ru is included in
the differential clock c G tru
The previous system for navigation equations is written in matrix notation as:

1
T
1
Pref 1 0u
2

T
Pref
2
0u
1 'ru

c G tru
n

Pref

T

n
1
0u
gAGE/UPC
where
Pref j { Pu j U 0 uj PRC j

33

Time synchronization issues:
For simplicity we have dropped any reference to measurement

epochs, but real-time implementations entail delays in data
transmission and the time update interval can be limited by
bandwidth restrictions.
Differential corrections vary slowly and its useful life can be up to
several minutes with S/A=off.
To reduce bandwidth, the reference station computes Pseudorange

Corrections (PRC) and Range-Rate Correction (RRC) for each
satellite in view, which are broadcast to every several seconds, up to
a minute interval with S/A=off.
The user computes the PRC at the measurement epoch as:

gAGE/UPC
PRC j (t ) PRC j (t0 ) RRC j (t t0 )

34
Data handling:
Reference station and user have to coordinate how the

measurements are to be processed:
Corrections must be identified with an Issue of Data (IOD) and
time-out must be considered.
Both receivers must use the same ephemeris orbits (which are
identified by the IODE).
If reference station uses a tropospheric model the same model
must be applied by the user.
If reference station uses the broadcast ionospheric model, the user
must do the same.
gAGE/UPC
Note: we have considered here only code measurements. The carrier based
positioning will be treated next, using double differences of measurements and
targeting the ambiguity fixing.
35
gAGE
Range Differential Correction Calculation
Broadcast Actual SV
SV Position Position

Range Uref Pref
Pseudoranges

PRC, RRC
Reference station
The reference station with known coordinates, computes pseudorange and

range-rate corrections: PRC= Uref Pref, RRC= 'PRC/'t .
gAGE/UPC
measurements, Puser + (PRC(t0) + RRC(t-t0)), removing SIS errors and

DGNSS with code ranges : users within a hundred of kilometres can obtain
corrections.36
gAGE
GODN
USN3 76 m
GODS
ftp://cddis.gsfc.nasa.gov/highrate/2013/
gAGE/UPC
1130752.3120 -4831349.1180 3994098.9450 gods

1130760.8760 -4831298.6880 3994155.1860 godn
1112162.1400 -4842853.6280 3985496.0840 usn3

37
gAGE GPS Standalone GPS Standalone USN3

GODN GODN
GODS
GODS
USN3
S/A=off
DGPS DGPS
GODN
GODN
GODS: reference
gAGE/UPC
USN3 station GODS: reference

USN3 station

38
gAGE GODS: reference PRC DGPS
station
GODN
GODS: reference
USN3 station
GODS: reference RRC

station DGPS
GODN
gAGE/UPC
GODS: reference
USN3 station

39
gAGE
Contents

1.1. Linear model
2.1. Introduction
gAGE/UPC

40
gAGE
research group of Astronomy and Geomatics Introduction: What Augmentation is?
To enhance the performance of the current GNSS

with additional information to:
Improve INTEGRITY via real-time monitoring
Improve ACCURACY via differential corrections
Improve AVAILABILITY and CONTINUITY
Satellite Based Augmentation Systems (SBAS)

E.g., WAAS, EGNOS, MSAS
Ground Based Augmentation Systems (GBAS)
E.g., LAAS
Aircraft Based Augmentation (ABAS)
gAGE/UPC
E.g., RAIM, Inertials, Baro Altimeter

41
gAGE
Why Augmentation Systems?

Current GPS/GLONASS Navigation Systems

cannot met the Requirements for All Phases
of Flight:
Accuracy
Integrity
Continuity
Availability
Marine and land users will also require

some sort of augmentation for improving
the GPS/ GLONASS performances.
gAGE/UPC

42
gAGE
research group of Astronomy and Geomatics WHY GNSS NEEDS AN AUGMENTATION ?
GPS Only Civil Aviation
PERFORMANCE CATEGORY I
Requirements
H. 13 m V. 22m ACCURACY (95%) H 16.0 m V 4.0 m
99% (RAIM) AVAILABILITY 99% to 99.990%
INTEGRITY 2.10-7/ approach

gAGE/UPC
?
Time to alarm 6 s
? CONTINUITY OF SERVICE 10-5 / approach

(10-6 / 15 s)

43
gAGE Accuracy: Difference between the measured position at any given

time to the actual or true position.
Even with S/A off a Vertical Accuracy < 4m 95% of time

GPS
GPS BBefore
cannotefbe
orguaranteed
e and
and After
Aftewith
r S/A
S/the
Aw was
as sswitched
witchGPS.
standalone ed o
off
ff
Colorado Springs, Colorado 2 May 2000
160
140 Horizontal Error (meters)
120 Vertical Error (meters)
100
Instantaneous Error (meters)
80
60
40
20
0
-20
-40
-60
-80 ANALYSIS NOTES
-100
-120 - Data taken from Overlook PAN Monitor Station,
equipped with Trimble SVeeSix Receiver
-140 - Single Frequency Civil Receiver
gAGE/UPC
-160 - Four Satellite Position Solution at Surveyed Benchmark

-180 - Data presented is raw, no smoothing or editing
-200
0 1 2 3 4 5 6 7 8 9 10
Time of Day (Hours UTC)

44
gAGE Integrity: Ability of a system to provide timely warnings to users or
research group of Astronomy and Geomatics to shut itself down when it should not be used for navigation.
Standalone GPS and GLONAS Integrity is Not Guaranteed
GPS/GLONASS Satellites:
Time to alarm is from minutes to hours
No indication of quality of service
Health Messages:
GPS up to 2 hours late
GLONASS up to 16 hours late
gAGE/UPC

45
gAGE
Continuity: Ability of a system to perform its function without
(unpredicted) interruptions during the intended operation.
Availability: Ability of a system to perform its function at initiation of

intended operation. System availability is the percentage of time
that accuracy, integrity and continuity requirements are met.
Availability and Continuity Must meet requirements
Continuity:
Less than 10-5 Chance of Aborting
a Procedure Once it is Initiated.
Availability:
gAGE/UPC
>99% for every phase of flight (SARPS).

46
gAGE
INTEGRITY
NSE
Confidence bound
gAGE/UPC
Less than 10-7 probability of true

Alert Limit error larger than confidence bound.
Time to alarm 6 s

47
gAGE
SBAS and GBAS Navigation Modes

SBAS
gAGE/UPC
GBAS

48
gAGE
Civil Aviation Signal-in-Space
Performance Requirements
Aviation Accuracy Accuracy Alert Alert Integrity Time Continuity Avail-

(H) 95% (V) 95% Limit (H) Limit (V) to alert ability
ENR 3.7 Km N/A 7400 m N/A 1-10-7/h 5 min. 1-10-4/h 0.99
(2.0 NM) 3700 m to to
1850 m 1-10-8/h 0.99999
TMA 0.74 Km N/A 1850 m N/A 1-10-7/h 15 s 1-10-4/h 0.999
(0.4 NM) to to
1-10-8/h 0.99999
NPA 220 m N/A N/A 1-10-7/h 10 s 1-10-4/h 0.99
(720 ft) 600 m to to
1-10-8/h 0.99999
APV-I 220 m 20 m 600 m 50 m 1-2x10-7 per 10 s 1-8x10-6 in 0.99
(720 ft) (66 ft) approach any 15 s to
0.99999
APV-II 16.0 m 8.0 m 40 m 20 m 1-2x10-7 per 6s 1-8x10-6 in 0.99
(52 ft) (26 ft) approach any 15 s to
0.99999
gAGE/UPC
CAT-I 16.0 m 6.0 - 4.0 m 40 m 15 -10 m 1-2x10-7 per 6s 1-8x10-6 in 0.99

(52 ft) (20 to 13 ft) approach any 15 s to
0.99999

49
gAGE
Contents

1.1. Linear model
2.1. Introduction
gAGE/UPC

50
gAGE
research group of Astronomy and Geomatics GBAS Concept
Most of the measurement errors are common:

clock, ephemeris, ionosphere and troposphere.
A common correction valid for any receiver within the LADGPS
area is generated and broadcast.
gAGE/UPC
The accuracy is limited by the spatial decorrelation of those

error sources (1m at 100Km).

51
gAGE GBAS Concept

The Ground Station (GS)
responsible for generating

and broadcasting carrier-
smoothed code differential
corrections and approach-
path information to user
aircrafts.
This system is used to
support aircraft operations
during approach and landing.
It is also responsible of detecting and alarming space-segment
and ground-segment failures.
The GS must insure that all ranging sources for which GBAS
corrections are broadcast are safe to use. If a failure occurs that
threatens user safety, the GS must detect and alert users (by not
gAGE/UPC
broadcasting corrections for the affected ranging source) within a

certain time-to-alert.
This is from [RD-5]

52
gAGE GBAS Components
The GBAS station involves

redundant Reference
Receivers and antennas,
redundant High Frequency
Data Broadcast equipment
linked to a single antenna.
The GS tracks, decodes, and monitors GPS satellite information and

generates differential corrections.
It also performs integrity checks on the generated corrections.
The correction message, along with suitable integrity parameters
gAGE/UPC
and approach path information, is then broadcast to airborne users

on a VHF channel, up to about 40km.
This is from [RD-5]

53
gAGE
Ground System Functional Flow Diagram
This is from [RD-5]
A variety of integrity monitor algorithms that are grouped into:

gAGE/UPC
Signal Quality Monitoring (SQM) Multiple Reference Consist Check (MRCC)

Data Quality Monitoring (DQM) VP-monitor
Measur. Quality Monitoring (MQM) Message Field Range Test (MFRT)

54
gAGE SQM: Targets satellite signals anomalies and local interference. Implements
tests for Correlation Peak Symmetry, Receiver Signal Power and Code-
Carrier Divergence.
DQM: Checks the validity of the GPS ephemeris and clock data for each
satellite that rises in view of the LGF and at each time new navigation
data messages are broadcast.
MQM: Confirms the consistency of the pseudorange and carrier-phase
measurements over the last few epochs to detect sudden step and any
other rapid errors.
MRCC: Examines the consistency of corrections for each satellite across all
reference receivers.
VP-monitor: Helps ensure a Gaussian distribution for the correction error
with zero mean and that the broadcast Vpr _ gnd overbounds the actual
errors in the broadcast differential corrections.
MFRT: Verifies that the computed averaged pseudorange corrections and
correction rates fit within the message field bounds. This is from [RD-5]
Executive Monitors (EXM-I and EXM-II) coordinate all previous monitors

gAGE/UPC
and combine failure flags.
Signal Quality Monitoring (SQM) Multiple Reference Consist Check (MRCC)

Data Quality Monitoring (DQM) VP-monitor
Measur. Quality Monitoring (MQM) Message Field Range Test (MFRT)
55
gAGE
gAGE/UPC

56
gAGE
Contents

1.1. Linear model
2.1. Introduction
gAGE/UPC

57
gAGE
SBAS Concept
The pseurorange error is split in its components.

Clock error
Ephemeris error
Ionospheric error
Local errors (troposphere, multipath, receiver noise)
gAGE/UPC
Uses a network of receivers to cover broad geographic area

58
gAGE
research group of Astronomy and Geomatics Error Mitigation
Error GBAS SBAS

component
Satellite clock Estimation and
Ephemeris Common Mode Removal each
error
Ionosphere Differencing component
Troposphere Fixed Model
Multipath and Carrier Smoothing by user

gAGE/UPC
Receiver Noise

59
gAGE
Three existing SBAS Systems

gAGE/UPC

60
gAGE
The European Geoestationary Navigation
Overlay SERVICE (EGNOS)
EGNOS is the European component of a Satellite

Based Augmentation to GPS.
EGNOS is being developed under the

responsibility of a tripartite group:
The European Space Agency (ESA)
The European Organization for the Safety of Air
Navigation (EUROCONTROL)
The Commission of the European Union.
gAGE/UPC

61
gAGE ECAC Area

(ECAC: European Civil Aviation Conference)
70
Artemis
IOR
AOR-E
60
50
Latitude ()
40
30
gAGE/UPC
20
-40 -30 -20 -10 0 10 20 30 40
Longitude ()

62
gAGE
research group of Astronomy and Geomatics EGNOS Architecture
GEO (x3)
GPS
RIMS
(x 37)
NLES
(x 6)
EWAN
gAGE/UPC
MCC 1 MCC 2 MCC 3 MCC 4 PACF ASQF

63
gAGE
gAGE/UPC

64
gAGE
EGNOS ground segment is composed of the following
stations/centres which are mainly distributed in Europe and are
interconnected between themselves through a land network.
37 RIMS (Ranging and Integrity Monitoring Stations) + seven

being deployed: receive the satellite signals and send this
information to the MCC centres.
4 MCC (Master Control Centres) receive the information from the

RIMS stations and generate correction messages to improve
satellite signal accuracy and information messages on the status
of the satellites (integrity). The MCC acts as the EGNOS system
'brain'.
6 NLES (Navigation Land Earth Stations): they receive the

correction messages from the CPFs for the upload of the data
stream to the geostationary satellites and the generation of the
GPS-like signal. This data is then transmitted to the European
gAGE/UPC
users via the geostationary Satellite
http://egnos-user-support.essp-sas.eu/egnos_ops/egnos_system/system_description/current_architecture

65
gAGE Master Control Center (MCC)

MCC is Subdivided into

CCF (Central Control Facility)
Monitoring and control EGNOS G/S
Mission Monitoring and archive
ATC I/F
CPF (Central Processing Facility)

Provides EGNOS WAD corrections
Ensures the Integrity of the EGNOS users
Utilises independent RIMS channels for checking of corrections
Real time software system developed to high software standards
4 MCCs are implemented in EGNOS

gAGE/UPC

66
gAGE
research group of Astronomy and Geomatics SBAS Differential Corrections and Integrity:
The RTCA/MOPS-DO 229C
gAGE/UPC
Do-229A, B, C, D

67
gAGE
Message Format
DIRECTION OF DATA FLOW FROM SATELLITE; MOST SIGNIFICANT BIT (MSB) TRANSMITTED FIRST
250 BITS - 1 SECOND

IODP (2 BITS) REPEAT FOR 12
REPEAT FOR 12 MORE SATELLITES UDREI MORE SATELLITES
PRCf
212-BIT DATA FIELD
13 12-BIT FAST CORRECTIONS 13 4-BIT UDREIs 24-BITS
PARITY
IODF (2 BITS)
6-BIT MESSAGE TYPE IDENTIFIER (= 2, 3, 4 & 5)
8-BIT PREAMBLE OF 24 BITS TOTAL IN 3 CONTIGUOUS BLOCKS
The corrections, even for

individual satellites are
distributed across several
250 bits individual messages.
One Message per second d
gAGE/UPC
All messages have identical format

68
gAGE SBAS Broadcast Messages (ICAO SARPS)
MSG 0 Don't use this SBAS signal for anything (for SBAS testing)
MSG 1 PRN Mask assignments, set up to 51 of 210 bits
MSG 2 to 5 Fast corrections
MSG 6 Integrity information
MSG 7 Fast correction degradation factor
MSG 8 Reserved for future messages
MSG 9 GEO navigation message (X, Y, Z, time, etc.)
MSG 10 Degradation Parameters
MSG 11 Reserved for future messages
MSG 12 SBAS Network Time/UTC offset parameters
MSG 13 to 16 Reserved for future messages
MSG 17 GEO satellite almanacs Many Message Types
MSG 18 Ionospheric grid point masks Coordinated Through
MSG 19 to 23 Reserved for future messages Issues Data (IOD)
MSG 24 Mixed fast corrections/long term
m ssatellite
atel
at ellilite
te e
error
rror
rr or corrections
cor
orre
rect
ctio
ions
ns
MSG 25 Long term satellite error corrections
MSG 26 Ionospheric delay corrections
gAGE/UPC
MSG 27 SBAS outside service volume degradation

MSG 28 to 61 Reserved for future messages
MSG 62 Internal Test Message
MSG 63 Null Message
69
gAGE
Issues of Data (IOD)
IODEk
IODE IODk Long-term No IOD: Fast
GPS repeat msg, Corrections
Corrections
s s
Ephemeris
Ephemeri small
(25) (2 - 5, 24)
changes
IODCk
IODP
GPS IODGk IODP
IODF
IODF j
IODP
Clock IODP
PRN Integrity
GLONASS
Mask Information n
DATA
(1) ( )
(6)
IODS IODP
IODP
Service Acceleration
Message Information
(7)
gAGE/UPC
(27)
Ionospheric
c Ionospheric
c
Mask Corrections
s
(18) IODI
IODI
(26)
70
gAGE
Message Time-Outs:
Users can operate even when missing Messages
Prevents Use of Very Old Data

Confidence Degrades When Data is Lost
IODF: Detect Missing Fast Corrections
1 second
System Latency
The Correction is tof: Time of applicability Last bit of message:

estimated by the (1st bit of message) tof+1sec
gAGE/UPC
master station
Correction time-Out

71
gAGE Associated Maximum En Route, Precision

Data Message Update Interval Terminal, NPA Approach
Types (seconds) Timeout (seconds) Timeout (seconds)
WAAS in Test Mode 0 6 N/A N/A

PRN Mask 1 60 None None
UDREI 2-6, 24 6 18 12
Fast Corrections 2-5, 24 60 (*) (*)
Long Term 24, 25 120 360 240
Corrections
GEO Nav. Data 9 120 360 240
Fast Correction 7 120 360 240
Degradation
Weighting Factors 8 120 240 240
Degradation 10 120 360 240
Parameters
Ionospheric Grid 18 300 None None
Mask
Ionospheric 26 300 600 600
gAGE/UPC
Corrections
UTC Timing Data 12 300 None None
Almanac Data 17 300 None None
(*) Fast Correction Time-Out intervals are given in MT7 [between 12 to 120 sec]
72
gAGE
research group of Astronomy and Geomatics PRN MASK (MT01)
Bit No 1 2 3 4 5 6 . 38 . 120 . 210
Value 0 1 0 1 1 0 1 1 0
PRN GPS GPS GPS GLONASS AORE
PRN 2 PRN 4 PRN 5 Slot 1 PRN 120
PRN mask 1 2 3 21 29
Number
Each MT01 contains its associated IODP
Assignment
Up to 51 satellites in 210 slots. PRN
Slot
Note: Each Correction set in 1-37 GPS/GPS Reserved
MT 2-5,5,6,7,24,25 its 38-61 GLONASS
characterized by its PRN-Mask

62-119 Future GNSS
gAGE/UPC
120-138 GEO/SBAS
number, between 1 to 51. 139-210 Future GNSS/GEO/SBAS/Pseudolites

73
gAGE
Example of message: Fast Corrections
(MT2-5,24)
Primarily Removed SA
Common to ALL users
Up to 13 Satellites Per Message
Pseudorange Correction /confidence Bound
Range Rate Formed by Differencing
UDRE degrades Over Time
Acceleration Term in MT 7
gAGE/UPC
Reset when new Message Received

74
gAGE Example of message: Fast Corrections
(MT2-5,24)
PRC (t
(t ) PRC n RRC n (t t n ) Y C1 PRC U * 't sat dt sat
PRC n PRC o
RRC n TGD IONO TROP
t n to
DIRECTION OF DATA FLOW FROM SATELLITE; MOST SIGNIFICANT BIT (MSB) TRANSMITTED FIRST
250 BITS - 1 SECOND

IODP (2 BITS) REPEAT FOR 12
REPEAT FOR 12 MORE SATELLITES UDREI MORE SATELLITES
PRCf
13 12-BIT FAST CORRECTIONS 13 4-BIT UDREIs 24-BITS

PARITY
IODF (2 BITS)
6-BIT MESSAGE TYPE IDENTIFIER (= 2, 3, 4 & 5)
8-BIT PREAMBLE OF 24 BITS TOTAL IN 3 CONTIGUOUS BLOCKS
gAGE/UPC
( RSSUDRE 0 >MT10@)
V 2i, flt V 2UDRE H 2 fc H 2rrc H 2ltc H 2er

75
gAGE Example of message: Ionospheric Corrections

(MT26)
Only Requiered for Precission Approach

Grid of Vertical Ionospheric Corrections
Users Select 3 o 4 IGPs that Surrounding IPP
5x5 or 10x10 for 55<Lat<55
Only 10x10 for 55<|Lat|<85
Circular regions for |Lat|>85
Vertical Correction and UIVE Interpoled to IPP

Both Converted to Slant by Obliquity Factor
gAGE/UPC
IGP: Ionospheric Grid Point

IPP: Ionospheric Pierce Point

76
gAGE
Global IGP Grid
W180 W140 W100 W60 W20 0 E20 E60 E100 E140
N85
N75
N65
N55
N50
S50
0 1 2 3 4 5 6 7 8
S55
S65
gAGE/UPC
S75
S85
Predefined Global IGP Grid

77
Fast Corrections
gAGE
Ephemeris +
+ Long Term Corrections MT1, MT2,5,24, MT6,
+ MT25, MT7, MT12,
Clocks UDRE MT9

+
Degradation Param.
MT10
IONO Corrections
+ MT18
IONO GIVE
+ MT26
Degradation Param.
y PRC UU** ''ttsat

C1 PRC sat
dt sat TGD IONO TROP
gAGE/UPC
V 2 V 2flt V UIRE
2
V air
2
V tropo
2

78
Navigation System Error
gAGE y = Gx
and Protection levels
= G T WG G T W y
-1
x
X >'N , 'E, 'U , cdt @
d N2 d NE d NV d NT
2

d NE dE d EV d ET
Px G T WG
1 2
d N2 d E2 d N2 d E2
d NE
2
2
2
d NV d EV dV dVT HPL 6.00
d d ET dVT d T2 2
NT
W PY1
VPL 5.33 dV
V 12 0

Py
0 2
VN V i2 V i2, flt V i2,UIRE V i2,air V i2,tropo

-KV-V V KV
gAGE/UPC
X ~ N (0,1)
P( X ! 5.33) 107

79
gAGE
Fast and Long-Term Correction Degradation
V

UDRE H fc H rrc H ltc H er , if
2
V 2 RSSUDRE 0 ( MT10)
i , flt
V 2UDRE H 2 fc H 2 rrc H 2ltc H 2er , if RSSUDRE 1 ( MT10)
t tu tlat
2
MT25
H fc a 2
MT2-6,24 tltc , v0 or v1
Cltc ,v 0 floor I ltc

t t
H ltc,v 0 IODF ,UDREi
ltc ,v 0 tu tof MT7
ai I fc ,i tlat
(when IODF z 3)
0 , if t0 t t0 I ltc _ v1
H ltc,v1
Clts _ lsb Cltc _ v1 max ^0, t0 t , t t0 I ltc _ v1` , otherwise
MT10
0 Neither fast nor long term corrections
Brrc , Cltc _ lsb , Cltc _ v1,
H er
have time out for precision approach
Cer Otherwise Iltc _ v1, Cltc _ v0 , Iltc _ v0 ,

IODFcurrent , IODFprevious z3
gAGE/UPC
IODFcurrent , IODFprevious z3
0 , ( IODF IODFprevious ) mod 2 1 Cer , RSSUDRE ,

H rrc
current

a I fc Brrc
4 't t t of , ( IODF current
IODFprevious ) mod 2 z1
Ciono _ ramp , Ciono _ step ,
Iiono , RSSiono
80
gAGE
Degradation of Ionospheric Corrections
1

R cos E 2
V UIRE Fpp2 V UIVE
2
2 2 1 e
Fpp
R h
e I
W x , y pp V n2,ionogrid , N MT10
N
V 2
UIVE n pp 4 or 3
n 1 Brrc , Cltc _ lsb , Cltc _ v1 ,
I ltc _ v1 , Cltc _ v 0 , I ltc _ v 0 ,
V GIVE H iono
2
V 2ionogrid
, if RSSiono 0 ( MT10)
2 Cer , RSSUDRE ,
V GIVE H iono
2
, if RSSiono 1 ( MT10)
Ciono _ ramp , Ciono _ step ,
I iono , RSSiono
Hiono t t

Ciono _ step floor I iono Ciono _ ramp t tiono MT26
gAGE/UPC
iono
tiono,GIVEi

81
gAGE
Users know the receiver-satellites geometry and can compute bounds on
the horizontal and vertical position errors.
These bounds are called Protection Levels (HPL and VPL). They provide
good confidence (10-7/hour probability) that the true position is within a

bubble around the computed position.
Protection
Level
True
Error
-KV-V V KV Tail area Probability P(VPE>VPL) < 10-7 /sample
N
V i2 V i2, flt V i2,UIRE V i2,air V i2,tropo
s V
gAGE/UPC
2 2
VPL KV Vi i
i 1
GEOMETRY

82
gAGE
EGNOS
Availability
Integrity
gAGE/UPC
Accuracy

DGNSS implementations: WADGNSS (SBAS)
83
gAGE
EGNOS
Availability
Integrity
Continuity
gAGE/UPC
Accuracy

84
gAGE
References

[RD-5] Gang Xie, Optimal on-airport monitoring of the integrity of GPS-based
landing systems, PhD Dissertation, 2004.
gAGE/UPC

85
gAGE
EGNOS Safety of Life Service Definition Document (Ref : EGN-SDD SoL, V1.0).
European Comission.
http://www.essp-sas.eu/downloads/vubjj/egnos_sol_sdd_in_force.pdf
12 GPS Standard Positioning Service Performance Standard [RD-3].

13 This is the typical range of ionospheric residual errors after application of the baseline Klobuchar
model broadcast by GPS for mid-latitude regions.
gAGE/UPC
SREW: Satellite Residual Error for the Worst user location.

UIVD: User Ionospheric Vertical Delay.
UERE: User Equivalent Range Error

86
gAGE
Lecture 7
Carrier-based Differential
Positioning.
Ambiguity Resolution
Techniques
gAGE/UPC

1
gAGE

gAGE/UPC
22 Jan 2015

2
gAGE
Contents
1. Linear model for DGNSS: Double Differences
1.1. Differential Code and carrier based positioning
1.2. Precise relative Positioning
1.3. The Role of Geometric Diversity
2. Ambiguity resolution Techniques
2.1. Resolving ambiguities one at a time
- Single-frequency measurements
- Dual-frequency measurements
- Three-frequency measurements
2.2 . Resolving ambiguities as a set: Search techniques
- Least-Squares Ambiguity Search Technique
gAGE/UPC
- LAMBDA Method

3
gAGE
Single difference (x) ruj { '(x) ruj (x)uj (x) rj
Pruj U ruj c G tru Truj I ruj K ru Q P

j
ru
Lruj U ruj c G tru Truj I ruj O Zruj O N ruj bru Q j

L ru
Double difference
(x) rujk { '(x) rujk (x) ru
k
(x) ruj (x)uk (x) rk (x)uj (x) rj
Pruk U ruk c G tru Truk I ruk K ru Q P

k
ru

j
ru
User Reference
P jk
ru U T I Q
jk
ru
jk
ru
jk
ru
jk
P ru
Station
Now are cancelled:

gAGE/UPC
Receiver clock
The same for carrier : Receiver code instrumental delays
Lrujk U rujk Truj I rujk O Zrujk O N rujk Q jk
L ru
Receiver carrier instrumental
delays
Carrier ambiguities N are integer
4
gAGE
research group of Astronomy and Geomatics (corrected by geometric range!!)
PRN06
PRN06
PLAN-GARR: 15km
PRN30
'( L1 U ) { L1ru
sat
U rusat PRN30 ' ( P1 U ) { L1ru
sat
U rusat
'( L1 U ) { Lruj U ruj c G tru Truj I ruj O Zruj O N ruj bru Q L ruj
'( P1 U ) { Pruj U ruj c G tru Truj I ruj K ru Q P ruj

Dif. Instrumental
gAGE/UPC

5
gAGE
Double-Difference of measurements
PLAN-GARR: 15km PLAN-GARR: 15km

'L1 'U { L1ru

sat
U rusat 'P1 'U { P1ru
sat
U rusat
Dif. wind-up: negligible
'( L1 U ) { Lrujk U rujk Trujk I rujk O Zrujk O N rujk Q L rujk
'( P1 U ) { Prujk U rujk Trujk I rujk Q P rujk Carrier ambiguities:

gAGE/UPC
constant
Dif. Tropo. and Iono. :
Small variations
6
gAGE
Single difference (x) ruj { '(x) ruj (x)uj (x) rj

j
ru
Lruj U ruj c G tru Truj I ruj O Zruj O N ruj bru Q j

L ru
where:
U ruj U0 ruj 0uj 'Uru 0 ruj site 0 ruj eph
j
Double difference
k
Prujk U rujk Trujk I rujk Q P

jk
ru

L ru
User Reference
Station
where: U rujk U0 rujk 0ujk 'Uru 0 rujk site 0kru eph

k
0 ruj eph
j
gAGE/UPC
being: U0 rujk { U0ujk U0 rjk ; 'rru { 'ru 'rr

7
gAGE
Exercise:
Consider the Single Differences of geometric range:

U ruj U0 ruj 0uj 'Uru 0 ruj site 0 ruj eph
j
where U ru Uuj U rj
j
Show that the Double Differences are given by:

U rujk U0 rujk 0ujk 'Uru 0 rujk site 0kru eph
k
0 ruj eph
j
being: U0 rujk { U0ujk U0 rjk ; 'rru { 'ru 'rr

gAGE/UPC

8
gAGE Linear model for Differential Positioning
k

jk
ru

L ru
where: U rujk U jk
0 ujk 'Uru 0 jkru site 0kru eph
k
0 jru eph
j
0 ru
For short baselines (e.g. up to 10 km) and if the reference 'rru { 'ru 'rr
station coordinates are accurately known, we can assume:
Trujk 0 ; I rujk 0 ; Zrujk 0 Note for baselines up to 10 km
the range error of broadcast
0 ru
j
eph
j
0 orbits is less than 1cm
site 0 'Uru 'Uu (assuming H ephj
10 m ).
With these simplifications, we have:

gAGE/UPC
Prujk U 0 rujk 0 ujk 'Uu Q P rujk

Lrujk U 0 rujk 0 ujk 'Uu O N rujk Q L rujk
User Reference
Station

Remark: Pru U 0 ru
jk jk

Pu jk U 0 ujk Pr jk U 0 rjk @ J. Sanz & J.M. Juan
9
gAGE
Differential code and carrier positioning
As with the SD, the left hand side of previous equations can be spitted
in two terms: one associated to the reference station and the other to
the user. Then, the differential corrections can be computed for code
and carrier as: jk jk
PRCP jk jk jk
U jk
0r
Pr ; PRCL U 0r
Lr
errors:
Pu jk U 0 ujk PRCP jk 0 ujk 'Uu Q P rujk
Lujk U 0 ujk PRCL jk 0 ujk 'Uu O N rujk Q L rujk
Where the carrier ambiguities N are integer numbers and

must be estimated together with the user solution.
For larger distances, the atmospheric propagation effects (troposphere,

gAGE/UPC
ionosphere) must be removed with accurate modelling. Wide area users

will require also orbit corrections.

Remark: Pru U 0 ru
jk jk

Pu jk U 0 ujk Pr jk U 0 rjk @ J. Sanz & J.M. Juan
10
gAGE
Differential code and carrier positioning
The user applies this differential correction to remove/mitigate common errors
Pu jk U 0 ujk PRCP jk 0 ujk 'Uu Q P rujk

where 0 ujk { 0ku 0 uj
Lujk U 0 ujk PRCL jk 0 ujk 'Uu O N rujk Q L rujk
Pref P R ,1 0u

1 R T
0u
0 0

'ru
R ,1
T
1

R
1 0
Pref L 0u 0u
O N ru
R ,1

R , n 1
Pref P
T
n 1 R
Pref R ,n 1 0u 0u 0 0
O N ru
R , n 1

L T
n 1
0 u 0 u 0 1
R

gAGE/UPC
where Pref R , k { P R , k U R , k PRC R , k 'ru

P u 0u P
Carrier
Pref L R ,k
{L R ,k
u U R ,k
0u
PRCL R ,k
ambiguities User Reference
Station

11
gAGE
Differential positioning
Position Position

Range Uref
Pref Pseudoranges

PRC
Reference station
The reference station with known coordinates, computes differential

corrections: PRCP jk U rjk Pr jk ; PRCL jk U rjk Lrjk
0 0
gAGE/UPC
The user receiver applies these corrections to its own measurements to

remove SIS errors and improve the positioning accuracy.

12
gAGE Correlations among the DD Measurements
We assume uncorrelated measurements (both code and carrier). Then, the
covariance matrix is diagonal:
PP V P2 I PL V L2 I where, we can assume: V P 50cm ; V L 5mm
Let X represent the code P or the Carrier L measurement.
The single difference (SD) and its covariance matrix can be computed as:
X uk
X ruk
1 1 0 0 X rk
Thence, if the measurements are
j 0 0 1 1 j ; uncorrelated, so are they in single
X ru Xu
j differences, but the noise is twice!
X r
V X2 0 0 0 1 0

1 1 0 0 0 V X2 1 0
gAGE/UPC
0 0
PX SD 2V X2 I
0 0 1 1 0 V 0 1
2
0 0

X
0 0 0 V X2 0 1

13
gAGE Correlations among the DD Measurements

Now, the double difference (DD) and its covariance matrix can be
computed as:
X ruk
X jk
ru 1 1 0 j
jl 0 1 1 X ru ;
X ru l
X ru
2V X2 0 0 1 0
1 1 0 2 1
PX DD 0 1 1 0 2V X2 0 1 1 2V X2
1 2
0 0 2V X2 0 1
Thence, even if the original measurements are uncorrelated, the

double differences are correlated.
gAGE/UPC
Note: The removal of the relative UserReference-station common bias

(e.g. relative receiver clock) in DD is done at the expense of one
observation and the introduction of a correlation between measurements.

14
gAGE
x x x
x { k
x R
x
' { rov ref ' { ' k
' R

Receiver errors common for all satellites do not affect positioning (as
they are assimilated in the receiver clock estimate). Thence:
gAGE/UPC
- Only residual errors in single differences between sat. affect absolute posit.
- Only residual errors in double differences between sat. and receivers affect
relative positioning.
15
gAGE
x x x
x { k
x R
x
' { rov ref ' { ' k
' R

When comparing SD and DD one might suggest that in the DD

formulation there is even further error reduction, positively influencing
the results in positioning. This is however not true, since in the SD
gAGE/UPC
case the mean value of unmodelled effects is absorbed by the receiver

clock. If the DD correlations are taken into account, the positioning
results in both cases are the same. However the DD formulation has
the
Master advantage
of Science in GNSS that it allows the direct estimation of the
@ J.ambiguities.
Sanz & J.M. Juan
16
gAGE
Contents
gAGE/UPC
- LAMBDA Method

17
gAGE Relative Positioning

The following relationship can be obtained

from the figure, where we assume that the Satellite-j
baseline is shorter than the distance to Uuj

the satellite by orders of magnitude: U rj
U ruj Uuj U rj uj Uru
uj
Applying the same scheme to a rru ru rr Reference
second satellite k
Station
U ruk Uuk U rk uk Uru

UU ruj
j
ru Uuj U
Uu rj U r
j j
Thence, the double differences of ru rr

ranges are: To Earth Centre
gAGE/UPC
U rujk U ruk U ruj uk uj Uru ujk Uru

18
gAGE Relative Positioning Satellite-j
U u
j
U rj
Thence, the double differences of ranges are: uj

rru ru rr Reference
Station
U rujk U ruk U ruj uk uj Uru ujk Uru U ruj Uuj U rj
As commented before, for short baselines (e.g. Trujk 0 ; I rujk 0

less than 10km), we can assume that ephemeris
Zrujk 0
and propagation errors cancel, thence:
jk
ru
Prujk U rujk Q P
jk
ru

L ru
Lrujk U rujk O N rujk Q L
jk
ru
Note that these equations allows

Prujk ujk Uru Q P rujk a direct estimation of the
gAGE/UPC
baseline, without needing an

Lrujk ujk Uru O N rujk Q L rujk accurate knowledge of the
reference station coordinates.
Master of Science in GNSS Note: but, at least the approximate coordinates of @ J. Sanz & J.M. Juan
reference station or user are needed to compute U 19

Prujk ujk Uru Q P rujk

where u jk { uk u j
Lrujk ujk Uru O N rujk Q L rujk
1 R T 0 0 Baseline
P R ,1
u u

vector
r ,u
R ,1 rru
Lr ,u
u
1

u
R T
1 0 R ,1
O N ru

Satellite-j

PrR,u, n 1 n 1 R T 0 0 U u
j
R , n 1 u u
O N R , n 1
ru U rj
Lr ,u

u
n 1

u
R T
0 1 uj
rrurru rruurrrr
gAGE/UPC
Reference
Station
DD Code and
Carrier
Carrier
ambiguities ru rr
measurements U ruj Uuj U rj

20
gAGE

'L1 { L1ru
sat
'P1 { P1ru
sat
Satellite-j
'L1 { L jk
ru U jk
ru T jk
ru I jk
ru OZ O N
jk
ru
jk
ru Q jk
L ru
Uuj
U rj
'P1 { Prujk U rujk Trujk I rujk Q P rujk uj

rru ru rr Reference
gAGE/UPC
Station
Variation of the Baseline U rujk ujk Uru U ruj Uuj U rj
projection over the unit ru rr
Line-Of-Sight vector
21
gAGE

'L1 { L1ru
sat
'P1 { P1ru
sat
IND1-IND2: 7 m IND1-IND2: 7m
gAGE/UPC
'L1 { L1ru
sat
'P1 { P1ru
sat

22
In this approach, the reference station broadcast its time-
tagged code and carrier measurements, instead of the

computed differential corrections.
Thence, the user can form the double differences of its own
measurements with those of the reference receiver, satellite by
satellite, and estimate its position relative to the reference receiver.
Notice that, the baseline can be estimated without needing an accurate

knowledge of reference the station coordinates. Of course, the
knowledge of the reference station coordinates would allow the user
to compute its absolute coordinates.
gAGE/UPC

23

Time synchronization issues:
There is an important and subtitle difference between the previous

approach of relative positioning (which does not need to know the
reference station coordinates) and the differential positioning
approach based on the knowledge of the reference station
coordinates.
The differential corrections wary slowly, and its useful life can be up
to several minutes with S/A=off.
But, the measurements change much faster. The range rate can be
up to 800m/s and, thence, a synchronizer error of 1millisecond can
lead up to more than 1/2 meter of error.
As commented before, real-time implementation entails also
latencies, that can be up to 2 seconds, thence, a extrapolation
gAGE/UPC
technique must be applied to the measurements to reduce error due

to latency and epoch mismatch (to <1cm if ambiguities are intended
to be fixed).
24
gAGE
research group of Astronomy and Geomatics PRC L1
RRC dL1
dtrec ~ 660 m/s

gAGE/UPC
RRC ~ 2cm/s
d U up to ~800 m/s

25
gAGE L1 L1
1 ms jump
of receiver
clock adjust
dtrec | 300 km/450 s 667 m/s
dL1 dL1
300Km=1ms
dL1 | d U dtrec
dtrec ~ 660 m/s

dL1 | d U dtrec
gAGE/UPC
d U up to ~800 m/s

26
COMMENTS
gAGE RRC= 'PRC/'t
Real-Time implementation entails delays in data
transmission, which can reach up to 1 or 2 s. RRC
Differential corrections vary slowly and its

useful life is of several minutes (S/A=off)
But, the measurements change much faster:
The range rate dU/dt can be up to 800m/s RRC ~ 1 cm/s
and, therefore, the range can change by
more than half a meter in 1 millisecond.
Moreover the receiver clock offset can be
up to 1 millisecond (depending on the
dL1 | d U dtrec
receiver configuration).
Thence, the reference station dL1
300Km=1ms
measurements must be :
Synchronized to reduce station clock
mismatch: station clock can be estimated
to within 1Ps H dtsta 1mm
gAGE/UPC
dtrec ~ 660 m/s

Extrapolated to reduce error due to
d U up to ~800 m/s
latency: carrier can be extrapolated with
error < 1cm.

Exercise:
Demonstrate the following relationship

between the baseline and the differential uj
range [*]:
2uj Uru e j
U j
U U
j j
j rru rru / 2
ru u r
u uj Uru ru rr
rru
User Reference Station

Comments:
The previous expression can be written as:
2uj Uru uj Uru / 2
U U j j
Z e rru
j j
with Z {
j
e j
u r
uj uj Uru uj Uru / 2
0 in Z and e leads to the approximate expression previously found.
j j
Taking rru
gAGE/UPC
Z j and e j depend on the baseline rru , which is the vector to estimate.

Nevertheless, it is not very sensitive to changes in such baseline and can be computed
iteratively, computing the navigation solution starting from rru 0 .
Master of Science
[*] This in GNSS
result is from [RD-7] @ J. Sanz & J.M. Juan
28
Solution
Consider the following relations:

uj
rru
j
r
j
u
uj Uru / 2 rj uj e j
e j rru / 2
uj Uru / 2 2uj Uru
rru ru rr
2
2 2
j
u Uru H j Uru rj uj User Reference Station
j
r uj j
r uj
U r
j

Uuj uj Uru uj
Then:
Uuj U rj Z j e j rru
gAGE/UPC
2uj Uru 2uj Uru

with: Z j { Z j e j
uj uj Uru uj uj Uru

29
gAGE
Contents

gAGE/UPC
- LAMBDA Method

30
gAGE
The Role of Geometric Diversity: Triple differences
Let us consider again the problem of relative positioning for short baselines.
We have previously found the following equation for DD carrier

measurements, assuming short baselines (e.g. < 10km)
This is from [RD-3]
L jk
ru Uru O N
k
u
j
u
jk
ru Q jk
L ru
As the ambiguities are constant along continuous carrier phase arcs, an

option could be to take differences on time. Thence, if the user and
reference receiver are stationary we can write the triple differences as:
G 2 G 1 T
G L 12
ru u u

G uk G uj Uru GQ L rujk
T
G L G G
13 3 1
G Lrujk ru
u u rru

1K
G Lru G K G 1
gAGE/UPC
T
where: u u
G Lrujk { Lrujk (t2 ) Lrujk (t1 ) For simplicity, we assign (j=1) to the
reference satellite
31
gAGE
The Role of Geometric Diversity: Triple differences
This is from [RD-3]
G 2 G 1 T
G L 12
ru u u

1 T
G uk G uj Uru GQ L rujk G L G u G u rru
13 3
G Lrujk ru

1K
G Lru G G
K 1 T
u u
Now, we have a clean equations system involving only the

baseline vector to estimate. But the geometry is very weak (the
associated DOP will be large number) and the position estimates
will be in general worse than those from double differences.
gAGE/UPC

32
gAGE Estimation of position and change in position:
the role of Geometric Diversity This is
from [RD-3]
Let us now consider a simple model for the estimation of the relative
position vector from SD carrier measurements, assuming short baselines
(e.g. <10km):
Lruj U ruj c G tru O N ruj bru Q j

L ru Lruj uj Uru d ru O N ruj Q L ruj
U ruj Uuj U rj uj Uru where d ru { c G tru bru
1 T 1 Satellite-j
L 1
ru u
O N ru
1
U j
2 T r
u
u
L
2
1 ru O N ru2 U rj

ru
d ru uj
K K
gAGE/UPC
ru rr
O N ru
rru
Lru K T 1
Reference
Station
U ruj Uuj U rj
ru rr
To Earth Centre 33
gAGE Estimation of position and change in position:

the role of Geometric Diversity This is
from [RD-3]
Previous system can be arranged as:
1 T 1 Considering now differences between two
L1ru u
O N ru
1

2 2 T r 2 epochs t0 and t1 , and assuming no cycle-slips:
Lru u 1 ru O N ru
d ru

r (t ) r (t )

K K L ru (t1 ) L ru (t0 ) G (t1 ) ru 1 G (t0 ) ru 0
Lru K T O N ru d ru (t1 ) d ru (t0 )
u 1
L ru G G rru (t1 ) rru (t1 ) rru (t0 ) G (t1 ) G (t1 )

1 (t ) 1 (t ) T 0
u 1 u 0

G r (t ) r (t ) 1
u (t1 ) u (t0 )
T
G (t1 ) ru 1 G (t1 ) G (t0 ) ru 0
1
0
L ru (t1 ) L ru (t0 ) G (t1 ) G (t0 )
G d ru (t1 )

d ru (t0 )

1 (t ) 1 (t ) T 0
u 1 u 0
Estimation of changes Estimation of absolute value

in baseline vector and of baseline vector is tied to
clock bias is tied to the change in geometry
gAGE/UPC
d ru (t0 ) cannot be
the geometry matrix matrix at time t1 . This would estimated at all!
at time t1 . This can be poor determined if such
be well determined. change is not significant.
34
gAGE
Contents
- Least-Squares Ambiguity Search Technique.
gAGE/UPC
- LAMBDA Method.

35
gAGE Ambiguity resolution Techniques

As a driven problem to study the ambiguity fixing, we will consider the
problem of differential positioning in DD for short baselines (e.g. < 10 km).

In general we will consider that we have Code and Carrier measurements in
different frequencies (q=1,2), i.e. P1, P2, L1, L2
Short baseline
P jk
q , ru U T I
jk
ru
jk
ru
jk
q , ru Q jk
Pq , ru ; q 1, 2 Truj 0 Pq jk,ru U rujk Q Pq
jk
ru ; q 1, 2
I qjk,ru 0
Ljk
q , ru U T I
jk j jk
q , ru Oq Z O N
jk jk
q , ru Q jk
Lq , ru Lqjk,ru U rujk Oq N qjk,ru Q jk
Zrujk
ru ru ru ru
0 Lq
To simplify notation, when different frequencies are

considered, we will remove the subscript ru.
K sat. in view K SD
K(K-1) DD, Pqjk U jk Q Pjk ; q 1, 2
but only K-1 DDs are
q
linearly independent Lqjk U jk Oq N qjk Q Ljk q
We assume the following measurement errors:

gAGE/UPC
V P | 0.5 m V P | 1m Take the highest elevation as the

jk
reference satellite to minimize

q q
V L | 0.5cm
q
V L | 1cm
jk
measurement error.
q
AsMaster
commented before, the ambiguity terms are integer numbers, and@we
of Science in GNSS can take benefit
J. Sanz & J.M. Juan
of this property to fix such ambiguities applying integer ambiguity resolution techniques.36
gAGE
Contents
gAGE/UPC
- LAMBDA Method.

37
gAGE Resolving ambiguities one at a Time

A simple trial would be (for instance using L1 and P1):
P1 jk U jk Q Pjk L1jk P1 jk
1 o L P1
jk
1
jk
O1 N Q o
1
jk jk
N 1jk
Ljk
U O1 N Q
jk jk jk
P1
O1 roundoff
1 1 L1
O1 20 cm
V P | 1m 1
1
jk
V N VP | 5
O1
jk jk
V L | 1cm
jk
1 1
1
N 1
Fail Too much error (5 wavelengths)!
N 1/ 2 Note that, assuming a Gaussian
Good N
N distribution of errors, V N jk 1/ 2
guarantee only the 68% of success
1
N 1/ 2
N 1 As the ambiguity is constant (between
cycle-slips), we would try to reduce
gAGE/UPC
uncertainty by averaging the estimate on

time, but we will need 100 epochs to
Similar results with L2, P2 measurements
reduce noise up to (but measurement
errors are highly correlated on time!)
38
gAGE
gAGE/UPC

39
gAGE
gAGE/UPC

40
gAGE
Contents
gAGE/UPC
- LAMBDA Method.

41
gAGE Resolving ambiguities one at a Time

Dual frequency measurements: wide-laning with the Melbourne-Wbbena
combination
NW N1 N 2
P1 jk
U Q jk jk
f1 P1 jk f 2 P2jk OW
c
86.2 cm
P1
PNjk U jk Q Pjk f1 f 2
P2 jk
U Q jk jk
P2
f1 f 2 N
f1 L1jk f 2 L2jk V P | V P / 2 | 71cm

U jk OW NWjk Q Ljk
jk jk
U O1 N Q LWjk
jk jk jk jk N 1
L1 1 L1
f1 f 2 W
V L | 6 V L | 6 cm
jk jk
1
U O2 N Q
W
jk jk jk jk
L 2 2 L2
p
L PNjk
jk
OW NWjk Q Pjk o LWjk PNjk
W N Wjk
OW
N
roundoff
Fixing N1 (after fixing NW)
1 71cm
V N | VP 0.8
L L
jk
1
jk
2 O1 N O2 N Q
1
jk
2
jk jk
L1 L2
jk
W OW N
jk
86.2cm
O1 O2 N1jk O2 NWjk Q Ljk L 1 2 Now, with uncorrelated
O1 19.0 cm
L1jk L2jk O2 N Wjk measurements from 10
gAGE/UPC
O2 24.4 cm N 1jk epochs will reduce noise

O2 O1 5.4 cm O1 O2 roundoff up to about .
V L | 1cm
jk
1 1.4cm
1
V N | 2V jk 1/ 4
N jk N jk N jk 1
jk
O1 O2 L1 5.4cm
2 of Science
Master 1 in GNSSW @ J. Sanz & J.M. Juan
42
gAGE
gAGE/UPC

43
gAGE
Once the integer ambiguities are known, the carrier phase measurements
become unambiguous pseudoranges, accurate at the centimetre level (in DD),
gAGE/UPC
or better.
Thence, the estimation of the relative position vector is straightforward
following the same approach as with pseudoranges.
44
gAGE
research group of Astronomy and Geomatics Exercises:
1) Consider the wide-lane combination of carrier phase measurements

f1 L1 f 2 L2
LW , where LW is given in length units (i.e. Li= Oi Ii ).
f1 f 2
c
Show that the corresponding wavelength is: OW
f1 f 2
Hint:
LW= OW IW ; IW = I1 I2
2) Assuming L1, L2 uncorrelated measurements with equal noise VL,

show that: 2
J 12 1 f1
VL V L ; J 12
J 12 1 f2
gAGE/UPC

45
gAGE
Contents

gAGE/UPC
- LAMBDA Method.

46
gAGE Three Frequency measurements:
We still consider the above problem of relative positioning in DD
for short baselines (e.g. < 10 km) Ionosphere, troposphere and wind-
up differential errors cancel.
GPS Frequency Wavelengths Combinations With three frequency systems,

L1 154 x 10.23 MHz O =0.190 m O O =0.054 m having two close frequencies it
is possible to generate an
L2 120 x 10.23 MHz O =0.244 m OW = 0.862 m
extra-wide-lane signal to
L5 115 x 10.23 MHz O =0.255 m OEW =5.861 m enable the single epoch
ambiguity fixing.
We drop here the superscript (jk) for simplicity
c c c
Oi ; OW ; OEW
fi f1 f 2 f 2 f5
Li U Oi N i Q L ; i 1, 2,5
i
J 12 1 f1 / f 2
2
VL V 5, 7 cm
J 12 (77 / 60) 2
f1 L1 f 2 L2 J 12 1 L
U OW NW Q L
1
f 2 / f5
W
2
LW
f1 f 2 W J 25 (24 / 23) 2
J 25 1
f 2 L2 f5 L5 VL V 33,3 cm
LEW U OEW N EW Q L EW
J 25 1 L 1
NW N1 N 2
f 2 f5
gAGE/UPC
EW
f1 P1 f 2 P2
J 12 1 N EW N 2 N5
VP V 0, 712 m
PN U Q P N
J 12 1 P 1
f1 f 2 N
J 25 1 Exercise:
f 2 P2 f5 P5 VP V 0, 707 m
PEN U Q PEN J 25 1 P Justify the previous
f 2 fin5 GNSS
EN 1
Master of Science
expressions for V 47
We still consider the above problem of relative positioning in DD

gAGE for short baselines (e.g. < 10 km) Ionosphere, troposphere and wind-
up differential errors cancel .
L1 U O1 N1 Q L V L | V L | 1cm
GPS Frequency Wavelengths Combinations
1 1 2
O =0.190 m O O =0.054 m

L2 U O2 N 2 Q L V P | V P | 1m
L1 154 x 10.23 MHz 2 1 2
L2 120 x 10.23 MHz O =0.244 m OW = 0.862 m

LW U OW NW Q L PN U Q P N
L5 115 x 10.23 MHz O =0.255 m OEW =5.861 m
W
PEN U Q P EN
NW N1 N 2 ; N EW N 2 N5
f1 / f 2
2
LEW PEN 1 0.71m J 12 (77 / 60) 2
N EW V N | VP 0.12
OEW f 2 / f5
2
5.861m J 25
OEW
EN
(24 / 23) 2
roundoff
EW
J 12 1
VP V 0, 71m
N
J 12 1 P 1
OEW N EW LEW LW 1 33.3cm J 25 1

N W V N | VL 0.39 VP V 0, 71m
OW roundoff W OW EW
86.2 cm EN
J 25 1 P 1
J 12 1
gAGE/UPC
VL V 5, 7 cm
L1 L2 O2 N W 1 1.4cm
W
J 12 1 L 1
N 1 V N | 2V 1/ 4
O1 O2 5.4cm J 25 1
O1 O2 1 L1
roundoff VL V 33,3 cm
EW
J 25 1 L 1

48
Exercise:
gAGE
Repeat the previous study for the Galileo signals E1, E5b and E5a
V L | V L | 1cm
Galileo Frequency Wavelengths Combinations
L1 U O1 N1 Q L 1 2
E1 154 x 10.23 MHz O =0.190 m O O =0.058 m

1
L2 U O2 N 2 Q L V P | V P | 1m
1 2
E5b 118 x 10.23 MHz O =0.248 m OW = 0.814 m
2
E5a 115 x 10.23 MHz O =0.255 m OEW =9.768 m LW U OW NW Q L W

PN U Q P N
PEN U Q P EN
f1 / f 2
2
LEW PEN 1 J 12 (77 / 59) 2
N EW V N | VP > @
OEW f 2 / f3
2
J 23 (118 /115) 2
OEW
EN
roundoff
EW
J 12 1
VP V > @
N
J 12 1 P 1
OEW N EW LEW LW 1 J 25 1
N W V N | VL > @ VP V > @
OW roundoff W OW EW EN
J 25 1 P 1
J 12 1
> @
gAGE/UPC
VL V
L1 L2 O2 N W 1
W
J 12 1 L 1
N 1
V N | 2V > @
O1 O2 O1 O2 L1
J 25 1
> @
1
roundoff VL V
EW
J 25 1 L 1

49
Exercise:
gAGE
Repeat the previous study for the Galileo signals E1, E5b and E5a
V L | V L | 1cm
Galileo Frequency Wavelengths Combinations

L1 U O1 N1 Q L 1 2
E1 154 x 10.23 MHz O =0.190 m O O =0.058 m

1
L2 U O2 N 2 Q L V P | V P | 1m
1 2
E5b 118 x 10.23 MHz O =0.248 m OW = 0.814 m
2
E5a 115 x 10.23 MHz O =0.255 m OEW =9.768 m LW U OW NW Q L W

PN U Q P N
PEN U Q P EN
f1 / f 2
2
LEW PEN 1 0.71m J 12 (77 / 59) 2
N EW V N | VP 0.07
OEW f 2 / f3
2
9.768 m J 23
OEW
EN
(118 /115) 2
roundoff
EW
J 12 1
VP V 0, 71m
N
J 12 1 P 1
OEW N EW LEW LW 1 54.9 cm J 25 1

N W V N | VL 0.67 VP V 0, 71m
OW roundoff W OW EW
81.4 cm EN
J 25 1 P 1
J 12 1
gAGE/UPC
VL V 5.4 cm
L1 L2 O2 N W 1 1.4 cm
W
J 12 1 L 1
N 1 V N | 2V 1/ 4
O1 O2 5.8cm J 25 1
O1 O2 1 L1
roundoff VL V 54.9 cm
EW
J 25 1 L 1

50
gAGE
Contents
gAGE/UPC
- LAMBDA Method.

51
gAGE
Resolving Ambiguities as a set
As a driven problem to study the ambiguity fixing, we will consider problem of
differential positioning in DD for short baselines (e.g. < 10 km). To simplify, we
will consider only carrier measurements at a single or dual frequency.

Static position Equations Unknowns
L12q (ti ) U 12 (ti ) N q12 X 12
Lq (ti )
Single frequency (K-1)* nt 3+(K-1)
Lq (ti ) U (ti ) N q X Lq (ti )
13 13 13 13
Dual frequency 2(K-1)*nt 3+2(K-1)
q 1, 2
Kin. position Equations Unknowns
L1K 1 (t ) U 1K 1 (t ) N 1K 1 X 1K 1 (t )
q i i q Lq i Single frequency (K-1)* nt 3* nt+(K-1)
Dual frequency 2(K-1)*nt 3* nt+2(K-1)
In principle, the estimation of ambiguities in this system K t 4, nt t 2 K t 5, nt t 4
is not a big problem if we can wait enough time
and the unmodelled errors are not so large. Each epoch brings a set of (K-1)
equations for each frequency.
Linear Model: Note: nt is the number of epochs
Lqjk (ti ) U jk (ti ) N qjk X Ljk (ti ) q Ljk (ti ) U 0 jk (ti ) 0 jk (ti ) 'Uru (ti ) O N jk Q L jk (ti )
gAGE/UPC
Prefit-residual
U jk
(ti ) U jk
(ti ) 0 jk (ti ) 'U (ti ) G(t)
0 y(t)
We can estimate all parameters (position and ambiguities)

as a set by considering the over-dimensioned system y (ti ) G (ti ) 'r (ti ) O N
of linear equations and solving it by the LS criterion.
52
gAGE Resolving Ambiguities as a set

G (ti ) 'r (ti ) O N (ti ) Single freq.: K=K-1
y (ti )

Dual freq.: K=2(K-1)
K u3
K 3 K
vector matrix vector vector
For static positioning, considering two epochs (for instance):

y (ti ) G (ti ) I (ti )
y (t ) G (t ) 'r O I N (t )
i 1 i 1 i 1
In general, mixing several epochs, we will write: y G 'r O A N

Using the least-squares criterion, we can look for a real valued 3-vector 'r
and a K -vector of integers N that minimizes the cost function (sum of squared
residuals):
Weighted norm can
c('r, N) y G 'r O A N be taken as well
gAGE/UPC
The problem can be easily reformulated for the kinematic case. Kalman filtering
can be applied as well.

53
gAGE Resolving Ambiguities as a set

Different strategies can be applied:
To Float the ambiguities (i.e. treating the ambiguities as real numbers).

To Search ambiguities over a limited set of integers to find the best
solution.
To solve as an Integer Least-Squares problem.
For an observation span relatively long, e.g. one hour, the floated ambiguities
would typically be very close to integers, and the change in the position
solution from the float to the fixed solution should not be large.
As the observation span becomes smaller, ambiguity resolution play a more

important role. But very short observation spans implies the risk of wrong
ambiguity fixing, which can degrade the position solution significantly.
The performance, is thence measured by:

gAGE/UPC
1. Initialization time
2. Reliability (or, correctness) of the integer estimates

54
gAGE Search techniques
Strategy:
Define a volume to be searched

Set up a grid within this volume
Define a cost function (e.g. the sum of squared residuals)
Evaluate the cost function at each grid point
Solution corresponds to the grid point with the lowest value of the cost function
Position domain Ambiguity domain

LSAS (Hatch, 1990)
Ambiguity Function Method (AFM) LAMBDA (Teunissen, 1993)
ARCE MLAMBDA (Chang et al. 2005)
OMEGA (Kim and Langley, 2000)
gAGE/UPC
FASF (Chen and Lachappelle, 1995)

IP (Xu et al., 1995)

55
gAGE Search techniques

A conceptually simpler approach would consist on:
Estimate the floated solution N and its uncertainty (e.g. N =2502347.74 cycles, V N 0.6 cycles)

Define as a volume to be searched e.g. r 3V N r2 cycles and
evaluate the cost function (the RMS residuals) over the 6 ambig.: 2502345, , 2502350
The previous search must be done for each satellite in view.

- If there are 5 satellites tracked 4 DD ambiguities 64 = 1 296 combinations
- If there are 8 satellites tracked 7 DD ambiguities 67 =279 376 combinations
The integer ambiguity solution corresponding to the smallest RMS residuals is used
to select the candidate. However if two or more candidates give roughly similar
values of RMS, the test can not be resolute.
A ratio test (of 2 or 3, depending of the algorithm) between the two smallest RMS
gAGE/UPC
is often used to validate the test.

If the ratio is under these values, no integer solution can be determined and is
better to use the floated solution.

56
gAGE
Contents
gAGE/UPC
- LAMBDA Method

57
gAGE LAMBDA Method

Consider again the previous problem of estimating 'r , a 3-vector of real
numbers, and N a (K-1)-vector of integers, which are solution of

y G 'r O A N min y G 'r O A N Wy
To better exploit the internal correlations [*], we consider now the

covariance Wy Py1
Let be the float solution 'r 'r P'r P'r ,N

; Cov
and covariance matrix: N N P'r ,N PN
It can be shown the following orthogonal decomposition:

2 2
y G 'r O A N
2

y G 'r O A N 'r 'r (N)
2

O2 N N
Wy Wy W'r ( N ) WN
gAGE/UPC
Residual of real-valued

floated solution 'r , N
[*] Remember that DD measurements are correlated, as already
seen.
58
gAGE LAMBDA Method
Thence, we have to find 'r a 3-vector of real numbers, and N a (K-1)-vector
of integers minimizing:
2 2
y G 'r O A N
2

y G 'r O A N 'r 'r (N)
2

O2 N N
Wy Wy W'r ( N ) WN
This term is irrelevant for This term can be This term must
minimization since it does made zero for be minimized
not depend on 'r and N any N over the integers
2
Float solution and covariance matrix:
min N N o N
WN

'r 'r
; Cov
P'r

P'r ,N

'r 'r (N)
'r W'r ,N WN1 N N
N
N P'r ,N PN

gAGE/UPC
The vectors 'r and N are often

W'r ,N P'r1,N referred to as the fixed user solution
WN1 PN
and fixed ambiguity.

59
gAGE LAMBDA Method

The integer search: Finding the integer vector N that minimizes the cost function
N N
2 T
c(N)
N N
WN N N
WN WN PN1
A diagonal WN matrix would mean that the integer ambiguity estimates are
uncorrelated.
If the weight WN matrix is diagonal, the minimizing of the cost function is trivial. The
best estimate is the float ambiguity rounded to the nearest integer.
N 2

2 2
1/ V 2 0 N1 N 1 N 2 N 2
c(N)
WN
N1N1
V N2 N V N2 N
0 1/ V 2
1 1 2 2 N1
N 2 N 2
Ellipse parallel to coordinate axes
In practice, the estimated (float) ambiguities are highly corre-

lated and the ellipsoidal region stretches over a wide range of
cycles. This is specially the case when the measurements are
limited to a single epoch or only a few epochs.
gAGE/UPC
Thence, points that appears much further away from the

floated solution may have lower values of cost function than
those which appear nearby. In this context, the search for
integer vectors can by extremely inefficient.
60
To improve the computational efficiency of the search, the float ambiguities
gAGE can be transformed so that the elongated ellipsoid turns into a sphere-like.
Thus, the search can be limited to the neighbours of the floated ambiguity.
The idea is to apply a transformation that decorrelates the ambiguities so that the matrix
W becomes diagonal. W is a positive definite matrix and thence, can be always
diagonalized (as a real-valued matrix) with orthogonal eigenvectors. But the problem here
is that the integer ambiguities N must be transformed preserving its integer nature!
Thence, we are looking for an integer-valued transformation matrix Z that makes the
matrix W as close as possible to a diagonal matrix (decorrelating as much as possible
the ambiguities) and with similar axes (spherical).
Moreover, the inverse of transformation
matrix Z-1 must be also integer, to transform
Nc ZN back the results after finding the ambiguities
PN c Z PN ZT
c
N
ZN Note that Z, Z 1 det(Z) 1
(i.e. it is a volume-preserving transformation)
Z
gAGE/UPC
Pictures
from
[RD-6]

61
gAGE
Exercise:
Show that:
Z, Z 1 det(Z) 1
That is, Z is a volume-preserving transformation

gAGE/UPC

62
From [RD-4]
gAGE Decorrelation: Computing the Z-transform
The following conditions must be fulfilled:

1. Z must have integer entries

2. Z must be invertible and have integer entries
3. The transformation Z must reduce the product of all
ambiguity variances.
1
Note that Z, Z det(Z) 1
(i.e. it is a volume-preserving transformation)
Gauss manipulation over matrix P=W-1 can be applied to find-out the matrix Z.
1 0 Transforms N2 (N1 remains unchanged)

Z1 D
pN1N1 pN N
1 1
PN
1 2
Di int pN N / pN N
pN1N 2 pN N 1 2 i i
2 2 1 D 2
Z2 0 1 Transforms N1 (N2 remains unchanged)

gAGE/UPC
Note: Inverse matrices have also integer entries

Start transforming first the
1 0 1 D 2
Z11 Z 21 element with largest variance.

D1 1 0 1

63
Gauss manipulation over matrix P=W-1 can be applied to find-out the matrix Z
gAGE
1 0 Transforms N2 (N1 remains unchanged)
Z1 D
pN1N1 pN N
1 1
PN
1 2
Di int pN N / pN N
pN1N 2 pN N 1 2 i i
2 2 1 D 2
Z2 0 1 Transforms N1 (N2 remains unchanged)

Example: 1.05 53.4 38.4 Example

N PN from [RD-4]
1.30 38.4 28.0

Step 1:
1 1 We transform first the element with
Z2 0 1 D 2 int >38.4 / 28.0@ 1 largest variance (in this case N1)

1 1 53.4 38.4 1 0 4.6 10.4 The half,
PN c Z 2 PN ZT2 0 1 38.4 28.0 1 1 10.4 28.0 at most!

In general,
Step 2: to increase the
1 0
gAGE/UPC
Z1 2 1 D1 int >10.4 / 4.6@ 2 number of small off-

diagonal elements,
we have to
1 0 4.6 10.4 1 2 4.6 1.2 transform first the
PN cc Z1 PN c Z1T 2 1 10.4 28.0 0 1 1.2 4.8
Master of Science in GNSS elements with
largest variance 64
Example and
1 1 1 0 pictures from
gAGE Z2 Z1 2 1
0 1 [RD-4]
53.4 38.4 4.6 10.4 4.6 1.2

PN 38.4 28.0 PN c 10.4 28.0 PN cc 1.2 4.8

1 0 1 1 1 1
Z Z1Z 2 2 1 0 1 2 3
gAGE/UPC

1 1 53.4 38.4 1 2 4.6 1.2
PN cc Z PN ZT 2 3 38.4 28.0 1 3 1.2 4.8

65
Example:
gAGE
1.05 53.4 38.4 4.6 1.2
N 1.30 PN 38.4 28.0 PN cc 1.2 4.8

1 1
Z 2 3

cc 1 1 1.05 0.25 0.25 0

N ZN 2 3 1.30 1.80 Ncc int 2
1.80
gAGE/UPC
0 3 1 0 2
N Z 1 2 1 2 2
2

66
Let P be a symmetric and positive-definite matrix:
gAGE
1 N2
p11 p12 p11 p22 w N 2c
P O1
p p22
2 N1c
12 1
O2 p11 p22 w O1
2 O2 I
w ( p11 p22 ) 2 4 p122 N1
O1 0
Pc 0
O2 tan 2I
2 p12
p11 p22
Example:
53.4 38.4 1.05

PN 38.4 28.0 N 1.30

O1 81.14 9.0
gAGE/UPC
81.14 0
PNc 0 O2 0.25 0.5
0.25
tan 2I 3.02 I 35$85
BACKUP
67
gAGE Consider again the previous problem of estimating 'r , a 3-vector of real
numbers, and N a (K-1)-vector of integers, which are solution of
G 'r O A N
The solution comprises the following steps:

'r P'r P'r ,N
;
1. Obtain the float solution and its covariance matrix: N
P'r ,N PN
2. Find the integer vector N which minimizes the cost function

N N WN PN1
2 T
c(N)
N N
WN N N
WN
a) Decorrelation: Using the Z transform, the ambiguity search space is

re-parametrized to decorrelate the float ambiguities.
b) Integer ambiguities estimation (e.g. using sequential conditional least-
squares adjustment, together with a discrete search strategy).
c) Using the Z-1 transform, the ambiguities are transformed to the
original ambiguity space.
gAGE/UPC
3. Obtain the fixed solution 'r, from the fixed ambiguities N.

y O AN G 'r
68
gAGE b) Integer ambiguities estimation
Several approach can be applied:

Integer rounding
Comment:
Integer bootstrapping In principle, the previous transformation Z
is not required by the estimation concept;
Integer Least-Squares
it is only to achieve considerable gain in
.. speed in the computation process [RD-5].
b1) Integer rounding

This is the simplest way.
Just to round-up the ambiguity vector entries
to its nearest integer
N int( N1 ), ,int( N K )
For instance, in the previous example:

gAGE/UPC
0.25 0
Ncc int 2
1.80

69
gAGE b2) Integer bootstrapping (from [RD-6])
It makes use of integer rounding, but it takes some of the correlations

between the ambiguities into account.
1. We start with the most precise ambiguity (here we will assume Nn)
2. Then, the remaining float ambiguities are corrected taking into account
their correlation with the last ambiguity.

N n int N n Using the triangular
decomposition
n1 n n

N n 1 int N n 1|n int N n 1 V N , N V N2 N n N n
PN LT D L

lij V N V N2

n
N1 int N n|I int N1 V N , N V N2 N i|I N i
j , N i|I i|I
i 2
1 i|I i|I

gAGE/UPC
N i|I Stands for the i-th ambiguity obtained through a conditioning of the
previous I ^i 1, , n` sequentially rounded ambiguities.

70
Example and pictures from [RD-6]
gAGE
N1 N 1 V N
2 , N1 2

V N2 N 2 N 2
Float
N2
Conditional
estimate
N1|2
N1
gAGE/UPC

N2 nint N 2 0 N1 nint N 1|2 nint N 1 V N , N V N2 N 2 N 2 1
2 1 2

71
gAGE b3) Integer Least Squares (ISL) (from [RD-6])
1. The target to find the integer vector N which minimizes the cost function
N N PN1
2 T
c(N)
N N
PN1 N N WN
PN 1
2. The integer minimiser is obtained through a search over the integer grid
points on the n-dimensional hyper-ellipsoid:

T
P 1 N N
N N d F2
N
Where F determines the size of search region.

The solution is the integer grid point N, inside the ellipsoid, giving the
minimum value of cost function c(N).
where: di V N2
i|I
Using the triangular decomposition: PN L DL

T
li j V N
j , N i|I
V 2
N i|I
N N N N
T
d F2
L1 D1 LT N N
T
d F2
D1 N N
Defining:
N N N
gAGE/UPC
2 2 2
N 1 N 2 N n
) o LT N
N LT (N N
N N N)
(N c( N )
1 2

n
d F2
d1 d2 dn

72
gAGE
N N N
2 2 2
N 1 N 2 N n
T

N N D 1 d F2
N N 1 2
d F2
n
c( N )
d1 d2 dn
But N i depends on N i 1 , , N n .
N n N n
N
L N
T
N)
(N
N
n
N i N i j N i l ji ; i n 1, n 2, ,1
j i 1
Search region bounds:

N n d n1/ 2 F d Nn d N n d n1/ 2 F

2 1/ 2 2 1/ 2
N n 1 d n1/21 F 2 N n N n dn d N n 1 d N n 1 d n1/21 F 2 N n N n dn

1/ 2 1/ 2

n 2 n 2
N 1 d11/ 2 F 2 N j N j dj d N1 d N n d11/ 2 F 2 N j N j dj
j 2 j 2
gAGE/UPC
Acceptance test: The integer ambiguity solution corresponding to the

smallest RMS residuals is used to select the candidate. However if two or more
candidates give roughly similar values of RMS, the test can not be resolute.
A ratio test (of 2 or 3, depending of the algorithm) between the two smallest
RMS is often used to validate the test.
73
Ellipsoid size: selecting the candidates for the acceptance test

gAGE
The size of the ellipsoidal search region N N D N N d F is controlled by F
T
1 2
Therefore, the performance of the search process is highly dependent on F:

A small F may result in a ellipsoidal region that fails to contain the solution.
A too large value for F may result in high time-consuming for the search process.
Search with enumeration: When the number of required candidates is at most

n+1 (with n=dim(N)), the following procedure can be applied to set the value F:
The best determined ambiguity is rounded to its nearest integer. The remaining
ambiguities are then rounded using their correlations with the first ambiguity:

N n nint N n

N n 1 nint N n 1|n nint N n 1 V N , N V N2 N n N n

based on the
n1 n n
bootstrapping

estimator

n
N1 nint N n|I nint N 1 V N , N V N2 N i|I N i
i 2
1 i|I i|I

In each step of the conditional rounding procedure, two candidates are taken: The
nearest and second-nearest, and conditional rounding is proceeded in both cases.
gAGE/UPC
If p candidates are requested, the values of cost function c(N) are ordered in
ascending order and F is chosen equal to the p-th value.
If more than n+1 candidates are requested, the volume of the search ellipsoid can be used
([RD-6]).
74
gAGE
Search with shrinking technique: practical example
This is an alternative to the previous strategy, based on shrinking the search
ellipsoid during the process of finding the candidates.
In the next example, we have to choose 6 candidates:

Round Ambiguty-2 to the second near integer
Fist candidate is the and round the new conditional estimate for
bootstrapped solution Ambiguity-1
gAGE/UPC
The other 5 candidates are found by The candidate with the

choosing the conditional Ambiguity-1 to largest F is removed.
the 2nd, 2rd, 4th and 5th nearest integers
Master of Science
Example in GNSS
and pictures from [RD-6] @ J. Sanz & J.M. Juan
75
Example and pictures

2nd nearest 3rd nearest
gAGE from [RD-6]
4th nearest is
out of the
region.
No more
candidates at
gAGE/UPC
this step
Thence, the best 6 candidates are found (in the ISL sense). The one
with
Master thein GNSS
of Science smallest cost function c(N) value is the actual ISL@ J.solution.
Sanz & J.M. Juan
76
gAGE
research group of Astronomy and Geomatics Acceptance Test
The integer ambiguity solution corresponding to the smallest RMS residuals

is used to select the candidate.
However if two or more candidates give roughly similar values of RMS, the
test can not be resolutive.
A ratio test (of 2 or 3, depending on the algorithm) between the two

smallest RMS is often used to validate the test.
If the ratio is under these values, no integer solution can be determined
and is better to use the floated solution.
N N
T
RMS
N N
PN1 N N
gAGE/UPC
PN 1

77
gAGE Examples with MATLAB (octave)

Note:
This document uses
the transposed
matrix ZT, but the
gAGE/UPC
principle is the
same.

78
gAGE Examples with MATLAB (octave)
Q { PN WN1
load large Q, a
[Qz,Zt,Lz,Dz,az,iZ] = decorrel (Q,a);
imagesc(Q) imagesc(Qz)
colorbar colorbar
Z
gAGE/UPC
Master of Science in GNSS Qz=Lz*diag(Dz)*Lz

79
gAGE [Qz,Zt,Lz,Dz,az,iZ] = decorrel (Q,a); Z=Zt

[L,D] = ldldecom(Q)
Qz=Lz*diag(Dz)*Lz
Q=L*diag(D)*L
az= Z*a Qz= Z*Q*Z

a= inv(Z)*az Q= inv(Z)*Qz* inv(Z)
Z =
3 0 -4 -3 -5 -4 -4 2 -2 1 -3 1
-0 -1 1 -1 -2 4 4 -3 4 1 0 1
3 5 -2 -2 1 -1 -2 1 -1 -4 -1 -1
-5 -2 3 2 4 3 -3 -2 -2 1 -3 -1
4 5 1 4 2 6 5 2 -4 1 2 -4
-8 -4 1 0 0 -3 2 3 2 -1 -0 4
4 -7 -0 1 0 -4 -1 -7 3 -5 -1 2
2 -1 -8 -1 2 -4 1 2 -4 2 2 -2
gAGE/UPC
-3 2 3 10 -8 -2 -5 0 -4 1 -4 0
-1 6 8 -1 2 1 2 7 3 -2 6 1
-8 7 -8 3 -6 -1 1 0 0 3 -1 -1
8 1 6 -3 5 4 -5 -3 0 -0 1 -3

80
load large Q, a
gAGE
Integer rounding
round(a)
[ - 28491 65753 38830 5004 -29196 -298 -22201 51236 30258 3899 -22749 -159]
Decorrelation + Integer rounding

[Qz,Zt,Lz,Dz,az,iZ] = decorrel (Q,a)
azfixed=round(az);
afixed=iZ*azfixed
[ -28537 65473 38692 4939 -29228 -504 -22237 51018 30150 3849 -22774 -320]
Decorrelation + bootstrapping
gAGE/UPC
azfixed=bootstrap(az,Lz);
afixed=iZ*azfixed
[ -28451 65749 38814 5025 -29165 -278 -22170 51233 30245 3916 -22725 -144]

81
gAGE
afixed=iZ*azfixed
[ -28451 65749 38814 5025 -29165 -278 -22170 51233 30245 3916 -22725 -144]
Decorrelation + ILS with enumeration search
[azfixed,sqnorm] = lsearch (az,Lz,Dz,6);

afixed=iZ*azfixed
c(N)
-28451 65749 38814 5025 -29165 -278 -22170 51233 30245 3916 -22725 -144 15.0
-28279 65862 38805 5170 -29061 -192 -22036 51321 30238 4029 -22644 -77 31.6
gAGE/UPC
-28727 65935 39032 4844 -29337 -178 -22385 51378 30415 3775 -22859 -66 33.9
-28546 66062 39027 4998 -29228 -83 -22244 51477 30411 3895 -22774 8 34.5
-28229 65518 38583 5197 -29056 -500 -21997 51053 30065 4050 -22640 -317 34.7
-28365 65586 38683 5084 -29124 -418 -22103 51106 30143 3962 -22693 -253 35.5

82
gAGE
afixed=iZ*azfixed
[ -28451 65749 38814 5025 -29165 -278 -22170 51233 30245 3916 -22725 -144]
Decorrelation + ILS with search-and-shrink
[azfixed,sqnorm] = ssearch (az,Lz,Dz,6);

afixed=iZ*azfixed
c(N)
-28451 65749 38814 5025 -29165 -278 -22170 51233 30245 3916 -22725 -144 15.0
-28279 65862 38805 5170 -29061 -192 -22036 51321 30238 4029 -22644 -77 31.6
gAGE/UPC
-28727 65935 39032 4844 -29337 -178 -22385 51378 30415 3775 -22859 -66 33.9
-28546 66062 39027 4998 -29228 -83 -22244 51477 30411 3895 -22774 8 34.5
-28229 65518 38583 5197 -29056 -500 -21997 51053 30065 4050 -22640 -317 34.7
-28365 65586 38683 5084 -29124 -418 -22103 51106 30143 3962 -22693 -253 35.5

83
gAGE References

[RD-5] P. Joosten and Tiberius, C. LAMBDA: FAQs. GPS Solutions (2202),
6:109-114.
[RD-6] Sandra Verhagen and Bofeng L., LAMBDA software package. MATLAB
implementation, Version 3.0. Mathematical Geodesy and Positioning,
Delft University of Technology.
gAGE/UPC
[RD-7] XW. Chang, CC. Paige, L. Yin, 2005, Code and carrier phase based
short baseline GPS positioning: computational aspects. GPs Solutions,
2005 (2005) 9:72-83, DOI 10.1007/s10291-004-0112-8 Springer.
84
gAGE
Least-Squares Ambiguity Search technique
This technique requires an approximate solution, which can be

obtained from code range measurements. The search area can be
defined by surrounding the approximate position by a 3V region
(i.e. 'r r G ).
1
y G 'r O N
N y G 'r
O
K-1 equations with Then , given the 3-D position, the (K-1) integer
3+(K-1) unknowns. ambiguities can be resolved automatically.
Thence, That is, the (K-1) integer ambiguities are
K-4 equations are redundant constrained to three degrees of freedom.
gAGE/UPC

85
'r
gAGE
The technique is based on exploiting the
constrains on the integer ambiguities: 'r V G (k ) 'r V
The tracked satellites into two groups: 4 sat. (with good DOP) + N-4 sat.
The primary group of 4 sat., is used determine the possible ambiguity sets N ( k )
(given an initial position estimate and its associated uncertainty 'r r V )
'r r V o N(k )
1
O
y G 'r G
(k ) o N(k )
Then, the corresponding position estimates 'r( k ) are computed.
G G
1
N(k ) o 'r( k ) T
G y O N ( k )
The remaining N-4 secondary satellites are used to eliminate candidates of

the possible ambiguity sets:
Each position estimate is checked against measurements from the
secondary group of N-4 satellites.
With the correct position estimate, the difference between the
gAGE/UPC
measured and computed carrier phase should be close to an integer

for each satellite pair. 1
O
y G 'r N (k ) (k ) ?
86
gAGE
Contents
gAGE/UPC
- LAMBDA Method

87
List of Acronyms
List of Acronyms
AGW Atmospheric Gravity Waves
ANTEX ANTenna EXchange format
APC Antenna Phase Centre
ARP Antenna Reference Point
ASCII American Standard Code for Information Interchange
A/S Anti-Spoofing
C/A Coarse/Acquisition
CDDIS Crustal Dynamics Data Information System
CODE Centre for Orbit Determination in Europe
CRF Celestial Reference Frame
CRS Conventional Celestial Reference System
CS Cycle Slip
DAT Data Analysis Tool
DCB Differential Code Bias
DLL Dynamic Link Library
DLR Deutsches Zentrum fur Luft- und Raumfahrt
DOP Dilution Of Precision
DoY Day of Year
DPC Data Processing Core
ECEF Earth-Centred, Earth-Fixed
ECI Earth-Centred Inertial
EMR Energy Mines and Resources
ENU East North Up
ESA European Space Agency
gAGE Research group of Astronomy and Geomatics
319
TM-23/2
GAL GALileo Satellite Identifier
GDOP Geometric Dilution Of Precision
GEO GEOstationary Satellite Identifier
GIPSY GPS Inferred Positioning SYstem
gLAB GNSS-LAB tool suite
GLO Glonass Satellite Identifier
Glonass GLObal NAvigation Satellite System
GNSS Global Navigation Satellite System
GNU GNUs Not Unix
GPS Global Positioning System
GRACE Gravity Recovery and Climate Experiment
GRAPHIC Group and Phase Ionospheric Calibration
GFree Geometry Free
GUI Graphical User Interface
HDOP Horizontal Dilution Of Precision
HTML HyperText Markup Language
IAC Information Analytical Centre
ICD Interface Control Document
IFree Ionosphere Free
IGL IGS Glonass orbit products
IGRF International Geomagnetic Reference Field
IGS International GNSS Service
IGST IGS Time
IODE Issue Of Data Ephemerides
IODN Issue Of Data Navigation
IONEX IONosphere map EXchange format
IRI International Reference Ionosphere
ITRF International Terrestrial Reference Frame
JPL Jet Propulsion Laboratory
LEO Low Earth Orbit
LS Least Squares
320
List of Acronyms
LSTIDs Large-Scale TIDs
MC Satellite Mass Centre
MCC Mission Control Centre
MJD Modified Julian Day
MM Monument Marker
MSDOS Microsoft Disk Operating System
MSTID Medium-Scale Travelling Ionospheric Disturbance
MSTIDs Medium-Scale Travelling Ionospheric Disturbances
MW MelbourneWubbena
NANU Notice Advisory to NAVSTAR Users
NASA National Aeronautics and Space Administration
NEU North East Up
NGA National Geospatial-Intelligence Agency
NSE Navigation System Error
OS Operating System
PCO Phase Centre Offset
PCV Phase Centre Variation
PDOP Precision Dilution Of Precision
PNG Portable Network Graphics
PPP Precise Point Positioning
PRN Pseudo-Random Noise
PVT Position, Velocity, Time
PZ-90 Parametry Zemli 1990 (Parameters of Earth 1990)
RINEX Receiver INdependent EXchange format
RK4 Fourth-order RungeKutta
RMS Root Mean Square
RO Radio Occultation
S/A Selective Availability
SBAS Satellite-Based Augmentation System
SED Storm Enhancement Density
SINEX Solution (Software/technique) INdependent EXchange format
321
TM-23/2
SIS Signal In Space
SISRE Signal In Space Range Error
SOHO SOlar Helioscopic Observatory
SP3 Standard Product #3
SPP Standard Point Positioning
SPS Standard Positioning Service
SSI Signal Strength Indicator
STEC Slant Total Electron Content
STROP Slant TROPospheric delay
SU Soviet Union
SV Space Vehicle
SVN Space Vehicle Number
TDOP Time Dilution Of Precision
TEC Total Electron Content
TECU Total Electron Content Unit
TGD Total Group Delay
TID Travelling Ionospheric Disturbance
TRF Terrestrial Reference Frame
TRS Conventional Terrestrial Reference System
TUM Technische Universitat Munchen
UPC Technical University of Catalonia
USNO United States Naval Observatory
UTC Coordinated Universal Time
VDOP Vertical Dilution Of Precision
WGS-84 World Geodetic System 84
ZTD Zenith Tropospheric Delay
322

GNSS Data Processing Theory Slides

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

GNSS Data Processing Theory Slides

Hochgeladen von

Copyright:

Verfügbare Formate

gAGE

GNSS Data Processing

@ J. Sanz Subirana & J.M. Juan Zornoza

These slides can be obtained either from the server http://www.gage.upc.edu,

Lecture 1: GNSS measurements and their combinations.

Lecture 2: Satellite orbits and clocks computation and accuracy.

Lecture 3: Position estimation with pseudorange.

Lecture 4: Introduction to DGNSS.

Lecture 5: Precise positioning with carrier phase (PPP).

Lecture 6: Differential positioning with code pseudoranges.

Lecture 7: Carrier based differential positioning. Ambiguity resolution techniques.

Web site: http://www.gage.upc.edu

Master of Science in GNSS @ J. Sanz & J.M. Juan

gAGE Specific Objectives:

characteristics, properties, combinations and applications.

To get tools and skills to process and analyze GNSS

Master of Science in GNSS @ J. Sanz & J.M. Juan

A ship determines its location from a set on lighthouses that

Knowing the emission time

With two lighthouses there

A ship determines its location from a set on lighthouses that

With three lighthouses a

The ranges are measured by means the

traveling time of the acoustic signal from

SUMMARY: The positioning system is based on:

To solve a geometric problem.

NOTE: the ranges are measured by means

the traveling time of the acoustic signal

Thence, the receiver coordinates are found solving a

One of the solutions is not

on the Earth surface.

Master of Science in GNSS @ J. Sanz & J.M. Juan

Master of Science in GNSS @ J. Sanz & J.M. Juan

Master of Science in GNSS @ J. Sanz & J.M. Juan

Master of Science in GNSS @ J. Sanz & J.M. Juan

delays are modeled

Master of Science in GNSS @ J. Sanz & J.M. Juan

xio  x1 yio  y1 zio  1

problem is linearized, and removed.

Master of Science in GNSS @ J. Sanz & J.M. Juan

[RD-1] J. Sanz Subirana, J.M. Juan Zornoza, M. Hernndez-Pajares, GNSS

Master of Science in GNSS @ J. Sanz & J.M. Juan

Master of Science in GNSS @ J. Sanz & J.M. Juan

Web site: http://www.gage.upc.edu

Master of Science in GNSS @ J. Sanz & J.M. Juan

These slides can be obtained either from the server http://www.gage.upc.edu,

Master of Science in GNSS @ J. Sanz & J.M. Juan

1. Review of GNSS measurements.

Master of Science in GNSS @ J. Sanz & J.M. Juan

1. Review of GNSS measurements.

Master of Science in GNSS @ J. Sanz & J.M. Juan

S L(1k ) (t ) aP X P( k ) (t ) D ( k ) (t ) sin(Z1t  IL1 )  aC X C( k ) (t ) D ( k ) (t ) cos(Z1t  IL1 )

Master of Science in GNSS @ J. Sanz & J.M. Juan

P(T ) c 'T c trec (T )  t sat (T  'T )

From hereafter we will call:

C1 pseudorange computed from XC(t) binary code (on frequency 1)

IL (T ) IL rec (T )  ILsat (T  'T )

From hereafter we will call:

C1 pseudorange computed from XC(t) binary code (on frequency 1)

P1 is an absolute measurement (unambiguous)

Each time that the receiver loose the

Code and Carrier Phase measurements

Code (unambiguous but noisier)

Carrier Phase (ambiguous but precise)

xio x1 yio y1 zio 1

S L(1k ) (t ) aP X P( k ) (t ) D ( k ) (t ) sin(Z1t IL1 ) aC X C( k ) (t ) D ( k ) (t ) cos(Z1t IL1 )

P(T ) c 'T c trec (T ) t sat (T 'T )

IL (T ) IL rec (T ) ILsat (T 'T )

U sta c (dtsta dt sat ) Tropsta Ion1sat