Lecture UPC 3.v0.0

gAGE
research group of Astronomy and Geomatics
Lecture 3
Position estimation with
pseudoranges
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
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.
24 April 2014
gAGE/UPC

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…
= ρ rec + c ⋅ ( dtrec − dt sat ) + ∑ δ k + ε

sat sat
C1rec
Linearising ρ around an ‘a priori’ receiver position ( x0,rec , y0,rec , z0,rec )
x0,rec − x sat y0,rec − y sat z0,rec − z sat

= ρ0,satrec + ∆xrec + ∆yrec + ∆zrec + c ( dtrec − dt sat ) + ∑ δ k
ρ sat
0, rec ρ sat
0, rec ρ sat
0, rec
where:
∆xrec =xrec − x0,rec ; ∆yrec =yrec − y0, rec ; ∆zrec = zrec − z0, rec
Prefit-residuals (Prefit)
sat
C1rec ∑δk
− ρ0,satrec + c dt sat −=
ρ sat
∆xrec +
ρ sat
∆yrec +
ρ sat
∆zrec + c dtrec
gAGE/UPC
0, rec 0, rec 0, rec
measurement computed unknown

6
For all satellites
gAGE  x0,rec − x sat1 y0,rec − y sat1 z0,rec − z sat1 
in view 
ρ0,satrec1 ρ0,satrec1 ρ0,satrec1
1
 
Prefit
Prefit  ∆∆xxrec
11
x  rec 

 22 

− x sat 2 y0,rec − y sat 2 z0,rec − z sat 2 ∆y 
 Prefit
Prefit 
0, rec
1  ∆yrec
rec 
=  ρ 0,satrec2 ρ0,satrec2 ρ0,satrec2 
 ........
........    ∆∆zzrec 
    rec 
 .......... 
Prefit
Prefit nn
x  ccdt
dtrec 
rec 
0, rec − x y0,rec − y sat n z0,rec − z sat n
sat n
 1
 ρ 0,satrecn ρ0,satrecn ρ0,satrecn 
Observations Unknowns
(measured/computed)
Measurements modelling:
Prefit residual is the difference between measured and modeled
pseudorange:
=
Prefit rec C1rec [measured] − C1rec [modelled]
sat sat sat
where:
gAGE/UPC
sat
C1rec = ρ rec
[modelled] sat
,0 − c ( 
dt sat
+ ∆rel sat
) + Trop sat
rec + Ion sat
1rec + TGD sat

7
gAGE
Contents

gAGE/UPC

8
gAGE
research group of Astronomy and Geomatics Code Pseudorange modeling
C1
The pseudorange modeling is based in the GPS Standard

Positioning Service Signal Specification (GPS/SPS-SS).
gAGE/UPC
sat
C1rec = ρ rec
[modelled] sat
,0 − c ( 
dt sat
+ ∆rel sat
) + Trop sat
rec + Ion sat
1rec + TGD sat

9
gAGE
gAGE/UPC
sat
C1rec = ρ rec
[modelled] sat
,0 − c ( 
dt sat
+ ∆rel sat
) + Trop sat
rec + Ion sat
1rec + TGD sat

10
gAGE
research group of Astronomy and Geomatics Geometric range
ρ rec
sat
Euclidean distance between satellite coordinates at emission time

and receiver coordinates at reception time.
(x − x0,rec ) + ( y − y0,rec ) + ( z − z0,rec )

2 2 2
ρ sat
0, rec = sat sat sat
Of course, receiver coordinates are not known (is the target of this
gAGE/UPC
problem). But ....

sat
C1rec = ρ rec
[modelled] sat
,0 − c ( 
dt sat
+ ∆rel sat
) + Trop sat
rec + Ion sat
1rec + TGD sat

11
(x − x0,rec ) + ( y − y0,rec ) + ( z − z0,rec )
gAGE 2 2 2
ρ sat
0, rec = sat sat sat
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 Earth’s 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 )
3.- Linearise the equations again, about the new improved estimates,
gAGE/UPC
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 Earth’s @Centre.
J. Sanz & J.M. Juan
12
gAGE
research group of Astronomy and Geomatics 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

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
gAGE T[em s]
The satellite offset
clock dt S can
Geomatics
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
research group of Astronomy and Geomatics Distance: ∆r
Distance ∆r
gAGE/UPC

15
gAGE
gAGE/UPC Variation in range: ∆ρ= ρemission − ρreception

16
gAGE
gAGE/UPC Vertical error comparison

17
gAGE
gAGE/UPC Horizontal error comparison

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
ρ’ 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:
T[ems]= trec(TR)-(C1/c+dt )
S
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
reception “∆t“ (CTS reference system at reception time is used to
gAGE/UPC
build the navigation equations).
(Xsat,Ysat,Zsat)CTS[reception] =R3(ωE ∆t).(Xsat,Ysat,Zsat)CTS[emission]

19
gAGE
research group of Astronomy and Geomatics Variation in range: ∆ρ= ρ’ – ρemission
ρ’
gAGE/UPC

20
gAGE

21
gAGE

22
gAGE
research group of Astronomy and Geomatics 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
sat
C1rec = ρ rec
[modelled] sat
,0 − c ( 
dt sat
+ ∆rel sat
) + Trop sat
rec + Ion sat
1rec + TGD sat

23
gAGE
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
9.636381916043D-01 2.153437500000D+02 3.056960010495D+00-8.030691653399D-09
gAGE/UPC
-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

24
gAGE
gAGE/UPC Range variation: satellite clocks

25
gAGE

26
gAGE

27
gAGE
research group of Astronomy and Geomatics Relativistic clock correction (∆rel)
• A constant component depending only on nominal value of satellite’s

orbit major semi-axis, being corrected modifying satellite’s clock
oscillator frequency*:
f 0' − f 0 1  v  ∆U
2
−10
=   + 2  −4.464 ⋅ 10
f0 2c c
• A periodic component due to orbit eccentricity (to be corrected by user

receiver):
µa r⋅v
∆ rel =−2 2 e sin( E ) = −2 2 ( seconds )
c c
Being µ=3.986005 1014 (m3/s2) universal gravity constant, c =299792458

(m/s) light speed in vacuum, a is orbit’s major semi-axis, e is its eccentricity,
E is satellite’s eccentric anomaly, and r and v are satellite’s 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 f’o=10.22999999543 MHz.

28
gAGE
gAGE/UPC Range variation: relativistic correction

29
gAGE

30
gAGE

31
gAGE Ionospheric Delay Ion f sat
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 I
Ion1 rec
signal frequency as given by: f12
Where I is number 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 = 2 2
f1 − f 2
• 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
ρ sat
0, rec − c ( 
dt sat
+ ∆rel sat
) + Trop sat
rec + Ion sat
1rec + TGD sat
32
gAGE Klobuchar model (klob.f)
Vertical delay
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 2

−1/ 2

layer at 350 Km in height. ) 1 − 
m (elev= E
cos (elev)  
  RE + h  
• The slant delay is computed
from the vertical delay at the
ionospheric Pierce Point (IPP),
gAGE/UPC
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
) 1 − 
m (elev= E
cos (elev)  
  RE + h  
gAGE/UPC
Ionospheric Layer
(350 Km in height)

34
gAGE
research group of Astronomy and Geomatics Klobuchar model
Klobuchar
30
coefficients
25
Time Delay (ns at 1.6 GHz)
20   2π (t − Φ ) 
 DC + A cos   (day )
  P 
Amplitude IonVERT =
15  DC ; if  2π (t − Φ )  > π (night )
  P  2
10
Being :
3 3
5
½*Period
=A
=
n
n 0=
∑=
α φ ; P ∑β φ
n
n
n
Dc=5ns n 0
φ = Geomagnetic Latitude
4 8 12 16 18 24
Local Time (hours) Where:
gAGE/UPC
DC= 5ns
−1/ 2
  R 
2
 Φ= 14 (ctt. phase offset)
) 1 − 
m (elev= E
cos (elev)  
  E R + h   t = Local Time

35
gAGE
(time, rsta, rsat,α0,α1,α2,α3,β0,β1,β2,β3) [Klob]  Iono

elev, φ
2 NAVIGATION DATA RINEX VERSION / TYPE

CCRINEXN V1.5.2 UX CDDIS 24-MAR- 0 00:23 PGM / RUN BY / DATE
IGS BROADCAST EPHEMERIS FILE COMMENT
0.3167D-07 0.4051D-07 -0.2347D-06 0.1732D-06 ION ALPHA
-0.2842D+05 -0.2150D+05 -0.1096D+06 0.4301D+06 ION BETA
-0.121071934700D-07-0.488498130835D-13 319488 1002 DELTA-UTC: A0,A1,T,W
13 LEAP SECONDS
END OF HEADER
1 99 3 23 0 0 0.0 0.783577561379D-04 0.113686837722D-11 0.000000000000D+00
0.191000000000D+03-0.106250000000D+01 0.487163149444D-08-0.123716752769D+01
-0.540167093277D-07 0.476544268895D-02 0.713579356670D-05 0.515433833885D+04
0.172800000000D+06-0.260770320892D-07-0.850753478531D+00 0.763684511185D-07
gAGE/UPC
0.957259887797D+00 0.241437500000D+03-0.167990552187D+01-0.823998608564D-08
0.174650132022D-09 0.100000000000D+01 0.100200000000D+04 0.000000000000D+00
0.320000000000D+02 0.000000000000D+00 0.465661287308D-09 0.191000000000D+03
0.172800000000D+06 0.000000000000D+00 0.000000000000D+00 0.000000000000D+00

36
gAGE
research group of Astronomy and Geomatics Range variation: Ionospheric correction
Klobuchar model
STEC: I ON SLANT
TEC: I on VER T
gAGE/UPC

−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

38
gAGE

39
gAGE
gAGE/UPC
Master of Science in GNSS

Backup @ J. Sanz & J.M. Juan
40
gAGE
Tropospheric Delay
Troposphere is the atmospheric layer placed between Earth’s surface
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:
Trop sat
= (d dry + d wet ) ⋅ m(elev) d dry= 2.3exp(−0.116 ⋅10−3 H ) meters
rec
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 ρ sat
0, rec − c ( 
dt sat
+ ∆rel sat
) + Trop sat
rec + Ion sat
1rec + TGD sat
41
gAGE
1.001
m(elev) =
0.002001 + sin 2 (elev)
gAGE/UPC

42
gAGE
research group of Astronomy and Geomatics Range variation: Tropospheric correction
Trop SLANT
gAGE/UPC
Trop VER TiCAL

RTCA-Do229C model

43
gAGE

44
gAGE

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:
K=sat
1, rec K1,rec + TGD sat
f12
K= K 2,rec + 2 TGD sat
sat
2, rec
f2
• K1rec may be assumed as zero (including it in receiver clock offset).

• TGDsat is transmitted in satellite’s 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]
of = ρ0,satrec − c ( dt sat + ∆rel sat ) + Troprec
Science in GNSS
sat
+ Ion1rec
sat
+ TGD sat
gAGE
gAGE/UPC Range variation: Instrumental delays (TGD)

47
gAGE

48
gAGE

49
gAGE
Measurement noise (thermal noise)
Antispoofing (A/S):
The code P is encrypted to Y. Wavelength σ noise Main
 Only theGPS
code Csignal
at (chip-length) (1% of λ) [*] characteristics
frequency L1 is available.
Code measurements
C1 300 m 3m
Unambiguous
P1 (Y1): encrypted 30 m 30 cm
but noisier
P2 (Y2): encrypted 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
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 λ/4  up to 5 cm for L1 and L2 [theoretically]
Typically: less than 1 cm
51
gAGE
gAGE/UPC Receiver and multipath noise
GPS standalone (C1 code) 10,000 €

52
gAGE
research group of Astronomy and Geomatics Receiver and multipath noise
Same environment!
gAGE/UPC
GPS standalone (C1 code) 100 €

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 = ρ0,satrec − c ( dt sat + ∆rel sat ) + Troprec
[modelled] 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
[modelled] sat
+ Ion1rec
sat
+ TGD sat
56
gAGE
1. Selection of orbital elements: From file brdc2860.98n,

select the last transmitted navigation message block before instant

t=38230 s (10h37m10s).
Transmission time:
979 208818  10h 0m 18s
PRN
14 98 10 13 12 0 0 +5.65452501178E-06 +9.09494701773E-13 +0.000000000000E+00
+1.28000000000E+02 -6.10000000000E+01 +4.38125402624E-09 +8.198042513605E-01
-3.31364572048E-06 +1.09227513894E-03 +5.67547976971E-06 +5.153795101166E+03
+2.16000000000E+05 -6.33299350738E-08 +1.00409621952E+00 -3.725290298462E-09
+9.73658001335E-01 +2.74031250000E+02
GPS week
+2.66122811383E+00 -8.081050495434E-09
GPS sec of week

-1.45720352451E-10 +1.00000000000E+00 +9.79000000000E+02 +0.000000000000E+00
+3.20000000000E+01 +0.00000000000E+00 -2.32830643654E-09 +1.280000000000E+02
+2.08818000000E+05 +0.00000000000E+00 +0.00000000000E+00 +0.000000000000E+00
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
14 98 10 13 12 0 0 +5.65452501178E-06 +9.09494701773E-13 +0.000000000000E+00
+1.28000000000E+02 -6.10000000000E+01 +4.38125402624E-09 +8.198042513605E-01
-3.31364572048E-06 +1.09227513894E-03 +5.67547976971E-06 +5.153795101166E+03
+2.16000000000E+05 -6.33299350738E-08 +1.00409621952E+00 -3.725290298462E-09
+9.73658001335E-01 +2.74031250000E+02 +2.66122811383E+00 -8.081050495434E-09
-1.45720352451E-10 +1.00000000000E+00 +9.79000000000E+02 +0.000000000000E+00
+3.20000000000E+01 +0.00000000000E+00 -2.32830643654E-09 +1.280000000000E+02
+2.08818000000E+05 +0.00000000000E+00 +0.00000000000E+00 +0.000000000000E+00
t = 38230 sec
t0= 12h 0m 0s= 43200 s
dt sat = a0 + a1 (t − t0 ) + a2 (t − t0 ) 2 = 5.65 ⋅10−6 s

gAGE/UPC
sat
[modelled] sat
+ Ion1rec
sat
+ TGD sat
58
gAGE
3. Satellite-receiver geometric range computation:
Use the following values (4789031, 176612, 4195008) as
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:
X= 11453479.346
T[emission]= Y= 122468524.004
38229.921 sec
 Orbit.f  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 “ωE ∆t” during
traveling time ∆t=T[reception]-T[emission].
gAGE/UPC
(Xsat,Ysat,Zsat)CTS[reception] =R3(ωE ∆t).(Xsat,Ysat,Zsat)CTS[emission]

61
gAGE
research group of Astronomy and Geomatics (Xsat,Ysat,Zsat)CTS[reception] =R3(ωE ∆t).(Xsat,Ysat,Zsat)CTS[emission]
 11453350.377   cos(ωE ∆t ) sin(ωE ∆t ) 0  11453479.346 

    
 122468589.797 =
  − sin(ω E ∆t ) cos(ω E ∆t ) 0  122468524.004 
 8245076.145    8245076.145 
   0 0 1  
CTS [ reception ] CTS [ emission ]
ωE ∆t =−5.74 ⋅10−6 rad . ( where ωE =7.2921151467 ⋅10−5 rad / sec)

ρ0,satrec
∆t =− =−0.079sec
c
( x sat − x0,rec ) + ( y sat − y0,rec ) + ( z sat − z0,rec )  23616673.3m

2 2 2
ρ0,satrec =
( 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.
62
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 − y0,rec ) + ( z − z0,rec ) = 23616699.124m

2 2 2
ρ satellite
0, receiver = sat sat sat
( 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] = ρ0,satrec − c ( dt sat + ∆rel sat ) + Troprec
of Science in GNSS
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)

14 98 10 13 12 0 0 +5.65452501178E-06 +9.09494701773E-13 +0.000000000000E+00
+1.28000000000E+02 -6.10000000000E+01 +4.38125402624E-09 +8.198042513605E-01
-3.31364572048E-06 +1.09227513894E-03 +5.67547976971E-06 +5.153795101166E+03
+2.16000000000E+05 -6.33299350738E-08 +1.00409621952E+00 -3.725290298462E-09
+9.73658001335E-01 +2.74031250000E+02 +2.66122811383E+00 -8.081050495434E-09
-1.45720352451E-10 +1.00000000000E+00 +9.79000000000E+02 +0.000000000000E+00
+3.20000000000E+01 +0.00000000000E+00 -2.32830643654E-09 +1.280000000000E+02
+2.08818000000E+05 +0.00000000000E+00 +0.00000000000E+00 +0.000000000000E+00
TGD= -2.32830643654E-09 * c= -0.69801 m

gAGE/UPC
sat
[modelled] sat
+ Ion1rec
sat
+ TGD sat
64
gAGE5. Relativistic clock correction:
PRN e sqrt(a)
14 98 10 13 12 0 0 +5.65452501178E-06 +9.09494701773E-13 +0.000000000000E+00

+1.28000000000E+02 -6.10000000000E+01 +4.38125402624E-09 +8.198042513605E-01
-3.31364572048E-06 +1.09227513894E-03 +5.67547976971E-06 +5.153795101166E+03
+2.16000000000E+05 -6.33299350738E-08 +1.00409621952E+00 -3.725290298462E-09
+9.73658001335E-01 +2.74031250000E+02 +2.66122811383E+00 -8.081050495434E-09
-1.45720352451E-10 +1.00000000000E+00 +9.79000000000E+02 +0.000000000000E+00
+3.20000000000E+01 +0.00000000000E+00 -2.32830643654E-09 +1.280000000000E+02
+2.08818000000E+05 +0.00000000000E+00 +0.00000000000E+00 +0.000000000000E+00
T[em ission] = E = 0.095 rad.

38229.921 s
 Orbi t .f  (eccentric anom aly)
µa =µ 3.986005 ⋅1014 m3 s −2
∆rel sat =
−2 −2.3 ⋅10−10 s
e sin( E ) =
c2 c = 299792458 m s −1
gAGE/UPC
sat
[modelled] sat
+ Ion1rec
sat
+ TGD sat
65
gAGE
6. Tropospheric correction
( ddry + d wet ) m(elev) =

sat
Troprec = 6.76m
−0.116⋅10−3 H
d dry = 2.3m
2.3e
See klob.f
d wet = 0.1m
1.001 π
m(elev) = =elev =
20.57 0.359rad
0.002001 + sin (elev)
2 180
H = 160m (heigh over the ellipsoid )
(x,y,z)rec [car2geo]  (Lon, Lat, H)rec

gAGE/UPC
sat
[modelled] sat
+ Ion1rec
sat
+ TGD sat
66
gAGE
7. Ionospheric correction
(time, rsta, rsat,α0,α1,α2,α3,β0,β1,β2,β3) [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
+1.2288E+05 -1.6384E+04 -2.6214E+05 +1.9661E+05 ION BETA
-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
[modelled] sat
+ Ion1rec
sat
+ TGD sat
67
gAGE
7. Compute the modeled pseudorange.
sat
C1rec = ρ rec
[modelled] sat
,0 − c ( 
dt sat
+ ∆rel sat
) + Trop sat
rec + Ion sat
1rec + TGD sat
ρ0,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
 rec [modelled] = 23615021.689m
sat
Trop sat
= 6.760 m C1
rec
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 =C1rec
sat
− C1[mod]rec
sat
=ρ rec
sat
− ρ0,satrec + c dtrec + K1rec + ε
In the previous example (PRN14 at t = 38230 s):

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

69
gAGE
Contents

gAGE/UPC

70
gAGE
research group of Astronomy and Geomatics 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…
= ρ rec + c ⋅ ( dtrec − dt sat ) + ∑ δ k + ε

sat sat
C1rec
Linearising ρ around an ‘a priori’ receiver position ( x0,rec , y0,rec , z0,rec )

= ρ0,satrec + ∆xrec + ∆yrec + ∆zrec + c ( dtrec − dt sat ) + ∑ δ k
ρ sat
0, rec ρ sat
0, rec ρ sat
0, rec
where:
∆xrec =xrec − x0,rec ; ∆yrec =yrec − y0, rec ; ∆zrec = zrec − z0, rec
Prefit-residuals (Prefit)
sat
C1rec ∑δk
− ρ0,satrec + c dt sat −=
ρ sat
∆xrec +
ρ sat
∆yrec +
ρ sat
∆zrec + c dtrec
gAGE/UPC
0, rec 0, rec 0, rec
measurement computed unknown

72
gAGE For all sat.  x0,rec − x sat1 y0,rec − y sat1 z0,rec − z sat1 
in view 
 ρ0,satrec1 ρ0,satrec1 ρ0,satrec1
1

Prefit
11
x  rec 

 22 

 Prefit
Prefit 
0, rec
1  ∆yrec
rec 
 ........
........    ∆∆zzrec 
    rec 
 .......... 
Prefit
Prefit nn
x  ccdt
dtrec 
rec 
sat n
 1
Observations Geometry of rays

(measured/computed)
 T sat n
ρ0 rec
−
ρ0 sat
rec
n
ρ
ˆ 0 sat
gAGE/UPC
rec
 sat n
ρT0 rec
ρˆ T0 rec
sat n
≡ Unitary Line-Of-Sight
ρ sat 1
0 rec vector from receiver
to satellite
73
(x,y,z) coordinates
gAGE  x0,rec − x sat1 y0,rec − y sat1 z0,rec − z sat1 
 1
 ρ0,satrec1 ρ0,satrec1 ρ0,satrec1 
Prefit
11
x  rec 

 22 

 Prefit
Prefit 
0, rec
1  ∆yrec
rec 
 ........
........    ∆∆zzrec 
    rec 
 .......... 
Prefit
Prefit nn
x  ccdt
dtrec 
rec 
sat n
 1

(measured/computed)
 Prefit1   −ρˆ T0 rec
sat 1
1
 2
 T sat 2 
−ρˆ T0 rec
sat n  Prefit
 ........ 
  −ρˆ
=  0 rec
1

 ∆rrec 
c dt 
   .... ...   rec 
 Prefit 
n  −ρˆ T sat n 1
(e,n,u) coordinates  0 rec
 Prefit1   − cos el1 sin az1 − cos el1 cos az1 − sin el1 1   ∆erec 
 2  
gAGE/UPC
 ∆n 
  − 2 2
− 2 2
− 2
= 
Prefit cos el sin az cos el cos az sin el 1  rec 
 ........   ..........   ∆urec 
     
 Prefit  − cos el sin az − cos el cos az − sin el 1  rec 
n n n n n n
c dt

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

Backup @ J. Sanz & J.M. Juan
75
 x0,rec − x sat1 y0,rec − y sat1 z0,rec − z sat1 
COMMENTS:
 1
gAGE  ρ0,satrec1 ρ sat1
0, rec ρ sat1
0, rec 
 Prefit1  x   ∆xrec 
 2
− x sat 2 y0,rec − y sat 2
z0,rec − z sat 2  ∆y 
 0,rec 1
 Prefit  =  ρ 0,satrec2 ρ0,satrec2 ρ0,satrec2   rec 
 ........   ∆zrec 
 
   ..........   
 Prefit n  x  c dtrec 
− x sat n y0,rec − y sat n z0,rec − z sat n
 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 Earth’s centre ) to linearise the equations
2.- With the pseudorange measurements and the navigation
equations, compute the correction ( ∆xrec , ∆yrec , ∆zrec ) to have
improved estimates:= ( x , y , z ) ( x , y , z ) + ( ∆x , ∆y , ∆z )
gAGE/UPC
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 
 ρ0,satrec1 ρ0,satrec1 ρ0,satrec1
1

Prefit
11
x  rec 

 22 

 Prefit
Prefit 
0, rec
1  ∆yrec
rec 
 ........
........    ∆∆zzrec 
    rec 
 .......... 
Prefit
Prefit nn
x  ccdt
dtrec 
rec 
sat n
 1

(measured/computed)
Thence, the basic linearized GPS measurement

equation can be written as:
yy = G x
gAGE/UPC
This is a linear system with, in general, n ≥ 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
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  +ε ⇒ y Gp+ε
=
gAGE/UPC

2
=
⇒
         n 
    N ×1 N × 2 2 ×1
 yN  m xN + n  yN   N 
x 1
79
gAGE  y1  m x1 + n  y1   x1 1 
 y  mx +n y   x 1 m
 2
=
⇒  2   2  +ε ⇒ y Gp+ε
=
2
        n 
 
    N ×1 N × 2 2 ×1
 yN  m xN + n  yN   N 
x 1
This is an over-determined (incompatible) system of equations
(due to the measurement noise ε).
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 G=
T
y GT G p ⇒
= pˆ (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  Least Squares
=
2) p (G G ) G y ⇔ min=
ˆ T
y − y min  ∑ ( yi − yi ) 
ˆ −1 T
ˆ
 i =1  solution
where
yˆ = G pˆ
But, what is the physical meaning

of the least square solution?
yî What does it mean the condition
yi
 N 2
gAGE/UPC
y − yˆ min  ∑ ( yi − yî )  ?
min=
xi x  i =1 

81
What is the physical meaning of the least square solution?
gAGE
y What does it mean to minimize
 N 2
y − yˆ min  ∑ ( yi − yî )  ?
min=
 i =1 
yî
yi − yî
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
move in the direction of the y axis.
yî
yi
 N 2
min  ∑ ( yi − yî ) 
 i =1 
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 − yˆ
i)
2
i i i
2 2
1 N k
= k ∑ ( yi − yî )
2
VTOTAL
2 i =1 k
k
k
k
yî k k
k
yi
k
 N 2
min (VTOTAL ) ⇒ min  ∑ ( yi − yî ) 
 i =1 
gAGE/UPC
 xi
N

x
min=
y − yˆ min  ∑ ( y − yˆ )=
 ⇔ pˆ (G T G ) −1 G T y
2
i i
 i =1  @ J. Sanz & J.M. Juan
84
gAGE Let be the basic linearized GPS measurement equation:
y =Gx
• Least Squares solution:

 2
min  ∑ ( yi − yˆ i ) 
2
min Y=
− Yˆ
ˆ = (G G ) Gt y
−1
t  i 
x
Yˆ = AXˆ
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
ˆ = ( G W G ) G Wy
−1  2
min = y − yˆ W min  ∑ wi ( yi − yî ) 
2
t t
x  i 
yˆ = G xˆ
85
Weighted Least Squares solution:
gAGE
y The same, but assuming different spring constants wi:
1 1
Vi= wi ∆yi2= wi ( yi − yî )
2
2 2 wN
1 N
= ∑ wi ( yi − yî )
2
VTOTAL
2 i =1
yî w1 wj
wi
yi
w2
 N 2
min (VTOTAL ) ⇒ min  ∑ wi ( yi − yî ) 
 i =1 
gAGE/UPC
 N xi 2  x
min= y − yˆ W min  ∑ wi ( yi − =
yî )  ⇔ pˆ (G T WG ) −1 G T Wy
 i =1  @ 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
min  ∑ ( yi − yˆ i ) 
2
min Y=
− Yˆ
ˆ = (G G ) Gt y
−1
t  i 
x
Yˆ = AXˆ
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
ˆ = ( G W G ) G Wy
−1  2
min = y − yˆ W min  ∑ wi ( yi − yî ) 
2
t t
x  i 
yˆ = G xˆ
88
Assuming that measurements Y have
gAGE random errors with zero mean and 1/ σ y21 0 
variance σ2 , and assuming that error sources  
for each satellite are uncorrelated with W=  
error sources for any other satellite, the  0 1/ σ y2n 

following weighted matrix may be used:  
wi =
1
 σ y2 ↑ ⇒ wi ↓
σ 2 i
− σ +σ
yi •- σ σ
greater error  less weight σ •= 300Km= 1msec
•Dt
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 = Py -1
W =P y , thence the
Minimum Variance Solution is found:
xˆ = ( G P G ) G t Py−1 y
t −1 −1
gAGE/UPC
Pxˆ = ( G P G)
And the error covariance matrix t −1 −1
for the estimation X is: 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:
• the measurement Y(n)

y ( n) (i.e., at t = t )
n
• the prediction of the state X(n), ˆ ( n − 1)

ˆ − ( n) from previous estimation x
x
1. Weighted average:
Let’s assume, that we have the
 y ( n) = G ( n) x( n) prediction xˆ − (n) , with Pthence, −
 −
ˆ (n)
x
it can be used to add an

ˆ ( n) = x( n)
x 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

Pxˆ − ( n ) 
xˆ ( n)   I 
Master of Science in GNSS  
94
gAGE −1
 y ( n)   G ( n)   Py ( n ) 0 
 − =  x( n) W=
 0

Pxˆ − ( n ) 
xˆ ( n)   I   
And the following solution of the previous equation system can be

found with some elemental algebraic manipulations:
ˆ = ( G t Py−1 G ) G t Py−1 y
−1
x
Pxˆ = ( G P G)
t −1 −1
y
ˆ ( n)
x Pxˆ ( n ) G t ( n) Py−(1n ) y ( n) + Pxˆ−−1( n ) x
ˆ − ( n) 
 
gAGE/UPC
−1
Pxˆ ( n ) G ( n ) P G ( n ) + P
t −1 −1 
 y(n) xˆ − ( n ) 
Master of Science in GNSS J. Sanz Subirana, JM. Juan Zornoza, M. Hernández-Pajares,
95
gAGE
2.- Prediction
Scalar case:
Let’s xˆn −1 be the state at epoch n-1 with variance σ x2ˆ 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:
xˆn− = φ xˆn −1
Thence, existing a linear relation between xˆn −1 and xˆn− , the
variance of the prediction will be: 2
σ xˆ − = φ 2 σ xˆ2n−1 + q 2
=
n
gAGE/UPC
An additional term is added to

account for modeling error!

96
gAGE
Generalization to the vector case:
xˆn− = φ xˆn −1
σ x2ˆ − φ 2 σ x2ˆn−1 + q 2
=
n
xn → x( n)
φ → Φ( n) Φ( n) : transition matrix
σ x2n
→ Px ( n ) Q( n) : process noise matrix
q 2
→ Q( n)
ˆ − ( n=
x ) Φ( n − 1) ⋅ x
ˆ ( n − 1)
gAGE/UPC
Pxˆ − ( n=
)
Φ ( n − 1) ⋅ Pxˆ ( n −1) ⋅ Φ t
( n − 1) + Q( n − 1)

97
gAGE Kalman filter (see kalman.f)
Measurements
ˆ − ( n)
x y ( n)
Pxˆ − ( n ) Py ( n )
Estimation
Prediction
−
(n) Φ xˆ (n − 1)
xˆ= Pxˆ ( n ) ⋅ G t Py−(1n ) y (n) + Pxˆ−−1( n ) xˆ − (n) 
xˆ (n) =
 
−1
Pxˆ − ( n ) Φ Pxˆ ( n −1)Φt + Q
= = Pxˆ ( n ) G Py ( n ) G + Pxˆ − ( n ) 
t − 1 − 1
 
ˆ (0)
x
Pxˆ (0) ˆ ( n)
x
Pxˆ ( n )
gAGE/UPC
Initialization

98
gAGE Kalman filter (classical version)
Measurements
ˆ − ( n)
x y ( n)
Pxˆ − ( n ) Py ( n )
Estimation
Prediction −1
−
xˆ (n) = xˆ − (n) + K (n) ⋅  y (n) − G (n) ⋅ xˆ − (n) 
xˆ= (n) Φ xˆ (n − 1) −1
K (n) = Pxˆ − ( n ) ⋅ G (n) ⋅ G (n) ⋅ Pxˆ − ( n ) ⋅ G t (n) + Py ( n ) 
t
Pxˆ − ( n ) Φ Pxˆ ( n −1)Φt + Q

=  
[I K (n) ⋅ G (n)] ⋅ Pxˆ − ( n )
Pxˆ ( n ) =−
ˆ (0)
x
Pxˆ (0) ˆ ( n)
x
Pxˆ ( n )
gAGE/UPC
Initialization

99
gAGE
Contents

gAGE/UPC

100
gAGE Some simple examples to define matrices Φ and Q
ˆ − ( n=
) Φ( n − 1) ⋅ x
ˆ ( n − 1)
x
Pxˆ − ( n ) =Φ ( n − 1) ⋅ Pxˆ ( n −1) ⋅ Φt ( n − 1) + Q( 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 σ2Dt . In
these conditions, matrices have the form:
1  0 
   
1 0
Φ( n) =   Q( n) =  
 1   0 
   
0  σ DT
2
  
gAGE/UPC
Being σ DT process noise associated to clock offset (in some way, the
2
uncertainty in clock value).
Master of Science in GNSS -σ@ J. Sanz & J.M.σJuan

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

* * *
Φ=1  x(n)=x(n-1)
Q=0 (no prediction error)
gAGE/UPC
We can assure that, the next x(n) will be

the same as x(n-1).
ˆ − (=
x n) Φ(n - 1) ⋅ x
ˆ ( n − 1)
Pxˆ − ( n )= Φ(n - 1) ⋅ Pxˆ ( n −1) ⋅ Φ t (n - 1) + Q(n - 1) 102
gAGE
White Noise process N(0, σ)
From a given epoch is not possible

to predict the following one
-2σ
+σ
-σ
* * *
+σ
-σ
+2σ
Φ=0  x(n)=0 (zero mean)
Q=σ2 (prediction error noise)
gAGE/UPC
We only can assume that, the next x(n)

can be x(n)=0 with a confidence σ.
ˆ − (=
x n) Φ(n - 1) ⋅ x
ˆ ( n − 1)
Pxˆ − ( n )= Φ(n - 1) ⋅ Pxˆ ( n −1) ⋅ Φ t (n - 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   σ dx
2

   
0 σ dy
2
Φ( n) =   Q( n) =  
 0   σ dz2 
   
 0   σ DT
2

2) In case of a slow moving vehicle, coordinates2 may be modeled as
random walk, process’ spectral density q = dσ , and the clock as a
dt
white noise:
1   qdx ∆t 
   
1 qdy ∆t
Φ( n) =   Q( n) =  
gAGE/UPC
 1   qdz ∆t 
   
0  σ DT
2
  

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

dσ 2
∆t
dt
** *
ˆ − (=
x n) Φ(n - 1) ⋅ x
ˆ ( n − 1)
Pxˆ − ( n )= Φ(n - 1) ⋅ Pxˆ ( n −1) ⋅ Φ t (n - 1) + Q(n - 1)
gAGE/UPC
Φ=1  x(n)=x(n-1) (the same value is assumed)

Q=(dσ
Master of Science in GNSS/dt)*∆t (but, with prediction error noise increasing
2 @ 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
research group of Astronomy and Geomatics References
[RD-1] J. Sanz Subirana, J.M. Juan Zornoza, M. Hernández-Pajares, GNSS

Data processing. Volume 1: Fundamentals and Algorithms. ESA TM-
23/1. ESA Communications, 2013.
[RD-2] J. Sanz Subirana, J.M. Juan Zornoza, M. Hernández-Pajares, GNSS
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

119
gAGE
Thank you
gAGE/UPC

120

Lecture UPC 3.v0.0

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture UPC 3.v0.0

Uploaded by

Copyright:

Available Formats

gAGE

research group of Astronomy and Geomatics

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. Measurements modelling and error sources

Master of Science in GNSS @ J. Sanz & J.M. Juan

1. Measurements modelling and error sources

Master of Science in GNSS @ J. Sanz & J.M. Juan

Master of Science in GNSS @ J. Sanz & J.M. Juan

= ρ rec + c ⋅ ( dtrec − dt sat ) + ∑ δ k + ε

x0,rec − x sat y0,rec − y sat z0,rec − z sat

0, rec 0, rec 0, rec

measurement computed unknown

Master of Science in GNSS @ J. Sanz & J.M. Juan

1. Measurements modelling and error sources

Master of Science in GNSS @ J. Sanz & J.M. Juan

The pseudorange modeling is based in the GPS Standard

Master of Science in GNSS @ J. Sanz & J.M. Juan

Master of Science in GNSS @ J. Sanz & J.M. Juan

Euclidean distance between satellite coordinates at emission time

(x − x0,rec ) + ( y − y0,rec ) + ( z − z0,rec )

problem). But ....

Master of Science in GNSS @ J. Sanz & J.M. Juan

Thence, as it will be shown in next lesson, the navigation problem will

and iterate until the change in the solution estimates is sufficiently

•The GPS signal travels from

•The satellite can move

The receiver time-tags are

An algorithm is needed to compute the satellite

As it is known, the pseudorange measurements link the “emission

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

receiver clock (TR) (receiver time tags).

Thence, the emission time in the satellite clock is:

Finally, since dt S = t S –T is the time offset between satellite clock (tS)

T[ems]= tems(TS)-dtS = trec(TR)-(C1/c+dtS)

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

The algorithm provided by the GPS/SPS-SS (orbit.f) supplies satellite

coordinates in an Earth-Fixed reference frame. To compute the

2. Compute satellite coordinates at emission time T[ems]

T[ems] [orbit]  (Xsat,Ysat,Zsat)CTS[emission]

build the navigation equations).

(Xsat,Ysat,Zsat)CTS[reception] =R3(ωE ∆t).(Xsat,Ysat,Zsat)CTS[emission]

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

• They are time-offsets between satellite/receiver time and GPS

- The receiver clock offset (dtrec) is estimated together with receiver

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

dtsat=a0 + a1(t-t0) + a2(t-t0)2 + ∆rel

Master of Science in GNSS @ J. Sanz & J.M. Juan

2 NAVIGATION DATA GPS RINEX VERSION / TYPE

-5.178787145843D-11 1.000000000000D+00 8.230000000000D+02 0.000000000000D+00

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

• A constant component depending only on nominal value of satellite’s