Carrier-Based Differential Positioning. Ambiguity Resolution Techniques

gAGE
Lecture 7
research group of Astronomy and Geomatics
Carrier-based Differential
Positioning.
Ambiguity Resolution
Techniques
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.
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
Linear model for Differential Positioning
Single difference (•) ruj ≡ ∆(•) ruj =•
( )uj − (•) rj
Pruj = ρ ruj + c δ tru + Truj + I ruj + K ru + ν P ruj

Lruj = ρ ruj + c δ tru + Truj − I ruj + λ ωruj + λ N ruj + bru + ν L ruj
Double difference
(•) rujk ≡ ∇∆ (•) rujk = (•) kru − (•) ruj = (•)uk − (•) rk − (•)uj − (•) rj 
Pruk = ρ ruk + c δ tru + Truk + I ruk + K ru + ν P ru

k

User Reference
P jk
ru = ρ jk
ru +T jk
ru +I jk
ru +ν jk
P ru
Station
Now are cancelled:

gAGE/UPC
• Receiver clock
The same for carrier : • Receiver code instrumental delays
Lrujk = ρ rujk + Truj − I rujk + λ ωrujk + λ N rujk + ν L rujk • Receiver carrier instrumental
delays
Master of Science in GNSS
Carrier ambiguities N are integer
@ J. Sanz & J.M. Juan
4
gAGE
Single-Difference of measurements
(corrected by geometric range!!)
PRN06
PRN06
PLAN-GARR: 15km
PRN30
∆( L1 − ρ ) ≡ L1ru
sat
− ρ rusat PRN30 ∆ ( P1 − ρ ) ≡ L1ru
sat
− ρ rusat
Dif. Wind-up: Very small
∆( L1 − ρ ) ≡ Lruj − ρ ruj= c δ tru + Truj − I ruj + λ ωruj + λ N ruj + bru + ν L ruj
∆( P1 − ρ ) ≡ Pruj − ρ ruj= c δ tru + Truj + I ruj + K ru + ν P ruj

Dif. Instrumental
gAGE/UPC
Dif. Receiver clock: delays and carrier

Main variations Common Dif. Tropo. and Iono. : ambiguities:
for all satellites Small variations constant
5
gAGE
Double-Difference of measurements
(corrected by geometric range!!)
PLAN-GARR: 15km PLAN-GARR: 15km

PRN06-PRN30 PRN06-PRN30
∇∆L1 − ∇∆ρ ≡ L1ru

sat
− ρ rusat ∇∆P1 − ∇∆ρ ≡ P1ru
sat
− ρ rusat
Dif. wind-up: negligible
∇∆( L1 − ρ ) ≡ Lrujk − ρ=jk

ru Trujk − I rujk + λ ωrujk + λ N rujk + ν L rujk
∇∆( P1 − ρ ) ≡ Prujk − ρ rujk = Trujk + I rujk + ν P rujk Carrier ambiguities:

gAGE/UPC
constant
Dif. Tropo. and Iono. :
Small variations
6
gAGE
Linear model for Differential Positioning
Single difference (•) ruj ≡ ∆(•) ruj =•
( )uj − (•) rj

Lruj = ρ ruj + c δ tru + Truj − I ruj + λ ωruj + λ N ruj + bru + ν L ruj
where:
ρ=j
ru ρ0 ruj − ρˆ 0uj ⋅ ∆rru − ρˆ 0 ruj ⋅ ε site + ρˆ 0 ruj ⋅ ε eph
j
Double difference
(•) rujk ≡ ∇∆ (•) rujk = (•) kru − (•) ruj = (•)uk − (•) rk − (•)uj − (•) rj 
Prujk = ρ rujk + Trujk + I rujk + ν P rujk

Lrujk = ρ rujk + Truj − I rujk + λ ωrujk + λ N rujk + ν L rujk User Reference
Station
where: ρ rujk = ρ0 rujk − ρˆ 0ujk ⋅ ∆rru − ρˆ 0 rujk ⋅ ε site + ρˆ 0kru ⋅ ε eph

k
− ρˆ 0 ru
j
⋅ ε eph
j
gAGE/UPC
being: ρ0 rujk ≡ ρ0ujk − ρ0 rjk ; ∆rru ≡ ∆ru − ∆rr

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

ρ=j
ru ρ0 ruj − ρˆ 0uj ⋅ ∆rru − ρˆ 0 ruj ⋅ ε site + ρˆ 0 ruj ⋅ ε eph
j
where ρ ru= ρu − ρ r
j j j
Show that the Double Differences are given by:

ρ rujk = ρ0 rujk − ρˆ 0ujk ⋅ ∆rru − ρˆ 0 rujk ⋅ ε site + ρˆ 0kru ⋅ ε eph
k
− ρˆ 0 ru
j
⋅ ε eph
j
being: ρ0 rujk ≡ ρ0ujk − ρ0 rjk ; ∆rru ≡ ∆ru − ∆rr

gAGE/UPC

8
gAGE Linear model for Differential Positioning
(•) rujk ≡ ∇∆(•) rujk = (•) kru − (•) ruj = (•)uk − (•) rk − (•)uj − (•) rj 
Prujk = ρ rujk + Trujk + I rujk + ν P rujk

Lrujk = ρ rujk + Truj − I rujk + λ ωrujk + λ N rujk + ν L rujk
where: ρ rujk = ρ jk
− ρˆ 0 ujk ⋅ ∆rru − ρˆ 0 jkru ⋅ ε site + ρˆ 0kru ⋅ ε eph
k
− ρˆ 0 jru ⋅ ε 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 ; ωrujk  0 Note for baselines up to 10 km
the range error of broadcast
ρˆ 0 ru
j
⋅ ε eph
j
0 orbits is less than 1cm
ε site  0 ⇒ ∆rru  ∆ru (assuming ε eph j
 10 m ).
With these simplifications, we have:

gAGE/UPC
Prujk − ρ 0 rujk = −ρˆ 0 ujk ⋅ ∆ru + ν P rujk

Lrujk − ρ 0 rujk = −ρˆ 0 ujk ⋅ ∆ru + λ N rujk + ν L rujk
User Reference
Station
Remark: Pru − ρ 0 ru = Pu − ρ 0 u − Pr − ρ 0 r
Master of Science in GNSS jk jk jk jk jk jk
( ) @ 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
PRCP jk
ρ 0 r − Pr ;
= jk jk jk
PRCL jk
ρ 0 r − Lr
=
• The user applies this differential correction to remove/mitigate common
errors:
Pu jk − ρ 0 ujk + PRCP jk = −ρˆ 0 ujk ⋅ ∆ru + ν P rujk
Lujk − ρ 0 ujk + PRCL jk = −ρˆ 0 ujk ⋅ ∆ru + λ N rujk + ν 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.
Master of Science in GNSS jk jk jk jk jk

(
Remark: Pru − ρ 0 ru = Pu − ρ 0 u − Pr − ρ 0 r
jk
) @ 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 − ρ 0 ujk + PRCP jk = −ρˆ 0 ujk ⋅ ∆ru + ν P rujk

where ρˆ 0 ujk ≡ ρˆ 0ku − ρˆ 0 uj
Lujk − ρ 0 ujk + PRCL jk = −ρˆ 0 ujk ⋅ ∆ru + λ N rujk + ν L rujk
The previous system for navigation equations is written in matrix notation as:
 Pref P R ,1   0u
(
 − ρˆ 1 − ρˆ R T
0u
) 0  0 

   ∆ru 
 R ,1  ( )
T
− ρ0u − ρ0u
ˆ 1
ˆ R
1  0 
 Pref L   
λ R ,1 
=  N
 ru 
  
      
  − ρˆ nu−1 − ρˆ Ru
( )   
R , n −1
 Pref P
T
0  0 
λ N ru 
R , n −1
 Pref R ,n −1   0 0
   
( )
L T
 − ρ
ˆ n −1
− ρ
ˆ R
0  1 
0u 0u 
gAGE/UPC
where Pref R , k ≡ P R , k − ρ R , k + PRC R , k ∆ru

P u 0u P
Carrier
Pref L R ,k ≡ LRu ,k − ρ 0 uR ,k + PRCL R ,k ambiguities User Reference
Station

11
gAGE
Differential positioning
Broadcast SV Actual SV
Position Position
Calculated Measured Puser

Range ρref
Pref Pseudoranges
Differential Message Broadcast

PRC
Reference station
(known Location) User
– The reference station with known coordinates, computes differential

corrections: PRCP jk =
ρ rjk − Pr jk ; PRCL jk =
ρ 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 = σ P2 I PL = σ L2 I where, we can assume: σ P  50cm ; σ 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 =
−  j ; uncorrelated, so are they in single
 X ru   0 0 1 1 Xu
 j differences, but the noise is twice!
 X r 
σ X2 0 0 0  1 0 
  
1 −1 0 0   0 σ X2 0   −1 0 
gAGE/UPC
0
PX SD =  2σ 2
I
 0 0 1 −1 0 0 σX2
0   0 1  X
 2  
 0 0 0 σ X   0 −1

13
gAGE Correlations among the DD Measurements
• Now, the double difference (DD) and its covariance matrix can be
computed as:
 X k

 X ru  1 −1 0   j 
jk ru
 jl  =    X ru  ;
 X ru  0 −1 1   l 
 X ru 
 2σ X2 0 0  1 0 
1 −1 0     2σ 2 2 1
PX DD =
0 −1 1   0 2σ 2
X 0  −1 −1  X 1 2 
  2   
 0 0 2σ 
X  0 1 
Thence, even if the original measurements are uncorrelated, the

double differences are correlated.
gAGE/UPC
Note: The removal of the relative “User”—“Reference-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
Single and double differences of receivers/satellites
•
∇
• ≡ •−
k R
•
∆≡ •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.
15
gAGE
Single and double differences of receivers/satellites
•
∇
• ≡ •−
k R
•
∆≡ •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,...)
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
research group of Astronomy and Geomatics 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 ρuj

the satellite by orders of magnitude: ρ rj
ρ ruj =ρuj − ρ rj =−ρˆ uj ⋅ rru
ρˆ uj
Applying the same scheme to a rru= ru − rr Reference
second satellite “k”

Station
ρ ruk =ρuk − ρ rk =−ρˆ uk ⋅ rru

ρρ==
j
j
ru
ru ρ ρ
− ρ
u
u
j
j
− ρ r
r
j
j
Thence, the double differences of ru rr

ranges are: To Earth Centre
gAGE/UPC
ρ rujk = − ( ρˆ uk − ρˆ uj ) ⋅ rru =
ρ ruk − ρ ruj = −ρˆ ujk ⋅ rru

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

rru= ru − rr Reference
Station
ρ rujk = − ( ρˆ uk − ρˆ uj ) ⋅ rru =
ρ ruk − ρ ruj = −ρˆ ujk ⋅ rru ρ=j
ru ρuj − ρ rj
As commented before, for short baselines (e.g. Trujk  0 ; I rujk  0

less than 10km), we can assume that ephemeris
ωrujk  0
and propagation errors cancel, thence:
Prujk = ρ rujk + Trujk + I rujk + ν P rujk P=jk
ru ρ jk
ru + ν jk
P ru
Lrujk = ρ rujk + Truj − I rujk + λ ωrujk + λ N rujk + ν L rujk Lrujk =ρ rujk + λ N rujk +ν L rujk
Note that these equations allows

P jk
−ρˆ ⋅ rru + ν
= jk jk
a direct estimation of the
gAGE/UPC
ru u P ru
baseline, without needing an
−ρˆ ujk ⋅ rru + λ N rujk + ν L rujk
Lrujk = 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 ρ 19
gAGE
research group of Astronomy and Geomatics Relative Positioning
−ρˆ ujk ⋅ rru + ν P rujk
Prujk =
where ρˆ u jk ≡ ρˆ uk − ρˆ u j
−ρˆ ujk ⋅ rru + λ N rujk + ν L rujk
Lrujk =
The previous system for navigation equations is written in matrix notation as:
 − ( ρˆ 1 − ρˆ R )T 0  0  Baseline
 PrR,u,1   u u

 R ,1     rru  vector
 Lr ,u   − ( ρ
ˆ u
1
− ρ
ˆ u )
R T
1  0   R ,1 
 =  λ N
 ru 
  Satellite-j
      
 PrR,u, n −1   − ( ρˆ n −1 − ρˆ R )T 0  0    ρuj
λ N ru 
R , n −1
 R , n −1   u u
 ρ rj
 Lr ,u   − ρˆ n −1 − ρˆ R T 0  1 
 (u u )  ρˆ uj
rrur= r −r
ru= ruu− rr r
gAGE/UPC
Reference
Station
DD Code and
Carrier
Carrier
ambiguities ru rr
measurements ρ= ρuj − ρ rj
j
ru

20
gAGE

∇∆L1 ≡ L1ru
sat
∇∆P1 ≡ P1ru
sat
Satellite-j
∇∆L1 ≡ L = − ρ
jk
ru
jk
ru +T jk
ru −I jk
ru +λω +λ N
jk
ru
jk
ru +ν jk
L ru
ρuj
ρ rj
∇∆P1 ≡ Prujk = − ρ rujk + Trujk + I rujk + ν P rujk ρˆ uj

rru= ru − rr Reference
gAGE/UPC
Station
Variation of the Baseline ρ rujk =

−ρˆ ujk ⋅ rru ρ=j
ru ρuj − ρ 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
gAGE Relative Positioning
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
PRC L1
RRC dL1
dtrec ~ 660 m/s

RRC ~ 2cm/s
gAGE/UPC
d ρ up to ~800 m/s
Receiver: JAVAD TRE_G3TH DELTA3.3.12

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 ρ + dtrec
dtrec ~ 660 m/s

gAGE/UPC
d ρ up to ~800 m/s

Receiver: JAVAD TRE_G3TH DELTA3.3.12
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 dρ/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
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 1µs  ε dtsta < 1mm
gAGE/UPC
dtrec ~ 660 m/s

•Extrapolated to reduce error due to
d ρ up to ~800 m/s
latency: carrier can be extrapolated with
error < 1cm.
Exercise:
Demonstrate the following relationship

between the baseline and the differential ρuj
range [*]:
 2ρuj + rru  eˆ j
ρ =
ρ −ρ =
j
− jj j
 ⋅ rru rru / 2
ru
 ρu + ρuj + rru
u r
 rru= ru − rr
 
User Reference Station
Comments:
The previous expression can be written as:
2ρuj + rru ρuj + rru / 2
j
− (ω eˆ ) ⋅ rru
ρ −ρ = j j j
with ω ≡
j
eˆ = j
j
u r
ρuj + ρuj + rru ρu + rru / 2
• Taking rru = 0 in ω
j j
and eˆ leads to the approximate expression previously found.
gAGE/UPC
• ω j and eˆ 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:

ρuj
rru= ρ rj − ρuj
ρuj + rru / 2 ρ rj + ρuj eˆ j

=eˆ j
= rru / 2
ρuj + rru / 2 2ρuj + rru
rru= ru − rr
( )
2 2
2ρuj + rru eˆ j ⋅ rru= ρ rj − ρuj = User Reference Station
= (
ρ rj − ρuj )( ρ j
r + ρuj )
(
= ( ρ rj − ρuj ) ρuj + rru + ρuj )
Then:
− (ω j eˆ j ) ⋅ rru
ρuj − ρ rj =
gAGE/UPC
2ρuj + rru 2ρuj + rru

with: ω j ≡ ω eˆ = j
j j
ρuj + ρuj + rru ρu + ρuj + rru

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]
− ( ρˆ − ρˆ ) ⋅ rru + λ N
L =
jk
ru
k
u
j
u
jk
ru +ν 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:
 δ L12ru   ( u
 − δ −δ u )
ρ
ˆ 2
ρ
ˆ 1 T 

 13   
− (δ ρˆ uk − δ ρˆ uj ) ( )
T
δ Lrujk = ⋅ rru + δν jk δ
⇒  ru  
L − δ ρ
ˆ 3
− δ ρ
ˆ 1
 rru + ν
= L ru
u u
     
 1K   
δ
 ru   − δ ρˆ K − δ ρˆ 1 
 ( u u) 
L
gAGE/UPC
T
where:
δ 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]
 δ L12ru   ( u u)
 1 T 
− δ ρ
ˆ 2
− δ ρ
ˆ

 13   1 T 
− (δ ρˆ uk − δ ρˆ uj )
δ Lrujk = ⋅ rru + δν ⇒  ru   ( u
δ − δ ρ − δ ρ u) 
3
jk L ˆ ˆ
= L ru rru + ν
     
 1K   
δ Lru  
( ˆ uK − δ ρˆ 1u ) 
T
 − δ ρ 
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 = ρ ruj + c δ tru + λ N ruj + bru + ν L ruj Lruj =−ρˆ uj ⋅ rru + d ru + λ N ruj + ν L ruj
ρ ruj =ρuj − ρ rj =−ρˆ uj ⋅ rru where d ru ≡ c δ tru + bru
 L1ru   ( u )
 −ρˆ 1 T
1 Satellite-j
  λ N ru
1
 ρuj
 2     
 Lru   ( −ρˆ u )
2 T
1  rru   λ N ru2  ρ rj
= + +ν
      d ru   ρˆ uj
 K    K
gAGE/UPC
rru= ru − rr
 Lru   −ρˆ K T  λ N ru 
( u ) 1
Reference
Station
ρ=j
ru ρuj − ρ 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:
 ( −ρˆ 1 )T 1 Considering now differences between two
L  1 u
  λ N ru
1

  
ru
   epochs t0 and t1 , and assuming no cycle-slips:
=
L
  
2
( −ρˆ )
2 T
1  rru   λ N ru2 
+ν
+
ru u
      d ru     rru (t1 )   rru (t0 ) 
 K 
 Lru   −ρˆ K T
  K
 λ N ru 
L ru (t1 ) − L= (t
ru 0 ) G (t1 
)  − G (t 0 
)  + ν
( u )

1  ru 1 
d (t )  ru 0 
d (t )
L ru G δ rru=
(t1 ) rru (t1 ) − rru (t0 ) −G (t1 ) + G (t1 )
 − ( ρˆ 1 (t ) − ρˆ 1 (t ) )T 0
 u 1 u 0

 δ r (t )   r (t )   ˆ1
− ( ρu (t1 ) − ρˆ 1u (t0 ) )
T 
L ru (t0 ) G (t1 )  ru 1  + ( G (t1 ) − G (t0 ) )  ru 0  + ν
L ru (t1 ) −=  0
G (t1 ) − G (t0 ) =
δ d ru (t1 ) 
 
 d ru (t0 )  


 − ( ρˆ 1 (t ) − ρˆ 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 =ρ jk
ru +Tjk
ru +I jk
q , ru +ν jk
Pq , ru ; q = 1, 2 Truj  0 ρ rujk +ν P rujk ; q =
Pq jk,ru =
q
1, 2
jk
I 0
Lqjk,ru = ρ rujk + Truj − I qjk,ru + λq ωrujk + λ N qjk,ru + ν Ljkq ,ru ρ rujk + λq N qjk,ru +ν L
q , ru
Lqjk,ru = jk
ωrujk  0 q
ru
To simplify notation, when different frequencies are

considered, we will remove the subscript “ru”.
K sat. in view  K ‘SD’
 K(K-1) ‘DD’, ρ jk +ν Pjkq ; q =
Pqjk = 1, 2
but only K-1 DDs are
linearly independent ρ jk + λq N qjk +ν Ljkq
Lqjk =
We assume the following measurement errors:
gAGE/UPC
σ P ≈ 0.5 m σ P ≈ 1m Take the highest elevation as the

⇒
jk
reference satellite to minimize

q q
σ L ≈ 0.5cm
q
σ 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):
P= jk
ρ jk
+ ν jk
 L1jk − P1 jk 
1 P1
→ L − P1 = λ1 N +ν →
jk jk jk jk
N1 =
ˆ jk
 
L1jk =ρ jk + λ1 N1jk +ν Ljk1
1 1 P1
 λ1  roundoff
λ1  20 cm
σ P ≈ 1m 1
1
jk
σ Nˆ  σP ≈ 5
λ1
jk jk
σ L ≈ 1cm
jk
1 1
1
N +1
Fail Too much error (5 wavelengths)!
N + 1/ 2 Note that, assuming a Gaussian
Good N̂
N distribution of errors, σ Nˆ 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 ½ (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-Wübbena
combination
= N1 − N 2
NW
P= jk
ρ jk +ν Pjk1 f1 P1 jk + f 2 P2jk λW =
c
 86.2 cm
P= = ρ jk +ν PjkN
1 jk
f1 − f 2
f1 + f 2
N
P=
2 ρ +ν
jk jk jk
P2
f1 L1jk − f 2 L2jk σ P ≈ σ P / 2 ≈ 71cm

L = ρ jk + λW NWjk +ν LjkW
=
jk jk
ρ + λ1 N +ν
jk
L =
jk jk jk jk N 1
f1 − f 2 σ L ≈ 6 σ L ≈ 6 cm
1 1 L1 W
jk jk
ρ + λ2 N +ν
W 1
L =
jk jk jk jk
2 2 L2
↓
L − P = λW N +ν
jk jk jk jk
→  LWjk − PNjk 
W N W PN Nˆ jk
= 
W
 λW  roundoff
Fixing N1 (after fixing NW)
1 71cm
σ Nˆ ≈ σP   0.8
L − L = λ1 N
jk
1
jk
2 1
jk
− λ2 N 2
jk
+ν jk
L1 − L2
jk
W λW N
jk
86.2cm
( λ1 − λ2 ) N1jk + λ2 NWjk +ν Ljk1 − L2
=
Now, with uncorrelated
λ1 = 19.0 cm
 L1jk − L2jk − λ2 Nˆ Wjk  measurements from 10
gAGE/UPC
λ2 = 24.4 cm Nˆ jk
=  epochs will reduce noise
λ2 − λ1 =
1
 λ − λ  roundoff
5.4 cm 1 2
up to about ¼.
σ L ≈ 1cm
jk
1
1 1.4cm
σ Nˆ ≈ 2σ jk   1/ 4
Nˆ=jk
Nˆ 1jk in−GNSS
Nˆ Wjk 1
jk
λ1 − λ2 L1 5.4cm
Master
2 of Science @ 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
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= λi φi ).
f1 − f 2
c
Show that the corresponding wavelength is: λW =
f1 − f 2
Hint:
LW= λW φW ; φW = φ1 – φ2
2) Assuming L1, L2 uncorrelated measurements with equal noise σL,

show that:
γ 12 + 1
2
 f 
=σ LW = σ L ; γ 12  1 
γ 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 λ1 =0.190 m λ2 −λ1 =0.054 m having two close frequencies it
is possible to generate an
L2 120 x 10.23 MHz λ2 =0.244 m λW = 0.862 m
extra-wide-lane signal to
L5 115 x 10.23 MHz λ5 =0.255 m λEW =5.861 m enable the single epoch
ambiguity fixing.
We drop here the superscript (jk) for simplicity
c c c
= λi = ; λW ; λEW
=
fi f1 − f 2 f 2 − f5
Li =ρ + λi N i +ν Li ; i =
1, 2,5
γ 12 + 1 = γ ( = )
2 2
σL = σ  5, 7 cm 12 f1 / f 2 (77 / 60)
f1 L1 − f 2 L2 γ 12 − 1 L
LW = ρ λW NW +ν LW
=+ γ 25 (= f 2 / f5 ) (24 / 23) 2
W 1
=
2
f1 − f 2
γ +1
f 2 L2 − f5 L5 σL = 25 σ  33,3 cm
LEW = ρ + λEW N EW +ν LEW
=
EW
γ 25 − 1 L 1
N= N − N W 1 2
f 2 − f5
gAGE/UPC
f1 P1 + f 2 P2
γ 12 + 1 = N 2 − N5
N EW
σP = σ  0, 712 m
P= = ρ +ν PN N
γ 12 + 1 P
1
f1 + f 2
N
γ +1 Exercise:
f P + f5 P5 σP = 25 σ  0, 707 m
= 2 2
PEN = ρ +ν PEN γ 25 + 1 P Justify the previous
f 2 + fin5 GNSS
EN 1
expressions for σ. 47
Master of Science @ J. Sanz & J.M. Juan
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 =ρ + λ1 N1 +ν L1 σ L ≈ σ L ≈ 1cm
GPS Frequency Wavelengths Combinations 1 2
λ1 =0.190 m λ2 −λ1 =0.054 m

ρ + λ2 N 2 +ν L2
L2 = σ P ≈ σ P ≈ 1m
L1 154 x 10.23 MHz 1 2
L2 120 x 10.23 MHz λ2 =0.244 m λW = 0.862 m

ρ λW NW +ν LW
LW =+ P=
N ρ +ν PN
L5 115 x 10.23 MHz λ5 =0.255 m λEW =5.861 m
ρ + λEW N EW +ν LEW
LEW = = ρ +ν PEN
PEN
NW =
N1 − N 2 ; N EW =
N 2 − N5
=γ 12 (=
f1 / f 2 )
2
 L − PEN  1 0.71m (77 / 60) 2
Nˆ EW =  EW σ Nˆ ≈ σP   0.12
 λEW = γ 25 (=
f 2 / f5 )
2
 λ EW  roundoff
EW EN
5.861m (24 / 23) 2
γ 12 + 1
σP = σ  0, 71m
N
γ 12 + 1 P 1
 λEW Nˆ EW − ( LEW − LW )  1 33.3cm γ 25 + 1

NW = 
ˆ  σ Nˆ ≈ σL   0.39 σP = σ  0, 71m
 λ W  roundoff W λW EW
86.2 cm EN
γ 25 + 1 P 1
γ 12 + 1
gAGE/UPC
σL = σ  5, 7 cm
 L1 − L2 − λ2 Nˆ W  1 1.4cm
W
γ 12 − 1 L 1
N1 = 
ˆ σ Nˆ ≈ 2σ   1/ 4
 λ1 − λ2 γ 25 + 1
λ1 − λ2 5.4cm
1 L1
  roundoff σL = σ  33,3 cm
EW
γ 25 − 1 L 1

48
Exercise:
gAGE
Repeat the previous study for the Galileo signals E1, E5b and E5a
σ L ≈ σ L ≈ 1cm
Galileo Frequency Wavelengths Combinations

L1 =ρ + λ1 N1 +ν L1 1 2
E1 154 x 10.23 MHz λ1 =0.190 m λ2 −λ1 =0.058 m σ P ≈ σ P ≈ 1m

ρ + λ2 N 2 +ν L2
L2 =
E5b 118 x 10.23 MHz λ2 =0.248 m λW = 0.814 m
1 2
E5a 115 x 10.23 MHz λ3 =0.255 m λEW =9.768 m ρ λW NW +ν LW

LW =+ P=
N ρ +ν PN
LEW = = ρ +ν PEN
PEN
=γ 12 (=
f1 / f 2 )
2
 L − PEN  1 (77 / 59) 2
Nˆ EW =  EW σ Nˆ ≈ σP  [ ]
 λEW =γ 23 (=
f 2 / f3 )
2
EW EN
(118 /115) 2
γ 12 + 1
σP = σ [ ]
N
γ 12 + 1 P 1
 λEW Nˆ EW − ( LEW − LW )  1 γ 25 + 1
NW = 
ˆ  σ Nˆ ≈ σL  [ ] σP = σ [ ]
 λ W  roundoff W λW EW EN
γ 25 + 1 P 1
γ 12 + 1
σ [ ]
gAGE/UPC
σL =
 L1 − L2 − λ2 Nˆ W  1
W
γ 12 − 1 L 1
N1 = 
ˆ σ Nˆ ≈ 2σ  [ ]
 λ1 − λ2 γ 25 + 1
λ1 − λ2
L1
σ [ ]
1
  roundoff σL =
EW
γ 25 − 1 L 1

49
Exercise:
gAGE
Repeat the previous study for the Galileo signals E1, E5b and E5a
σ L ≈ σ L ≈ 1cm
Galileo Frequency Wavelengths Combinations

L1 =ρ + λ1 N1 +ν L1 1 2
E1 154 x 10.23 MHz λ1 =0.190 m λ2 −λ1 =0.058 m σ P ≈ σ P ≈ 1m

ρ + λ2 N 2 +ν L2
L2 =
E5b 118 x 10.23 MHz λ2 =0.248 m λW = 0.814 m
1 2
E5a 115 x 10.23 MHz λ3 =0.255 m λEW =9.768 m ρ λW NW +ν LW

LW =+ P=
N ρ +ν PN
LEW = = ρ +ν PEN
PEN
=γ 12 (=
f1 / f 2 )
2
 L − PEN  1 0.71m (77 / 59) 2
Nˆ EW =  EW σ Nˆ ≈ σP   0.07
 λEW = γ 23 (=
f 2 / f3 )
2
EW EN
9.768 m (118 /115) 2
γ 12 + 1
σP = σ  0, 71m
N
γ 12 + 1 P 1
 λEW Nˆ EW − ( LEW − LW )  1 54.9 cm γ 25 + 1

NW = 
ˆ  σ Nˆ ≈ σL   0.67 σP = σ  0, 71m
 λ W  roundoff W λW EW
81.4 cm EN
γ 25 + 1 P 1
γ 12 + 1
gAGE/UPC
σL = σ  5.4 cm
 L1 − L2 − λ2 Nˆ W  1 1.4 cm
W
γ 12 − 1 L 1
N1 = 
ˆ σ Nˆ ≈ 2σ   1/ 4
 λ1 − λ2 γ 25 + 1
λ1 − λ2 5.8cm
1 L1
  roundoff σL = σ  54.9 cm
EW
γ 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
) ρ 12 (ti ) + N q12 + υ L12q (ti )
 L12q (ti =
 Single frequency (K-1)* nt 3+(K-1)

 Lq (ti =
13
) ρ 13 (ti ) + N q13 + υ L13q (ti ) Dual frequency 2(K-1)*nt 3+2(K-1)
 q = 1, 2
  Kin. position Equations Unknowns
 L1K −1 (t = ρ 1K −1
+ 1K −1
+ υ 1K −1

 q i ) (t i ) N q Lq
(ti ) 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 ≥ 4, nt ≥ 2 K ≥ 5, nt ≥ 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
) ρ jk (ti ) + N qjk + υ Ljkq (ti )
Lqjk (ti = Ljk (ti ) − ρ 0 jk (ti ) = −ρˆ 0 jk (ti ) ⋅ ∆rru (ti ) + λ N jk + ν L jk (ti )
gAGE/UPC
Prefit-residual
ρ=
jk
(ti ) ρ jk
(ti ) − ρˆ 0 jk (ti ) ⋅ ∆r (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 ) + λ N + ν
of linear equations and solving it by the LS criterion.
52
gAGE Resolving Ambiguities as a set

y (ti )= G (ti ) ∆r (ti ) + λ N + ν (ti )
Single freq.: K=K-1


Dual freq.: K=2(K-1)
K  ×3
K 3 K
vector matrix vector vector
For static positioning, considering two epochs (for instance):

 y (ti )   G (ti )  I   ν (ti ) 
=
 y (t )  G (t )  ∆r + λ I  N +  ν (t ) 
 i +1   i +1     i +1 
In general, mixing several epochs, we will write: y = G ∆r + λ A N + ν

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 + λ 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
research group of Astronomy and Geomatics Search techniques
A conceptually simpler approach would consist on:
• Estimate the floated solution N and its uncertainty (e.g. Nˆ =2502347.74 cycles, σ Nˆ = 0.6 cycles)
(
• Define as a volume to be searched e.g. ± 3σ Nˆ  ±2 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 + λ A N + ν min y − G ∆r − λ A N Wy
To better exploit the internal correlations [*], we consider now the

covariance Wy = Py−1
Let be the float solution  ∆rˆ   ∆rˆ   P∆rˆ P∆rˆ ,Nˆ 

 ˆ  ; Cov  ˆ  =  
and covariance matrix: N   N   P∆rˆ ,Nˆ PNˆ 
It can be shown the following orthogonal decomposition:

2 2
y − G ∆r − λ A N = y − G ∆rˆ − λ A N + ∆r − ∆rˆ (N) +λ N −N
2 2
ˆ 2 ˆ
Wy Wy W∆rˆ ( N ) WNˆ
gAGE/UPC
Residual of real-valued
(
floated solution ∆rˆ , N
ˆ )
[*] 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 − λ A N = y − G ∆rˆ − λ A N + ∆r − ∆rˆ (N) + λ2 N − N
2 2
ˆ ˆ
Wy Wy W∆rˆ ( N ) WNˆ
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

Float solution and covariance matrix:
2
min N − N
ˆ → N
WNˆ
 
 ∆rˆ   ∆rˆ   P∆rˆ
 ˆ  ; Cov   = 
P∆rˆ ,Nˆ 


ˆ ˆ −1
∆r =∆r (N) =∆r − W∆rˆ ,Nˆ WNˆ N − N
ˆ ( )
N  Nˆ  P ˆ PNˆ 
 ∆rˆ ,N  
gAGE/UPC
The vectors ∆r and N are often

W∆rˆ ,Nˆ = P∆−rˆ1,Nˆ referred to as the fixed user solution
WNˆ−1 = PNˆ
and fixed ambiguity.

59
gAGE LAMBDA Method
The integer search: Finding the integer vector N that minimizes the cost function
( ) ( )
2 T
c(N) =N −N
ˆ =N − N
ˆ WNˆ N − N
ˆ
WNˆ = PNˆ−1
WNˆ
• 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.
N̂ 2
( ) ( )
2 2
1/ σ ˆ ˆ
2
0  = N1 − N ˆ
1 N 2 − N ˆ
2
c ( N ) +
WNˆ =  
N1N1
σ N2ˆ Nˆ σ N2ˆ Nˆ
 0 1/ σ 2
 N̂1
 
1 1 2 2
Nˆ 2 Nˆ 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
N′ = Z N back the results after finding the ambiguities
PNˆ ′ = Z PNˆ ZT
ˆ ′ = ZN
N ˆ 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:
−1
Z, Z ∈  ⇒ 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 = 
 pNˆ1Nˆ1 pNˆ Nˆ 
α1 1 
PNˆ =  1 2
 α i = − int  pNˆ Nˆ / pNˆ Nˆ 
 pNˆ1Nˆ 2 pNˆ Nˆ   
1 α 2 
1 2 i i
2 2 
Z2 =    Transforms N1 (N2 remains unchanged)
0 1 
gAGE/UPC
Note: Inverse matrices have also integer entries

Start transforming first the
 1 0 1 −α 2 
Z1−1 = 
−1
= 0 element with largest variance.
1 
 Z
 −α1 1 
2


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 = 
 pNˆ1Nˆ1 pNˆ Nˆ 
α1 1 
PNˆ = 
1 2
 α i = − int  pNˆ Nˆ / pNˆ Nˆ 
 pNˆ1Nˆ 2 pNˆ Nˆ   
2 2  1 α 2 
1 2 i i
Z2 =    Transforms N1 (N2 remains unchanged)

0 1 
Example: 1.05  53.4 38.4  Example

N=
ˆ
 PNˆ =   from [RD-4]
1.30  38.4 28.0 
Step 1:
1 −1 We transform first the element with
Z2 =  − int [38.4 / 28.0] =
 α2 = −1 largest variance (in this case N1)
 0 1 
1 −1 53.4 38.4   1 0   4.6 10.4  The half,
=PNˆ ′ Z=P Z T
0 1  38.4 28.0  =
    at most!
  −1 1  10.4 28.0 
2 Nˆ 2
 
In general,
Step 2: to increase the
 1 0
gAGE/UPC
Z1 =   − int [10.4 / 4.6] =

α1 = −2 number of small off-
 −2 1  diagonal elements,
we have to
 1 0   4.6 10.4  1 −2   4.6 1.2  transform first the
= PNˆ ′′ Z= P
1 N ˆ′ Z T
1  −2 1  10.4 28.0  =  1.2 4.8
Master of Science in GNSS    0 1    elements with
largest variance 64
Example and
1 −1  1 0 pictures from
gAGE Z2 =  Z1 =  
  −2 1  [RD-4]
0 1 
53.4 38.4   4.6 10.4   4.6 1.2 

PNˆ =   PNˆ ′ =   PNˆ ′′ =  
38.4 28.0  10.4 28.0  1.2 4.8
 1 0  1 −1  1 −1
=Z Z=  −2 1  0 =
1   −2 3 
Z
1 2
gAGE/UPC
 
 1 −1 53.4 38.4   1 −2   4.6 1.2 
=
PNˆ ′′ Z=
PN ZT   38.4 28.0   −=   
 −2 3    1 3  1.2 4.8
65
Example:
gAGE 1.05  53.4 38.4   4.6 1.2 
N=
ˆ
 PNˆ =   PNˆ ′′ =  
1.30  38.4 28.0  1.2 4.8
 1 −1
Z= 
 −2 3 
 1 −1 1.05   −0.25   −0.25  0 

=
ˆ ′′
N Z= ˆ
N  =
 1.30   1.80  = ′′
N int= 1.80   2 
 −2 3        
gAGE/UPC
−1  0   3 1  0   2 

=N Z= 2 2 =    
   1  2   2 

66
Let P be a symmetric and positive-definite matrix:
gAGE
1 N2
p p12  ( p11 + p22 + w ) N 2′
P =  11 λ1=
p22  N1′
2
 p12 1
λ2= ( p11 + p22 − w ) λ1
2 λ2 φ
w= ( p11 − p22 ) 2 + 4 p122 N1
λ 0
P′ =  1
0 λ2  tan 2φ =
2 p12
p11 − p22
Example:
53.4 38.4 
PNˆ =  ˆ = 1.05 
 N 1.30 
38.4 28.0   
=λ1 =
81.14 9.0
gAGE/UPC
81.14 0 
′
PNˆ =  λ2 =
=
0.25
0.25 0.5
 0 tan 2φ= 3.02 ⇒ φ= 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
y = G ∆r + λ A N + ν
The solution comprises the following steps:

 ∆rˆ   P∆rˆ P∆rˆ ,Nˆ 
; 
1. Obtain the float solution and its covariance matrix:  N
ˆ 
  P∆rˆ ,Nˆ PNˆ 
2. Find the integer vector N which minimizes the cost function

( ) ( ) WNˆ = PNˆ−1
2 T
c(N) =N −N
ˆ =N − N
ˆ WNˆ N −N
ˆ
WNˆ
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 − λ A N = G ∆r + ν
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 
=N′′ int
=  1.80   2 
   

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  Nˆ n  Using the triangular
  decomposition
N n −1 =  ˆ 
int  N n −1|n  =

ˆ
n−1 n
−2
(
int N n −1 − σ Nˆ , Nˆ σ Nˆ N n − N n 

n
ˆ
) PNˆ = LT D L

    lij = σ Nˆ σ N−ˆ2
( )
n
N1 =  ˆ  int  N1 − ∑ σ Nˆ , Nˆ σ Nˆ N i|I − N i 
int  N n|I  = ˆ −2 ˆ ˆ
j , N i|I i|I
 i =2
1 i|I i|I

gAGE/UPC
Nˆ 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 =
2 1 2
(
Nˆ 1 − σ Nˆ , Nˆ σ N−ˆ2 Nˆ 2 − N 2 )
Float
N2
Conditional
estimate
N̂1|2
N1
gAGE/UPC
  
= =
N 2 nint  Nˆ 2  0
  N= 1 nint  ˆ =
N 
 1|2  nint 

N1 N 2 , N1 N 2
2 (
ˆ − σ ˆ ˆ σ −ˆ2 Nˆ − N =
2
 1
 )

71
gAGE b3) Integer Least Squares (ISL) (from [RD-6])
1. The target to find the integer vector N which minimizes the cost function
( ) ( ) WNˆ = PNˆ−1
2 T
c(N) =N − N
ˆ =N − N
ˆ PNˆ−1 N − N
ˆ
PN−ˆ 1
2. The integer minimiser is obtained through a search over the integer grid
points on the n-dimensional hyper-ellipsoid:
( ) ( )
T
ˆ ≤ χ2
ˆ P ˆ−1 N − N
N −N N
• Where χ2 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 = σ N−ˆ2
i|I
Using the triangular decomposition: PNˆ = L D L li j = σ Nˆ σ N−ˆ2

T
ˆ
j , N i|I i|I
( N − Nˆ ) ( ) ( N − N ) ( )
T
ˆ ≤ χ2
L−1 D−1 L−T N − N
T
 ≤ χ2
D−1 N − N
Defining:
(N ) + (N ) (N )
gAGE/UPC
2 2 2
  
1 − N1 2 − N2 n − Nn
 =N − L−T (N − N
N  − N =(N
ˆ ) → LT N (
ˆ − N)
= )
c( N ) + + ≤ χ2
d1 d2 dn

72
gAGE
( ) ( ) (N ) + (N ) (N )
2 2 2
  
1 − N1 2 − N2 n − Nn
T

N −N D −1  ≤ χ2
N −N =
c( N ) + + ≤ χ2
d1 d2 dn
But N i depends on Nˆ i +1 , , Nˆ n .
N n = Nˆ n
(
 − N = (N
LT N ˆ − N) )
∑ (N )
n
N i =Nˆ i + j − Nˆ i l ji ; i =n − 1, n − 2, ,1
j = i +1
Search region bounds:

N n − d n1/ 2 χ ≤ Nn ≤ N n + d n1/ 2 χ
( ( ) ) ( ( ) )
2 1/ 2 2 1/ 2
N n −1 − d n1/−21 χ 2 − N n − Nˆ n dn ≤ N n −1 ≤ N n −1 + d n1/−21 χ 2 − N n − Nˆ n dn

1/ 2 1/ 2
 2 n  1/ 2  
( ) ( )
n
 χ − ∑ N j − N j d j  ≤ N1 ≤ N n + d1  χ − ∑ N j − Nˆ j
2 2
N 1 − d1/ 2
1
ˆ  2
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
( ) ( )
T
The size of the ellipsoidal search region N − N  ≤ χ 2 is controlled by χ2
D−1 N − N
Therefore, the performance of the search process is highly dependent on χ2:
• A small χ2 may result in a ellipsoidal region that fails to contain the solution.
• A too large value for χ2 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 χ2 :
• 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  Nˆ n 
 
N n −1 = nint  Nˆ n −1|n  =
 ( )
nint  Nˆ n −1 − σ Nˆ , Nˆ σ N−ˆ2 Nˆ n − N n 

based on the
n−1 n n
bootstrapping

estimator
   
( )
n
N1 = nint  Nˆ n|I  = nint  Nˆ 1 − ∑ σ Nˆ , Nˆ σ N−ˆ2 Nˆ 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 χ2 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 χ2 is removed.
the 2nd, 2rd, 4th and 5th nearest integers
Master of Science
Example in GNSS from [RD-6]
and pictures @ J. Sanz & J.M. Juan
75
Example and pictures
2nd nearest 3rd nearest
gAGE
research group of Astronomy and Geomatics 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
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.
( ) ( )
T
RMS =N − N
ˆ = N −N
ˆ PNˆ−1 N − N
ˆ
gAGE/UPC
PN−ˆ 1

77
gAGE
research group of Astronomy and Geomatics 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)
WNˆ−1
Q ≡ PNˆ =
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
azfixed=bootstrap(az,Lz);
gAGE/UPC
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-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.
[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
research group of Astronomy and Geomatics 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 3σ region
(i.e. ∆rˆ ± δ ).
1
y = G ∆r + λ N + ν N=
ˆ ( y − G ∆rˆ )
λ
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
∆rˆ
gAGE
The technique is based on exploiting the
constrains on the integer ambiguities: ∆rˆ − σ δ (k ) ∆rˆ + σ
• 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 ∆rˆ ± σ )
∆rˆ ± σ → N (=
k)
1
λ
( y − G ( ∆rˆ + δ ) )
(k ) → N(k )
Then, the corresponding position estimates ∆rˆ( k ) are computed.
(G G )
−1
N(k ) =
→ ∆rˆ( k ) T
G  y − λ 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
λ
( y − G ∆rˆ ) − N (k ) (k ) ∈ ?
86
gAGE
Contents
gAGE/UPC
- LAMBDA Method

87

Carrier-Based Differential Positioning. Ambiguity Resolution Techniques

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

Carrier-Based Differential Positioning. Ambiguity Resolution Techniques

Uploaded by

Copyright:

Available Formats

gAGE

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

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

Pruj = ρ ruj + c δ tru + Truj + I ruj + K ru + ν P ruj

Pruk = ρ ruk + c δ tru + Truk + I ruk + K ru + ν P ru

Pruj = ρ ruj + c δ tru + Truj + I ruj + K ru + ν P ruj

Now are cancelled:

Dif. Wind-up: Very small

∆( L1 − ρ ) ≡ Lruj − ρ ruj= c δ tru + Truj − I ruj + λ ωruj + λ N ruj + bru + ν L ruj

∆( P1 − ρ ) ≡ Pruj − ρ ruj= c δ tru + Truj + I ruj + K ru + ν P ruj

Dif. Receiver clock: delays and carrier

PLAN-GARR: 15km PLAN-GARR: 15km

∇∆L1 − ∇∆ρ ≡ L1ru

Dif. wind-up: negligible

∇∆( L1 − ρ ) ≡ Lrujk − ρ=jk

∇∆( P1 − ρ ) ≡ Prujk − ρ rujk = Trujk + I rujk + ν P rujk Carrier ambiguities:

Pruj = ρ ruj + c δ tru + Truj + I ruj + K ru + ν P ruj

Prujk = ρ rujk + Trujk + I rujk + ν P rujk

where: ρ rujk = ρ0 rujk − ρˆ 0ujk ⋅ ∆rru − ρˆ 0 rujk ⋅ ε site + ρˆ 0kru ⋅ ε eph

being: ρ0 rujk ≡ ρ0ujk − ρ0 rjk ; ∆rru ≡ ∆ru − ∆rr

Consider the Single Differences of geometric range:

Show that the Double Differences are given by:

being: ρ0 rujk ≡ ρ0ujk − ρ0 rjk ; ∆rru ≡ ∆ru − ∆rr

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

Prujk = ρ rujk + Trujk + I rujk + ν P rujk

With these simplifications, we have:

Prujk − ρ 0 rujk = −ρˆ 0 ujk ⋅ ∆ru + ν P rujk

Lujk − ρ 0 ujk + PRCL jk = −ρˆ 0 ujk ⋅ ∆ru + λ N rujk + ν L rujk

Where the carrier ambiguities N are integer numbers and

For larger distances, the atmospheric propagation effects (troposphere,

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

Master of Science in GNSS jk jk jk jk jk

Pu jk − ρ 0 ujk + PRCP jk = −ρˆ 0 ujk ⋅ ∆ru + ν P rujk

where Pref R , k ≡ P R , k − ρ R , k + PRC R , k ∆ru

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

Calculated Measured Puser

Differential Message Broadcast

– The reference station with known coordinates, computes differential

– The user receiver applies these corrections to its own measurements to

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

covariance matrix is diagonal:

PP = σ P2 I PL = σ L2 I where, we can assume: σ P  50cm ; σ L  5mm

Let X represent the code P or the Carrier L measurement.

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

Thence, even if the original measurements are uncorrelated, the

Note: The removal of the relative “User”—“Reference-station” common bias

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

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

case the mean value of unmodelled effects is absorbed by the receiver

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

baseline is shorter than the distance to ρuj

second satellite “k”

ρ ruk =ρuk − ρ rk =−ρˆ uk ⋅ rru

Thence, the double differences of ru rr

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

Thence, the double differences of ranges are: ρˆ uj

As commented before, for short baselines (e.g. Trujk  0 ; I rujk  0

Note that these equations allows

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