Global Positioning System Reference
In-Depth Information
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
[25
Fi
gure 7.16 Example of real-time precise point positioning from Starfire at Redondo
an
d Melbourne on September 18, 2002.
The horizontal position error generally stays within
1 dm.
Lin
—
-
——
No
*PgE
time PPP described in Hatch et al. (2002) applies the model tropospheric refraction,
thus alleviating the need to estimate the troposphere by the user (which is adequate
since StarFire does not offer centimeter-level accuracy but global decimeter accuracy
in the horizontal, and perhaps slightly worse accuracy in the vertical). Figure 7.16
demonstrates the performance of StarFire.
[25
7.7 RELATIVE POSITIONING
In relative positioning, the vector between two stations is determined when two re-
ceivers observe simultaneously. If more than two receivers observe at the same time,
we speak of a session network consisting of all the co-observing stations. Session
solutions result in a set of correlated vectors between the stations. In relative posi-
tioning, one tends to use double- or triple-difference observations, although single-
difference observations could be used as well. Typically, one station is held fixed, i.e.,
coordinates
x
k
are known, and the coordinates
x
m
of the other station are estimated.
We discussed the details as to the accuracy required for the location of the known
station as a function of baseline length and satellite ephemeris accuracy in Section
5.35. The formulation given in that section can be readily generalized for the case
when both receivers occupying the endpoints of a baseline are in motion.
7.
7.1 Using Pseudoranges
In our customary notation the superscript
p
denotes the base satellite. If
S
satellites
are observed at the same instant, then the superscript
q
takes on
S
−
1 values, i.e.,
there are
S
−
1 independent double-difference equations (7.38),
