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
Combining (8.8) and (8.10) gives the linearized form
t
+
u
+
s
u
+
Qu
−
x
=
o
(8.11)
For the purpose of distinguishing various approaches, we call the transformation
(8.8) model 1. The seven transformation parameters
(
)
can be
estimated by a least-squares. The Cartesian coordinates
u
and
x
are the observations.
Equation (8.11) represents a mixed model
f
(
∆
x,
∆
y,
∆
z, s, ε,
ψ
,
ω
o
. See Section 4.4 for addi-
tional explanations of the mixed model adjustment. Each station contributes three
equations to (8.8).
A variation of (8.8), called model 2, is
a
,
x
a
)
=
t
+
u
0
+
(
1
+
s)
R
(
u
−
u
0
)
−
x
=
o
(8.12)
[30
where
u
0
is the vector in the system
(u)
to a point located somewhere within the
network that is to be transformed. A likely choice for
u
0
is the centroid. All other
notation is the same as in Equation (8.8). If one follows the same procedure as
described for the previous model, i.e., omitting second-order terms in scale and
rotation and their products, then (8.12) becomes
Lin
—
1.3
——
Lon
*PgE
t
+
u
+
s (
u
−
u
0
)
+
Q
(
u
−
u
0
)
−
x
=
o
(8.13)
Model 3 uses the same rotation point
u
0
as model 2, but the rotations are about
the axes
(n, e, u)
of the local geodetic coordinate system at
u
0
. The
n
axis is tangent
to the geodetic meridian, but the positive direction is toward the south; the
e
axis is
perpendicular to the meridian plane and is positive eastward. The
u
axis is along the
ellipsoidal normal with its positive direction upward, forming a right-handed system
with
n
and
e
. Similar to Equation (8.12), one obtains
[30
t
+
u
0
+
(
1
+
s)
M
(
u
−
u
0
)
−
x
=
o
(8.14)
If
(
λ
0
,h
0
)
are the
geodetic coordinates for the point of rotation
u
0
, it can be verified that the
M
matrix is
η
,
ξ
,
α
)
denote positive rotations about the
(n, e, u)
axes and if
(ϕ
0
,
R
3
(
λ
0
)
R
2
(
90
M
=
−
ϕ
0
)
R
3
(
α
)
R
2
(
ξ
)
R
1
(
η
)
R
2
(
90
−
ϕ
0
)
R
3
(
λ
0
)
(8.15)
Since the rotation angles
(
η
,
ξ
,
α
)
are differentially small, the matrix
M
simplifies to
M
(
λ
0
,ϕ
0
,
η
,
ξ
,
α
)
= α
M
α
+ ξ
M
ξ
+ η
M
η
+
I
(8.16)
where
−
λ
0
0
sin
ϕ
0
cos
ϕ
0
sin
M
α
=
−
sin
ϕ
0
0
cos
ϕ
0
cos
λ
0
(8.17)
cos
ϕ
0
sin
λ
0
−
cos
ϕ
0
cos
λ
0
0