International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004 
ZTD Local Time (hh:mm) 
25 11:23 11:56 12:30 ..13:03 13:36 14:10 14:43 15:16 
— Forward | | 
Backward | 
v 
24 | 
| 
> 
ng \ | 
T | PN Lt , | 
v UM | Peru lj 
A m 
} 
4 
th Tropospheric Delay (m) 
Zeni 
nN 
  
  
RG te dt rene grrr ete Talon LT 
401000 403000 405000 407000 409000 411000 413000 415000 
GPS Time (s) 
Figure 4: Zenith Tropospheric Delay 
Position Error (Backward Pass) Local Time (hh:mm) 
1123 11:56 12:30 13:03 13:36 14:10 14.43 15:16 
Fed thle I E Fei. —P3— i mec etc A n 
Error (m) 
  
| 
-02 | [== Latitude (RMSE = 0.024 m) 
| Longilude (RMSE = 0.022 m) 
| —— Height (RMSE = 0.08 m) 
  
“401000 403000 405000 407000 409000 411000 413000 415000 
GPS Time (s) 
Figure 5: Position Errors between P? and JPL's GIPSY 
OASIS II Solution 
Table 1: Position Error Statistics (cm): P? vs JPL's 
GIPSY/OASIS II 
  
  
  
  
  
  
  
  
Latitude Longitude Height 
Mean 1.7 1.4 -4.3 
Std. Dev. ].7 1.7 6.8 
RMSE 2.4 22 8.0 
  
  
4.2 Aircraft — High Dynamics (-800 km/h) 
The second data set was flown on September 21, 2003, with a 
flight duration of approximately 3.75 h. Maximum aircraft 
speed was up to 810 km/h. Final orbit and clock products from 
the International GPS Service (IGS) were used, with resolutions 
of 15 minutes and 5 minutes, and stated accuracies of 5 cm and 
0.1 ns, respectively [IGS Data & Products, 2003]. 
The number of satellites and PDOP, aircraft altitude, and the 
forward and backward estimations of the zenith tropospheric 
delay and receiver clock offset are shown in Figures 6, 7, 8 and 
9, respectively. 
The results obtained using P^ software were then compared to 
the data supplier's own in-house multi-reference station DGPS 
solution and also with JPL's GIPSY/OASIS II solution. The 
comparison with the in-house DGPS solution is shown in 
Figure 10, and with JPL's solution in Figure 11. 
Given in Table 2 are the position error statistics for the 
comparison between P" and the in-house DGPS solution while 
848 
given in Table 3 are the position error statistics for the 
comparison between P and JPL's GIPSY/OASIS II solution. 
There appears to be a relatively strong correlation between the 
height error and the receiver clock offset when comparing the 
P^ solution to the in-house DGPS solution, as shown in Figure 
10. Furthermore, a large offset appears for the height 
component. Both these effects are significantly reduced in 
Figure 11 when the two PPP solutions are compared, suggesting 
that the DGPS solution is less accurate compared to the two 
PPP solutions. This is also reflected in the statistics shown in 
Tables 2 and 3. The increased height error in the latter one-third 
of the data set is probably due to the higher PDOP, which is 
shown in Figure 6. 
NumSVs.& PDOP Local Time (hh:mm) 
  
  
  
  
  
> 23:36 00:10 00:43 01:16 01:50 02:23 
f ï ; 
7} 
6} 
a 
Ó 
a 
2 5 
1 tes | 
à — IST] | 
z PDOP | 
o - 
35 
E 
z 
3i 
2i 
1 i i ; 
13000 15000 17000 19000 21000 23000 
GPS Time (s) 
Figure 6: Number of Satellites and PDOP 
Height Local Time (hh'mm) 
3 4 
10500 ; 23:36 00:10 00:43 01:16 01:50 un 
10000| 
9500| 
9000 | 
Height (m) 
8500! 
8000 
  
7500: i i i 1 il 
13000 15000 17000 19000 21000 23000 
GPS Time (s) 
Figure 7: Altitude 
Local Time (hh: mm) 
_00:10 0043 0116 01:50 022 
| x Forward | : 
L—— Backward | 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
Zenith Tropospheric Delay (m) 
  
13000 15000 17000 19000 - 21000 - 23000 
GPS Time (s) 
Figure 8: Zenith Tropospheric Delay 
Int