3.1 Comparing the value of ambiguities 
The correctness of OTF ambiguity resolution 
can be confirmed by comparing the results from 
OTF and static initialization. Table 3 presents 
the resutls of the ambiguity resolution of BAS1- 
airborne GPS and BAS2-airborne GPS by OTF 
and static initialization Both methods give the 
same result. The distance between BASI and 
airborne GPS is 1.5km and the distance between 
BAS2 and airborne GPS is 12.8km Figure 1 
presents the 3D coordinate difference between 
the baseline resolved by OTF method and the 
static baseline resolved by Trimvec Plus in the 
static initialization period. From the figure, we 
can see that centimeter level accuracy can be 
obtained while ambiguities are correctly fixed. 
3.2 Comparing the kinematic baseline 
Because static initialization had been preformed 
before aircraft took off, kinematic baseline 
results from OTF method can be compared with 
the results from conventional kinematic 
positioning method. 3D coordinate difference 
from two methods and the distances between the 
aircraft and reference station are shown in figure 
2. The vertical line in the figure means the 
moment that ambiguities fixed. We can see from 
the figure that the difference is less that 10cm by 
using two methods while the distance between 
airborne GPS and reference station is 40km and 
the altitude of aircraft is 2500m(the height 
difference between airborne GPS and reference 
is larger than 2400m). It demonstrates that the 
OTF method described in this paper is effective 
for kinematic baseline long as 40km and height 
difference larger than 2400m. 
3.3 Comparing the kinematic figuration 
closure from different reference stations by 
means of OTF method 
Closure check is a common used method in 
static positioning. It can be used in kinematic 
positioning to check the confidence of GPS 
kinematic data. Kinematic figuration closure is 
formed by two kinematic baselines from two 
reference stations and a static baseline between 
reference stations. The closure error of the 
kinematic closure loop can be used to evaluate 
the confidence of kinematic data procession 
Figure 3 shows the 3D kinematic closure errors 
of loop BAS1-BAS2-airborne GPS with 
kinematic baseline from OTF method and loop 
distance. It can be seen from figure 3 that the 
closure errors are about 0.5m while the 
ambiguities of one set kinematic data are fixed 
and the closure errors are rapidly decreased to 
0.1m while the ambiguities of both set kinematic 
data are fixed It is remarkable that the closure 
distance is 115km including 12.8km which is 
distance between two reference stations. The 
average kinematic baseline length is about 50km 
and the altitude of aircraft is 2500m. It indicates 
that the OTF method described in this paper is 
effective for mid-range kinematic positioning. 
4 CONCLUSION 
It is shown from the data processing of airborne 
kinematic GPS positioning that the OTF method 
proposed in this paper is capable of rapid and 
efficient solving for carrier phase ambiguities for 
the case of long distance as 50 km and height 
difference of 2400 meters and break the 
limitation of short distance and low height 
difference of previous OTF methods. After 
complete fixing ambiguity, differential 
positioning accuracy better than one decimeter 
can be obtained. In addition, this paper also 
executes three evaluations for the reliability of 
OTF solution(see Table 3 and Figure 1,2,3). If 
systematic errors can be further reduced, this 
OTF method will be capable for the case of long 
distance as 100 km and height difference of 
10000 meters. 
7B-5-6