|
Fuzzy processing on GPS data to improve positioning accuracy, before and after S/A is turned off
The GDOP and the sum of Signal to Noise Ratio (SNR) are used as input fuzzy variables to the fuzzy processing unit. Fuzzy system output is defined as Reliable Factor. The main block diagram of fuzzy processing is shown in Fig.3 [12].
Fig.3. The main block diagram of fuzzy processing [12]
SNR and GDOP are divided into three and four segments for partition the rule space, respectively. R.F. is fuzzified with a singleton membership function. The membership functions are shown in Fig.4.
Fig.4. Membership functions: (a) GDOP, (b) SNR and (c) R.F. [12]
The approximate reasoning method is used for the inference process and the center of area method is employed for the defuzzification [13]. There are twelve rules in the rule base, which are shown in Table2.
Table2. Twelve rules in the rule base of proposed fuzzy system of Fig. 3 [12]
| R.F. |
SNR |
| S |
MS |
MB |
B |
| GDOP |
S |
S |
MS |
MB |
B |
| M |
S |
S |
MS |
MB |
| B |
S |
S |
S |
MS |
Experimental Results
At first, 3926 original data for a fix position were collected on the building of Computer Control and Fuzzy Logic Research Lab in the Iran University of Science and Technology. Then, the reliable factor of the fix position is obtained after the fuzzy processing.
As shown in Table3 and Table4, the reliable factor is compared with a desired value. If it exceeds the desired value, these fix positions are selected. By this principle, the appropriate fix positions can be selected from the original ones. Fuzzy selected positions have low value deviations.
Table3. Deviation x, Deviation y and Deviation z with desired Reliable Factor value (S/A on)
| R.F. value |
The number of fuzzy selected position |
D.x [m] |
D.y [m] |
D.z [m] |
| 0.0000 |
3926 |
196 |
219 |
190 |
| 0.1250 |
2271 |
160 |
219 |
89 |
| 0.2000 |
2185 |
160 |
219 |
83 |
| 0.2500 |
2182 |
160 |
135 |
83 |
| 0.3000 |
1694 |
160 |
125 |
83 |
| 0.4000 |
1689 |
160 |
125 |
83 |
| 0.4500 |
1524 |
160 |
125 |
73 |
| 0.4600 |
1257 |
146 |
121 |
73 |
| 0.4650 |
1052 |
135 |
105 |
72 |
| 0.4700 |
804 |
135 |
69 |
60 |
| 0.4750 |
198 |
77 |
49 |
42 |
| 0.4755 |
143 |
68 |
49 |
41 |
| 0.4756 |
129 |
67 |
49 |
41 |
| 0.4757 |
124 |
47 |
41 |
41 |
| 0.4758 |
118 |
47 |
41 |
41 |
| 0.4759 |
110 |
47 |
37 |
38 |
Table4. Deviation x, Deviation y and Deviation z with desired Reliable Factor value (S/A off)
| R.F. value |
The number of fuzzy selected position |
D.x [m] |
D.y [m] |
D.z [m] |
| 0.0000 |
3926 |
24 |
46 |
56 |
| 0.1250 |
2498 |
24 |
36 |
48 |
| 0.2000 |
1858 |
22 |
27 |
41 |
| 0.2500 |
1724 |
22 |
24 |
37 |
| 0.3000 |
684 |
21 |
21 |
35 |
| 0.3100 |
443 |
19 |
16 |
35 |
| 0.3125 |
342 |
14 |
16 |
15 |
| 0.3150 |
251 |
14 |
15 |
13 |
| 0.3200 |
222 |
14 |
13 |
11 |
| 0.3400 |
103 |
6 |
9 |
8 |
| 0.3550 |
15 |
4 |
7 |
8 |
According to Table3 and Table4, deviations of position components decrease due to increasing of Reliable Factor.
|
|
