الفهرس 
GPS System and ErrorsOnly 14 pages are availabe for public view
 |  | from | 167 |  |  |
| | | from | 167 | | |
Abstract
Many surveying projects need high accuracy of GPS positioning, which is available when using expensive receivers and relative techniques that deal with two or more synchronized receivers. This technique rises the cost and the field effort. The other positioning technique of GPS is the Single Point Positioning (SPP), which deals with only one receiver. The current technique reduces the cost and effort in the field, but the resulting quality is not suitable for many surveying requirements. The expected positional error reaches few meters.
A special case of SPP is the Precise Point Positioning (PPP) that deals with one receiver as SPP, considering other factors such as precise products and models to reduce the effect of different errors. This technique can reach the decimeters level accuracy especially when using dual frequency receivers. PPP needs scientific software or online services for post processing. Each scientific package provides different models to reduce the effect of GPS errors.
The current research aims to evaluate a special technique depending on single frequency GPS data, trying to raise the quality of PPP. To achieve the main goal of this thesis, three scientific software were used, which are GPS-S, RTKLIB and gLAB. Data of three different fields were used covering different regions in the world. The maximum distance between stations in any test field reaches about 5km.
The main objective of the first phase of the thesis was to select the optimum parameters and services which used in postprocessing stage to reach the highest possible quality of static PPP using gLAB and RTKLIB postprocessing software, and study the effect of code smoothing on the static PPP considering GPS-S in case of RTKLIB. The results and analysis prove that gLAB is better than RTKLIB with internal consistency of few centimeters and RMSE of few decimeters, and the code smoothing has a negative effect on the static PPP.
The main objective of the second phase was to test different mathematical models that produce positional corrections from permanent stations near the tested points to select and validate the optimums depending on synchronized data, and study the effect of these corrections on the static PPP positioning, considering the optimums of the first phase. The obtained results provided the 1st polynomial model as the optimum for gLAB and the EN solver model for RTKLIB. The results proved that the 1st polynomial case produces better results than the other case and has a positive effect on the static PPP positioning.
The role of the third phase is to apply the optimums of the previous two phases when producing corrections depending on non-synchronized data. The results showed the positive effect of the 1st polynomial non-synchronized corrections on the static PPP positioning when the data of four synchronized permanent stations are available within 7 days before or after the observational time of the tested points. This impression is not reflected on the EN solver case. In case of gLAB, if the time lag between the permanent stations and the unknown points observations are more than seven days, it is preferred to use the static PPP processing considering the optimums of gLAB software without any non-synchronized corrections.