In this paper we focus on the navigation processing module of a low cost GPS receiver, in order to study the GPS errors characterization and modeling. The navigation data are recorded using a low cost GPS receiver antenna, and post-processed using GPS post-processing algorithms used in RTKLIB C++ package. This method allowed us to study and simulate errors variation in the case of a single positioning mode. An important focus was given to the Dilution of Precision (DOP) characterization of GPS errors, on the computed solution by a software post-processing. The DOP parameter showed a good indication of the quality of the GPS positioning solution, when using good satellite geometry. For navigation processing, it is usually convenient to choose the lowest value of DOP (DOP < 4) in order to minimize the GPS errors. But often, this is to choose some satellites with low elevations angles. As a solution for this issue, the best case is obtained when choosing a minimum DOP with best and optimum elevation angle. The experience was carried out using a fixed antenna receiver, where the effects of other sources of errors such as multipath were reduced. In this case, we compare the performances of using all the satellites in view, to choosing the optimum geometry obtained using 4 satellites generated by a quasi-optimal algorithm. This two are employed to be compared and to calculate the constellation's DOP and the corresponding receiver's position error. Results show that it s better to deal with best satellites geometry, and an accurate correlation between DOP and resulting errors can be obtained