GPS: Statistical models to reproduce the pseudoranges in different scenarios and their reliability evaluation Original Research Article

The GPS (Global Positioning System) navigation process is based on the determination of distances between user and, at least, four satellites of the GPS constellation. These distances have inherent errors due to different sources. Due to these errors these distances are called pseudorange. The more aggressive one is the intentional degradation introduced by the Selective Availability (SA). The SA and other errors prevent GPS to be used in critical real-time applications that requires submetric accuracy, like aircraft final approaching and landing. Statistical modeling is being applied to the pseudorange in order to estimate the true satellite-user ranges, and therefore to reduce errors and to obtain information that allows service integrity monitoring. A statistical pseudorange modeling methodology based on the AutoRegressive Integrated Moving Average (ARIMA) and Reliability Evaluation, established on Confidence Interval (CI) for standard deviation is presented. This methodology was applied on three study cases: two of them with de SA activated (ON) and one with SA OFF. The modeling identified statistical characteristics of the pseudorange and more, in the case that the SA was off the pseudorange errors show a stationary-like behavior. This is an encouragement to continue this work and to test other techniques suitable for stationary data.