Report for NOAA/NGS on accuracy analysis of various NGS ionosphere estimation models
Advanced Search
Select up to three search categories and corresponding keywords using the fields to the right. Refer to the Help section for more detailed instructions.


This Document Has Been Replaced By:



This Document Has Been Retired


Up-to-date Information

This is the latest update:

Report for NOAA/NGS on accuracy analysis of various NGS ionosphere estimation models
  • Published Date:


Filetype[PDF-298.58 KB]

  • Alternative Title:
    Accuracy analysis of various NGS ionosphere estimation models
  • Description:
    "In this report, a preliminary demonstration of the accuracy analysis of several methods of estimating the ionospheric corrections, based on reference GPS permanently tracking network, is presented. The ionospheric reference 'truth' was selected as the double difference (DD) ionospheric correction, estimated from the dual frequency carrier phase data after fixing the DD ambiguities. The five ionospheric models that were tested are: 1. International GPS Service (IGS) global ionospheric map (GIM). IGS GIM is a combination of several different ionosphere models provided by IGS analysis centers (Schaer, 1999). 2. Absolute, smoothed pseudorange-based method -- denoted as P4. For details, see Wielgosz et al. (2003) 3. Absolute carrier phase-based model, decomposed from double differenced ionospheric delays, denoted as MPGPS. For details, see Kashani et al. (2004) 4. Absolute model based solely upon undifferenced dual frequency ambiguous carrier phase data -- NGSa (Smith, 2004) 5. Carrier phase double difference-based tomographic method -- NGSb (Spencer et al., 2004). Methods 1-4 assume that the ionosphere is an infinitesimal single layer, while method 5 considers the ionosphere as a 3D layer. The GPS data collected by the Ohio CORS stations -- COLB and SIDN, on August 31, 2003, were selected for the analysis. The data were divided into two one-hour sessions. The first (night) session includes the data collected between 08:00-09:00 UT (3-4 am local time), when the total electron content (TEC) reached its daily minimum; the second (daytime) session, from the period 17:00-18:00 UT (1-2 pm local time), corresponds to the highest daily TEC level. This session allocation allows assessing the influence of varying ionospheric conditions on the model comparison results. The average PDOP was 2.9 during the night session and 3.9 during the daytime session (this would be relevant for the future positioning tests). The data-sampling interval was 30 seconds, and a total of 120 epochs per session were processed. The satellite elevation angles for the night-time and the day-time tests are plotted in figures 3 and 14, respectively. Model 1 (GIM) provides absolute ionospheric delays in the IONEX format (grid). Thus, the ionospheric delays were interpolated for the user location using the linear interpolation. The Universal Kriging method was used to interpolate ionospheric delays provided by Models 2 (P4) and 3 (MPGPS). Ionospheric delays in Model 4 (NGSa) were derived from the orthogonal function coefficients for the radial profiles, and in Model 5 (NGSb), the ionospheric delays were obtained directly from the phase measurement data used in this test. The corrections were provided to us as L1 absolute slant ionospheric delays (one value per satellite per epoch) in a modified RINEX format, so no interpolation was applied. It should be noted that NGSb model, at this point, includes only two DD, since the method is based on the trajectory cross-over, and no interpolation was applied to derive corrections for other satellites. Once this is available, similar analysis, but for the station position coordinates, will be carried out"--Outline.
  • Place as Subject:
  • Main Document Checksum:
  • File Type:
  • Supporting Files:
    No Additional Files
No Related Documents.

You May Also Like: