Error Modeling for Differential GPS.

Abstract

Differential GPS (DGPS) positioning is used to accurately locate a GPS receiver based upon the well-known position of a reference site. In utilizing this technique, several errors sources contribute to position inaccuracy. This thesis investigates the error in DGPS operation and attempts to develop a statistical model for the behavior of this error. The model for DGPS error is developed using GPS data collected by Draper Laboratory. The Marquardt method for non-linear curve-fitting is used to find the parameters of a first order Markov process that models the average errors from the collected data. The results show that a first order Markov process can be used to model the DGPS error as a function of baseline distance and time delay. The model's time correlation constant is 3847.1 seconds (1.07 hours) for the mean square error. The distance correlation constant is 122.8 kilometers. The total process variance for the DGPS model is 3.73 meters square.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 27, 1995
Accession Number
ADA302794

Entities

People

  • Gregory S. Bierman

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Space

DTIC Thesaurus Topics

  • Accuracy
  • Artificial Satellites
  • Astronautics
  • Computational Science
  • Cross Correlation
  • Curve Fitting
  • Data Processing
  • Data Reduction
  • Data Sets
  • Global Positioning Systems
  • Information Processing
  • Information Science
  • Markov Models
  • Markov Processes
  • Navigation
  • Standards
  • Three Dimensional

Readers

  • Approximation Theory.
  • Computational Modeling and Simulation
  • Positioning, Navigation, and Timing (PNT) Technology.

Technology Areas

  • Space