Probabilistic approach to detect and correct GNSS NLOS signals using an augmented state vector in the extended Kalman filter

Changhui Jiang, Bing Xu, Li Ta Hsu

Research output: Journal article publicationJournal articleAcademic researchpeer-review

Abstract

Non-line-of-sight (NLOS) global navigation satellite system (GNSS) signals are a major factor that limits the GNSS positioning accuracy in urban areas. An advanced GNSS signal processing technique, the vector tracking loop (VTL), has been applied to NLOS detection and correction, and its feasibility and superior performance have been reported in recent studies. In a VTL-based GNSS receiver, the navigation solutions (i.e., position, velocity and time (PVT)) are used to predict the signal tracking loop parameters. The difference between the predicted signal and the received signal within the code discriminator output can be used to detect NLOS reception. We generate the probability of NLOS detection by modeling the code discriminator outputs using Gaussian fitting. If this probability is larger than a predefined threshold, NLOS reception is deemed to occur. Then, the NLOS-induced pseudorange measurement bias is estimated as a state variable in the state vector, i.e., an augmented state vector is created for the extended Kalman filter. Two GPS L1 C/A signal datasets from a static test and a dynamic test are investigated using the proposed algorithm. The experimental results indicate that when NLOS reception is present, the proposed approach outperforms the other two methods, i.e., the standard VTL method without considering NLOS reception and the VTL-based NLOS detection and correction method with multicorrelators, in terms of the positioning performance. In addition, the proposed approach has a lower computational load than the VTL method with multicorrelators.

Original languageEnglish
Article number72
JournalGPS Solutions
Volume25
Issue number2
DOIs
Publication statusPublished - Apr 2021

Keywords

  • Augmented state vector
  • Gaussian fitting
  • GNSS
  • Kalman filter
  • NLOS
  • Vector tracking loop

ASJC Scopus subject areas

  • Earth and Planetary Sciences(all)

Cite this