Analysis of the Relationship between Scintillation Parameters, Multipath and ROTI

Chendong Li, Craig M. Hancock, Nicholas A. S. Hamm, Sreeja Vadakke Veettil, Chong You2020 · DOI

TL;DR

This paper explores the relationship between ionospheric scintillation (a major GNSS error source) and more accessible parameters like multipath (MP) and rate of change of total electron content (ROTI). By analyzing data from equatorial and polar regions, the study demonstrates that MP and ROTI can indeed reflect scintillation activity, offering a potential pathway to monitor and mitigate these errors using standard GNSS receivers.

What Problem Does This Paper Address?

Global Navigation Satellite System (GNSS) performance, particularly positioning accuracy, is significantly degraded by ionospheric scintillation. This phenomenon, characterized by rapid fluctuations in signal amplitude (S4) and phase (σφ), can lead to signal loss and substantial positioning errors. While S4 and σφ are direct measures of scintillation, their generation requires specialized GNSS receivers, which are not widely available. This limits the global study and mitigation of scintillation. The paper addresses this by investigating whether more commonly available parameters, multipath (MP) and ROTI, can serve as proxies for scintillation, thereby enabling broader monitoring and research.

Methods

The study employs a multi-faceted approach to analyze the relationships between MP, ROTI, S4, and σφ:

  • Time Series Analysis: Visual inspection of time series plots to observe temporal correlations during scintillation events.
  • 2D Spatial Mapping: Generation of mean value maps and occurrence percentage maps to visualize spatial patterns and identify regions of high activity.
  • Structural Similarity Index (SSIM) and Pearson Correlation Coefficient (CC): Quantitative measures to assess the similarity and linear correlation between the spatial maps of different parameters.
  • Variograms and Cross-Variograms: Geostatistical tools used to investigate the spatial dependence and cross-correlation structure of the parameters.
  • Data Preprocessing: A 30-degree satellite elevation mask was applied to mitigate the influence of near-ground multipath, and data was normalized using P-Norm to allow for direct comparison of parameters with different scales.

Datasets

The study utilized GNSS data from three stations:

  • PRU2 and SAO0P: Located in Sao Paulo, Brazil (equatorial region).
  • SNA0P: Located in Antarctica (polar region).

Data was collected over six days in total, with parameters acquired at a 1-minute sampling interval. Specific dates used include:

  • September 8, 2017, and September 13, 2017 (SAO0P)
  • March 12, 2011 (PRU2)
  • April 2, 2016, October 13, 2016, and May 9, 2016 (SNA0P)

Key Insights and Ideas

  • MP and ROTI as Scintillation Proxies: The study provides strong evidence that MP and ROTI can indeed serve as indicators of ionospheric scintillation, especially when using standard GNSS receivers. This is a crucial finding for expanding scintillation monitoring capabilities.
  • Spatial and Temporal Co-occurrence: Scintillation events, characterized by high values of S4 and σφ, are consistently observed to occur at similar times and in similar spatial regions as elevated MP and ROTI values.
  • Complementary Nature of MP and ROTI: The research suggests that combining MP and ROTI might offer a more comprehensive representation of scintillation than either parameter alone, potentially capturing different aspects or types of scintillation.
  • Regional Differences: While relationships are observed in both equatorial and polar regions, the specific correlations and patterns can vary, highlighting the need for region-specific analysis. For instance, in equatorial data, MP correlated more with S4, while ROTI was more relevant to σφ.
  • Propagation Patterns: The study observed that MP, ROTI, S4, and σφ tend to share similar propagation directions, which could be valuable for predicting scintillation movement.

Key Results

  • High Correlation: For 70% of the outputs, the Structural Similarity Index (SSIM) and Correlation Coefficient (CC) between parameters were greater than 0.7. Many SSIM and CC values exceeded 0.6, with some reaching over 0.8, indicating a high degree of spatial and temporal similarity.
  • Spatial Structure: Variograms and cross-variograms revealed a common spatial structure for the parameters, with a range of approximately 5° to 6° for the Brazil data.
  • MP vs. S4 and ROTI vs. σφ: Generally, MP showed a stronger correlation with S4, while ROTI showed a stronger correlation with σφ, particularly in the equatorial region.
  • Antarctic Data Nuances: For Antarctic data, S4 was often not considered due to low values, but σφ, MP, and ROTI showed fluctuations and correlations, with MP and ROTI often correlating with σφ.

Limitations and Future Work

  • Limited Data: The study used only six days of data from three stations. A broader network with more comprehensive data is needed for more robust conclusions.
  • Separating MP Effects: While a 30° elevation mask was used, the paper acknowledges the need for more detailed studies to definitively separate real multipath effects from scintillation-induced MP.
  • Regional and Temporal Variations: The study suggests that relationships can vary between different dates and regions, indicating that further investigation into these dependencies is required.
  • Theoretical Basis: The authors suggest that future work should focus on investigating the relationship based on first principles and underlying physical formulas.
  • Mitigation Potential: While the study demonstrates the potential for using MP and ROTI for monitoring, further research is needed to explore how these relationships can be practically applied for mitigating scintillation errors.

Summary updated

BibTeX
@misc{li2020analysis,
  author = {Chendong Li and Craig M. Hancock and Nicholas A. S. Hamm and Sreeja Vadakke Veettil and Chong You},
  title = {{Analysis of the Relationship between Scintillation Parameters, Multipath and ROTI}},
  year = {2020},
  doi = {10.3390/s20102877},
}
Figure