Asynchronous RTK Method for Detecting the Stability of the Reference Station in GNSS Deformation Monitoring

Yuan Du, Guanwen Huang, Qin Zhang, Yang Gao, Yuting Gao2020 · DOI

TL;DR

This paper introduces a novel asynchronous RTK (ARTK) method to accurately detect shifts in GNSS reference stations, a critical issue that can lead to misinterpretation of deformation monitoring data. By comparing ARTK with synchronous RTK (SRTK) results, the method reliably distinguishes between reference station movement and actual monitoring site displacement, enhancing the accuracy and trustworthiness of deformation monitoring systems.

What Problem Does This Paper Address?

Global Navigation Satellite Systems (GNSS), particularly Real-Time Kinematic (RTK) positioning, are widely used for monitoring ground deformation and geological hazards. RTK relies on a stable reference station to establish a local coordinate frame, allowing for precise measurement of relative displacements at monitoring sites. However, a significant challenge arises when the reference station itself becomes unstable due to factors like tectonic plate movement, subsidence, or even human activity. In such cases, any observed displacement at the monitoring station could be erroneously attributed to the monitoring site's movement, when in reality, it's the reference station that has shifted. This misinterpretation can lead to false alarms or missed critical events, undermining the reliability of deformation monitoring. Existing methods often struggle to detect these reference station shifts in near real-time, especially for short-term, high-rate monitoring applications.

Methods

The core innovation of this paper is the development of a strategy that combines both synchronous RTK (SRTK) and asynchronous RTK (ARTK) positioning.

  • SRTK: This is the standard RTK approach where the reference station's position from the current epoch is used to establish the reference frame for calculating the monitoring station's position.
  • ARTK: This novel approach uses the reference station's position from a previous epoch (e.g., a few minutes prior) to establish the reference frame for the current epoch's monitoring station position.

The key idea is that if the reference station is stable, the positions calculated by SRTK and ARTK for the monitoring station should be very close and consistent over time. However, if the reference station shifts between the "previous" epoch used by ARTK and the "current" epoch used by SRTK, a discrepancy will emerge between the ARTK and SRTK results for the monitoring station. This discrepancy directly indicates a shift in the reference station.

The paper defines a threshold based on the root mean square error (RMSE) of the difference between ARTK and SRTK positions. If the difference exceeds three times this RMSE for a sustained period, it signals a reference station drift. The ARTK method is chosen for its resilience to short-term reference station displacements, as its reference frame is anchored to a prior, potentially more stable, epoch.

Datasets

The paper utilizes two types of data for evaluation:

  1. Simulated GNSS Data: High-rate (1 Hz) GNSS data was generated using two receivers (CD01 and CD02). CD01 was fixed, while CD02 was placed on a moving platform to simulate slow displacement. This allowed for controlled testing of the method under known displacement scenarios, including scenarios where the "reference" station was intentionally moved.
  2. Actual Landslide Monitoring Data: Dual-frequency BeiDou data with a 1-second sampling interval was collected from the Heifangtai landslide in Gansu Province, China. This real-world data allowed for validation of the method in a complex geological environment during a critical event (landslide collapse).

Key Insights and Ideas

  • The "Ghost Displacement" Problem: The paper vividly illustrates how a shifting reference station can create the illusion of movement at the monitoring site, a critical flaw in traditional RTK.
  • ARTK as a Stability Anchor: By using a slightly older reference frame for ARTK, the method effectively "freezes" the reference station's position, allowing any subsequent shift to be clearly detected by comparing it to the current (SRTK) reference frame.
  • Detecting the "Invisible": The ARTK/SRTK comparison acts as a built-in quality control for the reference station's stability, a feature often missing or less effective in standard RTK.
  • Short-Term Focus is Key: The method is specifically designed for detecting short-term, centimeter-level shifts, making it ideal for high-rate deformation monitoring and early warning systems.

Key Results

  • Simulated Data Validation: The method successfully detected simulated shifts in the reference station, distinguishing them from movements of the monitoring station. The difference between ARTK and SRTK results clearly indicated when the reference station moved, and this detection occurred within a short timeframe (e.g., 15 seconds in one simulation).
  • Landslide Monitoring Success: During the final collapse phase of the Heifangtai landslide, the ARTK and SRTK methods showed consistent results, indicating that the observed displacements were solely due to the monitoring station's movement and not a reference station shift. This provided confidence in the data's integrity during a critical event.
  • Reliable Displacement Determination: The proposed strategy consistently provided reliable displacement determinations, crucial for applications like landslide early warning systems.

Limitations and Future Work

  • ARTK Accuracy: The paper acknowledges that the positioning accuracy of the ARTK model is generally lower than that of the SRTK model. This necessitates the use of a threshold (3x RMSE) to reliably detect shifts.
  • Short Time Window: The method is optimized for short time periods (e.g., asynchronous step size not exceeding 1.5 minutes) to mitigate the accumulation of asynchronous errors. This limits its direct applicability to very long-term, low-rate monitoring without further adaptation.
  • Improving ARTK Accuracy: The authors suggest that improving the positioning accuracy of the ARTK technique, possibly by increasing the sampling rate of clock differences, would enhance the timeliness and effectiveness of reference station shift detection.
  • Further Application: The paper highlights the potential for this technique to be widely promoted and applied in geological disaster monitoring.

Summary updated

BibTeX
@misc{du2020asynchronous,
  author = {Yuan Du and Guanwen Huang and Qin Zhang and Yang Gao and Yuting Gao},
  title = {{Asynchronous RTK Method for Detecting the Stability of the Reference Station in GNSS Deformation Monitoring}},
  year = {2020},
  doi = {10.3390/s20051320},
}

Loading related content...

Figure