TL;DR
This paper introduces a novel method for Triple-Carrier Ambiguity Resolution (TCAR) that significantly improves reliability by simultaneously addressing ionospheric bias and observation noise. By cleverly exploiting the relationships between different ambiguity resolution steps, the proposed technique achieves more robust and accurate positioning, especially over longer baselines.
What Problem Does This Paper Address?
High-precision relative positioning using Global Navigation Satellite Systems (GNSS) relies heavily on accurately resolving integer ambiguities in carrier phase measurements. Triple-frequency GNSS offers the potential for faster and more reliable ambiguity resolution, but its effectiveness is hampered by two main factors: ionospheric bias and combined observation noise. Ionospheric bias, caused by the Earth's ionosphere bending GNSS signals, becomes more significant with longer baselines. Combined observation noise, inherent in the measurements, can be amplified by certain signal processing techniques. Traditional methods struggle to effectively mitigate both issues simultaneously, often leading to reduced reliability, particularly in challenging scenarios like long baselines or high ionospheric activity. This paper aims to overcome this limitation by developing a TCAR method that tackles both problems concurrently.
Methods
The core of the proposed method, termed Ionosphere-Free and Variance-Restricted (IFVR) TCAR, lies in its multi-step approach to ambiguity resolution: Extra-Wide-Lane (EWL), Wide-Lane (WL), and Narrow-Lane (NL).
- Exploiting Ambiguity Links: The key innovation is the utilization of "ambiguity links" – mathematical relationships between the ambiguities resolved at each step. This allows information from a successfully resolved step (like EWL) to inform and constrain the subsequent steps (WL and NL).
- Ionosphere-Free Combinations: For each step, the method constructs "ionosphere-free" combinations of the triple-frequency signals. This effectively eliminates the influence of ionospheric bias on the ambiguity resolution process.
- Variance Restriction: Crucially, the IFVR method doesn't just eliminate ionospheric bias; it also actively restricts the amplified observation noise that often accompanies ionosphere-free combinations. This is achieved by carefully designing the combinations to keep the noise-to-wavelength ratio at a sufficiently low level.
- Multi-Step Refinement: The process involves creating multiple independent ionosphere-free combinations at the WL and NL stages. These combinations are then used together in a least-squares estimation to further enhance the robustness and reliability of the ambiguity resolution.
Datasets
The proposed IFVR method was tested using actual triple-frequency BeiDou Navigation Satellite System (BDS) data. The experiments utilized data from four different baselines, ranging from a relatively short 7.7 km to a longer 68.8 km. These datasets were collected over 24-hour periods with a 30-second sampling interval.
Key Insights and Ideas
- Synergy of Ambiguity Links and Noise Restriction: The paper demonstrates that the true power of the IFVR method comes from the synergistic effect of using ambiguity links to constrain subsequent steps while simultaneously restricting the amplified noise. This dual approach is what allows for reliable ambiguity resolution even in the presence of significant ionospheric disturbances.
- Mitigating the Ionosphere-Free Contradiction: The long-standing challenge in TCAR has been the trade-off between eliminating ionospheric bias (which amplifies noise) and restricting noise (which can leave residual ionospheric effects). The IFVR method effectively breaks this trade-off by carefully designing combinations that achieve both goals.
- Robustness to Ionospheric Variability: The experimental results show that the IFVR method performs well even when the double-differenced ionospheric delay is substantial (up to 72.4 cm for the 68.8 km baseline). This highlights its resilience to challenging ionospheric conditions.
- Importance of NL Step: The paper emphasizes that the NL ambiguity resolution step is particularly sensitive to ionospheric bias and noise. The IFVR method's success in this step is crucial for overall positioning accuracy.
Key Results
- The proposed IFVR method achieved high reliability in ambiguity resolution across all three steps (EWL, WL, NL) for all tested baselines.
- The standard deviation (STD) of residuals for the combined observations at each step was successfully restricted to within 0.15 cycles, even for the longest baseline (68.8 km).
- When compared to state-of-the-art methods like GBIR (Geometry-Based and Ionosphere-Reduced) and GFIF (Geometry-Free and Ionosphere-Free), the IFVR method showed superior performance, particularly at longer baselines. For the 68.8 km baseline, the empirical fixed rate for the NL step improved significantly (e.g., from 22.5% for GBIR to 92.1% for IFVR).
- Positioning accuracy was also improved, with the IFVR method demonstrating more consistent and reliable results, especially in the ambiguity-fixed mode.
Limitations and Future Work
- Computational Complexity: While not explicitly detailed as a limitation, the multi-step nature and multiple combinations at WL and NL stages might imply increased computational load compared to simpler methods.
- Future Work: The authors suggest extending the application of the IFVR method to even longer baselines (100 km or 1000 km) for real-time kinematic (RTK) positioning. They also plan to further test its performance with other GNSS constellations like GPS and Galileo.
Summary updated