By Shahabuddin Amerudin
Authors: Joshua Critchley-Marrows, Marco Fortunato and William Roberts
Publication Date: April 6, 2020
The article discusses a new methodology aimed at achieving sub-meter Global Navigation Satellite System (GNSS) accuracies in consumer devices such as smartphones. The goal is to enable applications like augmented reality and visually impaired navigation to function with higher precision. The article explores the challenges of achieving high accuracy in mass-market GNSS receivers, especially in the context of real-time positioning. It also introduces an alternative approach to Precise Point Positioning (PPP) to improve accuracy in challenging receiver environments.
Key Points:
- Challenges in Mass-Market GNSS Receivers: Mass-market GNSS receivers in consumer devices have limitations such as limited observation data from satellite constellations and restricted access to multi-frequency measurements. Additionally, the design of smartphones, with shared antennas and other communication hardware, can introduce noise and interference, affecting positioning accuracy. As a result, the typical accuracy achievable in ideal conditions is a few meters.
- Sub-Meter Accuracy Goals: The article sets a benchmark of achieving 50 cm accuracy for smartphone positioning due to the increasing demand for location-based apps requiring higher precision.
- The FLAMINGO Initiative: The methodology presented in the article is developed as part of the FLAMINGO initiative, which enables real-time PPP and Real-Time Kinematic (RTK) GNSS positioning on smartphones. The FLAMINGO service has achieved sub-meter positioning accuracies in real-time, but it relies on base station infrastructure within approximately 30 km of the user.
- Methodology Overview: The proposed methodology involves modifications to traditional GNSS processing. It includes a preprocessing stage to remove poor GNSS observables, a combination of single-frequency PPP with dual-frequency ionospheric-free combination, and signal-to-noise ratio and elevation-based noise weighting.
- Poor-Measurement Rejection: The preprocessing stage filters out poor GNSS observables using three detection strategies: Code-Minus-Carrier, Phase Range Rate, and Pseudorange Rate. These strategies identify and remove observations with errors caused by factors like multipath or cycle slips.
- SFDF Combination: To compensate for the mix of single-frequency and dual-frequency observations from smartphones or mass-market receivers, the methodology proposes the use of SFDF (Single Frequency, Dual Frequency) combination. This combination reduces error terms and improves vertical accuracy.
- Model Weighting: The methodology uses both elevation and signal-to-noise ratio (SNR) as weights in the GNSS processing algorithm. The combination of these weights enhances the precision of the positioning solution.
- Implementation and Test Results: The methodology is tested on real-time receiver trials, and the results are compared to an idealized post-processing scenario. The real-time solutions show less accuracy compared to the post-processed ideal case due to the challenges and limitations of smartphone environments. The SFDF model and SNR weighting show improvements in vertical accuracy.
Conclusion
The article presents a novel methodology to achieve sub-meter GNSS accuracies in consumer devices like smartphones. While the ideal case demonstrates high accuracy in post-processing, real-time implementation faces challenges due to the complex and noisy environment of smartphones. Nevertheless, the methodology shows promise for improving vertical accuracy and provides valuable insights for future GNSS research and development in consumer applications. Achieving sub-meter accurate PPP in real-time remains a goal with significant potential benefits.
Suggestion for Citation: Amerudin, S. (2023). A Review: Accuracy for the Masses: Real-Time Sub-Meter in a Consumer Receiver? [Online] Available at: https://people.utm.my/shahabuddin/?p=6586 (Accessed: 31 July 2023).