Navigating Ephemeris Data in a Cloud-Based System: Strategies for Robust GNSS Measurements

In the realm of Global Navigation Satellite Systems (GNSS), the accuracy of our positional data relies significantly on one critical component - ephemeris data. This complex set of variables acts as the bedrock for precise location determination. However, navigating the intricacies of ephemeris data acquisition and management, particularly within a distributed system, presents a unique set of challenges. Let's delve deeper into these challenges and explore comprehensive solutions to ensure robust GNSS operations.

^‍What Exactly Is Ephemeris Data?

GNSS positioning fundamentally operates on the principle of trilateration, which involves calculating a location based on distances to multiple known points - satellites, in our case. For this calculation to be accurate, we must have precise knowledge of where each satellite is in space. This is where ephemeris data comes into play. It essentially provides a detailed orbital map for each satellite, describing its trajectory and velocity with great precision for a given period of time. Moreover, it includes accurate timing information from the satellite's onboard atomic clock for the same period, another critical component for position determination.

Each satellite broadcasts its own specific ephemeris data set, and this information is not static. The updates are frequent, occurring every half hour, hour, or two hours, depending on the specific GNSS constellation.

‍

Acquiring and Managing Ephemeris Data, A Multifaceted Challenge

While ideally, every GNSS receiver would serve as a flawlessly reliable source of ephemeris data, the reality is far more complex. The process of acquiring, interpreting, and managing this data is fraught with potential pitfalls, for example these key challenges:

Data Handling Errors: Mistakes can occur during the initial decoding of the data stream. For instance, a receiver might misinterpret a specific value within a data field, leading to inaccuracies in orbital calculations. This happens frequently due to the weak error detection and correction schemes of most available data streams. Another potential issue is the mixing of data fields. Values intended for one field might inadvertently be assigned to another, or data from one satellite might be erroneously associated with a different one.
‍
Constellation-Specific Issues: Each GNSS constellation (GPS, Galileo, GLONASS, Beidou, etc.) has its own unique characteristics and known quirks. An effective ephemeris management system must be capable of accommodating these variations and addressing potential errors that are specific to each constellation, or better yet, bypassing them altogether when possible.
‍
Security Concerns: In some applications, the integrity and authenticity of ephemeris data are of paramount importance. Measures must be in place to prevent data tampering or spoofing.
‍
Ephemeris and Satellite Health: Satellites can occasionally experience malfunctions. When this happens, the satellite is usually taken down for maintenance; however, the satellite may still transmit "junk" ephemeris data which is marked as ‘unhealthy’. While this may not be a problem in and of itself, the ‘unhealthy’ flag itself, may at times be erroneous, either marking a source as invalid when it is or vice-versa. Treating invalid data as valid can lead to significant errors.
‍
Latency and Synchronization: In a decentralized system, data might arrive at different times from various sources. This may cause out of date data to arrive after its validity period, causing errors. Ensuring proper synchronization and handling potential latency issues is crucial to maintaining data integrity.

The above example of a single erroneous parameter ( √a ) in ephemeris data received from GPS satellite 23, can cause miscalculation of satellite position and hence the miscalculation of the rover position. Even in a seemingly small error such as this, the impact on the final position can, and usually will, be significant.

These challenges are impossible to solve when dealing with a single source of ephemeris, such as a single receiver. This is true for standard navigation (the standalone receiver scenario), and for augmentation services (such as DGPS and RTK), that still rely on the receiver’s own ephemeris data. A decentralized system containing many sources, such as Tupaia’s cloud service, can use its many sources as a potential solution. This, however, is not a trivial solution, as the large number of sources may compound the issues significantly when conflicts and inconsistencies arise between the sources. It is therefore imperative to establish robust strategies for resolving discrepancies and ensuring the accuracy and reliability of the selected ephemeris data.

‍

Exploring Potential Solutions: A Range of Strategies

To effectively address these challenges, we must consider a range of potential solutions.

‍

The "First Come, First Served" Approach

One of the simplest methods for a decentralized system is to accept and utilize the first available ephemeris data set and stick with it until a newer version (with a more recent timestamp) arrives. This approach is straightforward to implement, but it carries significant risk. What if that initial ephemeris data is flawed or inaccurate? If that is the case, our system would essentially be "locked in" to using that faulty data until the next update cycle. This could lead to prolonged periods of inaccurate orbital calculations and, consequently, degraded positioning performance, as well as missing detection of cases where health changes suddenly for any given satellite.

‍

The "Latest is Best" Approach

Another seemingly logical strategy is to always use the most recent ephemeris data received. While this might seem like a good way to avoid persisting with outdated or erroneous data, it has its own set of potential problems. While no flawed data will be used for long, which is good, there may be cases where a perfectly valid dataset might be replaced by incorrect data simply because it arrived at a later time. This could create potentially extended periods where the orbital data being used is in error after being perfectly viable, which will have an effect on the positioning solutions. This method sacrifices stability for recency and may lead to frequent, albeit short, errors.

‍

The "Weighted by Source Reliability" Approach

This method takes the data from all available sources but gives them a weighting based on how reliable each source has proven to be in the past. This can be a static weighing depending on source type, or a dynamic one that is based on the sources’ history of errors. When choosing which ephemeris to utilize, this method will lean toward those with the highest weight but still take into account recency as well. While this method does effectively eliminate sources that have systemic errors or faults, this does not solve the occurrence of sporadic errors that may rise randomly.

‍

Dynamic Consensus and Verification: A More Robust Strategy

To overcome the shortcomings of these simpler methods, Tupaia has developed a solution that is both dynamic and based on a consensus mechanism. This involves establishing a database to allow us to consider the consistency of different ephemerides from the same timestamp, which may be weighted for source reliability, and remove any outliers. The system also incorporates a mechanism to handle exceptions for critical changes that are legitimate. By seeking the largest consensus among multiple sources, we can minimize the impact of individual errors and maximize the reliability of our GNSS measurements.

Striving for Reliability in Ephemeris Data Management

Effectively managing ephemeris data within a cloud-based system is a delicate balancing act. We must be responsive to updates, ensuring that we have access to the latest and most accurate information. At the same time, we must guard against the inclusion of erroneous or unreliable data, as it affects positioning calculations significantly. By designing and implementing a sophisticated, dynamic system that prioritizes consensus and verification, Tupaia can overcome the inherent challenges and uphold the high levels of accuracy that are essential for modern GNSS applications, which current augment systems, like RTK, cannot do. While it is not a fool-proof system, it does uniquely minimize the time we spend in error to the lowest possible amount. Tupaia’s cloud service is not just a technical necessity; it is one of the very foundations upon which reliable, robust and precise location services are built.

‍

- Alon Kivilevitz, Algorithms Engineer at Tupaia.

‍

Interested in Learning More? If you’d like to learn more about our precise positioning solutions or see how we can help your business, feel free to contact us. Let’s work together to bring precision to your next project!

Book a Demo Want to see our technology in action? Contact us at info@tupaia-pos.com to book a demo and discover how we can support your specific needs.

‍