Millions of Android smartphones were quietly involved in one of the largest crowdsourced navigation projects ever
- Phone-based maps globally outperform Klobuchar’s traditional ionospheric model
- Underserved regions such as Africa benefit from improved ionosphere data
- Geomagnetic storms are monitored with high precision using smartphone networks
Mapping the ionosphere is critical for improving the accuracy of navigation systems, but existing methods face significant limitations. Although ground-based GNSS stations provide detailed maps of ionospheric total electron content (TEC), their coverage is uneven, leaving large gaps in areas of low coverage.
Now, researchers from Google Research, Mountain View, CA, USA, have demonstrated an innovative solution using millions of Android smartphones as a distributed network of sensors in a study published in Nature.
Despite being less accurate than traditional instruments, these devices effectively double measurement coverage, provide accurate ionosphere data and address long-standing infrastructure gaps.
The influence of the ionosphere on navigation
The ionosphere, a layer of ionized plasma extending from 50 to 1,500 km above Earth, significantly affects the Global Navigation Satellite System (GNSS) signals by introducing positioning errors. Traditional ground-based GNSS stations, while accurate, suffer from limited spatial coverage, leaving underserved regions vulnerable to inaccuracies.
The Google Research study takes advantage of the billions of smartphones equipped with dual-frequency GNSS receivers to address coverage gaps. Unlike conventional GNSS stations, smartphones are mobile, widely distributed and capable of capturing large amounts of data.
By merging and averaging measurements from millions of devices, researchers achieved accuracy comparable to that of dedicated measuring stations, and were even able to resolve phenomena such as solar storms and plasma density structures.
Android’s GNSS API facilitated the collection of satellite signal data, such as travel times and frequencies, which were then used to estimate ionospheric total electron content (TEC).
The study found that while individual phone measurements are noisier than those from monitoring stations, their collective data provided robust and reliable results.
Comparisons showed that the phone-based TEC model outperformed existing methods such as the Klobuchar model, which is widely used in mobile phones.
Using data from Android phones, the researchers doubled the measurement coverage of the ionosphere compared to traditional methods, and were able to map plasma bubbles over India and South America and the storm-enhanced density over North America during a geomagnetic storm in May 2024.
They also observed mid-latitude troughs over Europe and equatorial anomalies – a phenomenon previously inaccessible due to the sparse coverage of stations.
In particular, regions such as India, South America and Africa, which are often not served by traditional monitoring networks, have benefited significantly from this approach, which provided real-time, high-resolution TEC maps.