Reviews are noting discrepancies between GPS readings from wearable devices and smartphones. We look at how wearables manufacturers can improve position accuracy.
Location tracking with GPS is becoming a staple feature in many consumer devices, from watches and shoes to high-end fitness trackers.
But GPS positioning on its own is often not accurate enough to provide the location and navigation accuracy consumers have come to expect. With more location-aware products coming on to the market, some consumer product reviews have started to highlight inconsistencies between (for example) the readings provided by a fitness tracker and a smartphone over the same course.
The smartphone advantage
So if both devices are using GPS to obtain a position, how do such discrepancies occur? The answer is that smartphones often have a few extras that the average wearable device does not. As multi-purpose devices, phones include a variety of additional sensors and data sources that can better inform the device of its location and offer more stable readings than GPS alone.
The importance of time-to-first-fix
The time it takes to initially “lock-on” to a GPS reading is often the first casualty of using GPS on its own. The ability to quickly get a GPS reading is called Time to First Fix, and it’s an essential part of any wearable device—especially one designed to keep up with the pace of its user’s lifestyle.
The modern smartphone can achieve a rapid Time to First Fix. By using mobile data and information from the cell radio, a smartphone can roughly find its location even before its user turn the GPS on. So when the device actually starts trying to find GPS satellites to connect to, it already knows roughly where to look.
Wearable devices tend not to include cellular functionality, but that doesn’t mean they can’t cache satellite position data and restore it at a later point, allowing for warm-start or hot-start time to first fix. Achieving this in a device depends on a variety of factors, from the system software, to how the GPS chipset is integrated.
Aside from time to first fix, the average smartphone also has an advantage when it comes to measuring distances. Accelerometers, gyroscopes, and other inertial sensors allow smartphones to roughly calculate distance travelled even if GPS signals are disrupted or lost.
Whether that disruption is caused by multipath signal refractions from tall buildings, covered areas or atmospheric interference, only a wearable with a comprehensive set of inertial sensors can compete with the modern smartphone’s ability to compensate for the loss of signal.
How can you ensure GPS tracking inaccuracy in your wearable device?
For obvious reasons, location tracking accuracy is something that users of fitness trackers and running watches take very seriously. But for manufacturers, it can be a tricky thing to ensure—especially if your device is using GPS as its only source of location data.
Many factors can affect the accuracy of GPS or other Global Navigation Satellite System (GNSS) readings. For device manufacturers, there are four main areas to be aware of when integrating a GNSS chipset:
Interference – Electromagnetic noise from other RF elements within the device can interfere with GNSS signals
Antenna placement – The size and location of the GNSS antenna can have a major impact on its ability to acquire and maintain a lock on satellite signals
GNSS shielding – The type of materials used in the device case and GNSS shield can significantly affect signal reception
The software layer – Different software algorithms can interpret GNSS signal data in vastly different ways
How can I overcome these challenges?
For manufacturers developing new consumer devices, there are unfortunately no hard and fast rules to overcoming these four challenges. They can even be compounded by design constraints such as small form factors and sleek outlines that can prevent optimal antenna placement.
Software can also be just as important as hardware. Considering GNSS chips can be power-intensive, the software responsible for putting GNSS functions to sleep (to help save the battery during inactivity) can have an impact on time to first fix if the software isn’t properly calibrated.
For general GNSS functionality, the only way to ensure accuracy is by rigorously testing prototype designs with different chipsets, antenna positions, shielding techniques, software algorithms and interference levels. Only by thoroughly testing multiple permutations can you identify the chipset and design that deliver the most accurate positioning, navigation and timing capabilities.
Live satellite signals are inadequate for testing purposes
But testing every permutation with “live” satellite signals can be time-consuming, resource-intensive and – since the satellites in the sky are constantly changing position, meaning test conditions are never the same – unreliable.
In some cases – for example, if you wanted to test how well your device works with signals from the not-yet-fully-operational Chinese satellite system BeiDou – testing with live signals is actually impossible.
A far more effective approach is to combine lab-based simulation with a Record & Playback solution. With a GNSS signal simulator, you can test your device in any combination of conditions, and repeat the same test conditions over and over again to assess the impact of your modifications.
Then, when you are confident your device works well in simulated test conditions, you can significantly speed up and lower the cost of field testing by recording the RF signal environment at a test site, and playing it back to the device in the lab.
Find out more
Spirent can help you find the right combination of simulation and Record and Playback to ensure your commercial device delivers the right levels of GNSS accuracy and reliability for your customers.
For more information on how to use GNSS Simulation and Record & Playback solutions to test your devices’ positioning and navigation features, read our eBook: Four Considerations to Prevent Damaging Your Wearable Brand, or contact us.