English
Should manufacturers use live satellite signals to test location-aware devices?
One challenge for manufacturers of location-aware devices is to ensure that every single unit leaving the factory will perform exactly as it should.
A malfunctioning product will be found out as soon as it gives users an inaccurate position or delivers navigation instructions that are patently wrong. And thanks to the power of social media, news of a poorly-functioning product can go viral, causing reputational and commercial damage.
Functional testing is vital
The only way to ensure a product is fit for purpose is by conducting functional tests on the internal circuitry of every unit coming off the production line–including the in-built GPS/GNSS receiver.
The challenge is how best to approach functional GPS/GNSS testing so that it can be done swiftly, accurately and in concert with the many other tests performed on the product.
The temptation to use live signals
Manufacturers may be tempted to take what looks like the simplest and most cost-effective route: using the available satellite signals in the sky above the factory. This type of “live sky” testing generally entails capturing the overhead signals via an antenna on the roof and rebroadcasting them on the factory floor, where they can be applied to the device under test via coaxial cable or Bluetooth.
Risks and limitations of live sky testing
However, this approach has many issues and limitations that seriously compromise the accuracy, efficiency and rigour of the functional test, including:
Local Interference:
There will inevitably be a great number of different RF signals being reflected around the test area at the same time as the rebroadcast GNSS signal, which may interfere with any tests being conducted.
Repeatability:
Live sky testing is not a repeatable test method because satellite constellations constantly change in position and signal strength, as does the presence of interference. So conditions that cause an error during one test run cannot be replicated during subsequent test runs.
Control:
If an error is identified, with live sky testing there is no way to replicate or manipulate the conditions that caused the error in order to pinpoint its source. A test engineer can’t remove a condition from the environment to assess whether it was that condition that caused the error.
Scalability:
Live sky testing lacks scalability in three respects. It doesn’t adequately test products for global markets; it can’t test for signals and satellites that aren’t yet operational, and it isn’t suitable for a test regime applied across multiple development sites.
In short, the live sky in one location can never reflect the diversity of conditions in which a GPS system or application may have to operate. And conducting live sky tests at multiple locations can see testing costs quickly spiral out of control.
Why simulation is a better solution for GPS/GNSS production testing
The limitations outlined above are the main reasons why most manufacturers of GPS and GNSS-enabled products prefer to use radio-frequency (RF) simulators, rather than live sky signals, for functional testing.
With a simulator, the test engineer can be sure of applying exactly the same conditions and test parameters to every device coming off the production line. And as well as ensuring repeatability and control, a simulator can also be used to generate signals from constellations that are not available in the sky above the factory–either because they serve a different part of the world, or because they are not yet fully operational.
A simulator can also integrate readily with the other automated test equipment (ATE) in a standard 19” rack, increasing efficiency and throughput.
And now there’s a new generation of RF simulators designed specifically for rapid, accurate functional testing in high-volume production environments, there’s even less reason for manufacturers to run the risk of bringing malfunctioning products to market because of sub-standard testing.
