Knowing where you are with GNSS

  • November 5, 2020
  • Steve Rogerson

IMC executive editor Steve Rogerson talks with Bernd Heidtmann, product manager at U-Blox, about using GNSS for wearables and asset trackers.

Until recently, GNSS was often not considered for a number of IoT applications due to its high power consumption. Small edge devices such as asset trackers and fitness wearables have enough problems stretching battery life without having to get their position from satellites.

But this is now changing, as the ability to embed GNSS tracking is now coming in at much lower power, hence its integration into smart watches and various industrial applications. One company pushing the boundaries with this is Swiss firm U-Blox, and I caught up with Bernd Heidtmann, the company’s product manager, to talk about its latest innovation – the M10, a receiver platform that can track four GNSS constellations at once.

This was designed fully in-house for very low-power positioning applications such as sports watches and asset and livestock trackers.

“People prefer the precision of GNSS over cellular localisation services,” said Heidtmann. “More and more applications are using it, but one of the key problems is power consumption. By reducing power consumption, this opens up the market for GNSS. Container tracking is one of the growth applications.”

The M10 consumes 12mW in continuous tracking mode, which is five times lower than the firm’s previous M8 technology. It also gets its first position much quicker when it is turned on. And it is has a special Super-E mode that can reduce power consumption even more, but we will come back to that later.

“With the M8, the power consumption was far higher and we did see limitations there,” said Heidtmann. “The M8 came out in 2015 and at the time it was up to date. It was good. But the M10 lets us resolve some issues such as working with a small antenna in a smart watch, for example. If the battery can’t last a couple of months on continuous tracking, you wouldn’t design GNSS in.”

It is the device’s high RF sensitivity that lets it work well with small antennas, making it suitable for compact product designs. In sport watches, for instance, highly dynamic positioning accuracy is guaranteed during a run in cities, woods or under an open sky, while preserving battery life.

It is also just over a third smaller than the M8 with the chip size being just 4 by 4mm in a QFN package.

So what about Super-E. Well, this did exist with the M8, and it could bring its continuous tracking mode down from over 60mW to less than 10mW. But the M10 is starting at 12mW and so Super-E brings it down to a quarter of that. But there are drawbacks in that the update rate is far lower so the CPU can sleep between updates.

Because it can track up to four GNSS constellations at once, it can deliver positioning data even in difficult environments such as deep urban canyons. The receiver’s Super-S technology helps distinguish positioning signals from background noise to capture data even when satellite signals are weak.

In a test in Perth in Australia, U-Blox pitted the M8 and M10 against each other and found that in easier environments with no tall buildings, both had very similar accuracy, but when the device was taken to a more built-up area then the M10 performed much better.

Knowing the error is also important so designers of end products can allow for that. For example, in an autonomous lawn mower, the designer can allow an appropriate margin to make sure it stays on the grass and doesn’t start cutting down the flowers. Similarly, in a drone, knowing the error means it can be kept safely out of no-fly zones.

“It is all about knowing the error,” said Heidtmann. “We have put functionality in the receiver so we do know the error. We can see the maximum error with a confidence level of 95%. We also have weak signal compensation technology that can improve position and speed accuracy by more than 25%.”

Another problem with position technology can be security. Jamming and spoofing are becoming big threats. There are two types of jamming. The first is deliberate where someone will try to jam the signal using really high power, and Heidtmann admitted there was not a lot they could do about that. The other is where there could be interference from, say, a cellular transmitter.

The M10 uses jamming and spoofing detection technology and reports this to the host. There are also plans to be ready for the Galileo OS-NMA service that can validate whether a signal is good or bad. This has not yet been enabled in the M10 as it will need testing once the service is switched on, probably next year some time.

Engineering samples of the M10 are available this month and initial production is planned for the third quarter of next year.

“There could be some interesting use cases,” said Heidtmann. “We expect there will be more applications using this.”