GPS Getting 1m Accuracy – Forget GLONASS, What about Galileo and BeiDou / BDS?

GPS-Track-Polar-M400Currently, many sports devices use GPS for positioning. There is a constellation of up to 31/32 GPS satellites at any one time doing their thing. Not all are operational at any one time.

GPS is the name specifically for the American positioning system.

It’s fairly hard to get GPS accuracy below 5m, indeed the precision of GPS is 5m, and hence working out how fast you are going and exactly where you are is difficult. But, add in a runner’s footpod or a cyclist’s wheel-based speed sensor and you can get your speed/pace more accurately defined.

I’d argue that, with a decent hardware implementation, the GPS method of working out speed/pace has been good enough for most people, most of the time.

However, for navigational purposes, some athletes REALLY need to know CONFIDENTLY and EXACTLY where they are.

Imagine a tiny little antenna bouncing about as you run. All the swinging and bouncing when you run is HARD to keep track of. Even harder when you hide the antenna by earing the watch the ‘wrong way around’. It’s easier when cycling to keep track, but hard when running.

Further, imagine a shoddy little GPS sports watch where all the bits have been thrown together with no electromagnetic shielding and with components touching each other.

Electrical Interference = Bad. ‘Design & Tuning’ = Good.

If you further imagine the 31/32 GPS satellites going around the Earth, then clearly your watch/device can’t see them all as many will always be over the horizon. If a building temporarily gets in the way it can see less of the satellites and trees & other obstructions can cause problems too. So maybe in a real-world scenario, you could be reduced to line of sight access to only 4 or 5 satellites. Potential accuracy will most likely suffer.

BUT. The Russians also have a satellite navigational system called GLONASS. Essentially it’s the same kind of thing as GPS but with 24/7 satellites and a level of precision/accuracy between 4.5m and 7.5m

So, simplistically, there are now more satellites to have a better stab at working out where you are.

OK you already knew that, I know.

I don’t think many of us have seen an increase in the precision of GPS+GLONASS. ie it does not appear to be telling us where we are to within an accuracy of, say, 4m rather than 5m. What it might have done is increase the chance that a location could still be established when line of sight to a GPS satellite is lost. So a forest runner may notice a difference more than a road cyclist.

The Europeans belatedly want to get in on the act too, possibly because the Americans and Russians can simply turn off their systems if they wanted to (and yes I know the Americans have said they won’t).

So we have the GALILEO ‘constellation’ of satellites. Most of them should be in orbit by the end of 2017 and operational by 2020. They will comprise 24 operational satellites (out of 30) but critically they are touted to give a PUBLIC level of precision to 1m and the possibility of encrypted accuracy down to 1cm (centimetre!).

Of course, the Chinese, being a global power, have to have one as well. That’s the snappily named BeiDou system (BDS). Again we are looking at 2020 for a global system from them and their system comprises 21 satellites (out of 35) but their public accuracy is planned to be 5m although encrypted accuracy will be 10cm.

India is also working on their version called IRNSS but that looks like being a low-altitude regional system rather than global, so that’s little use to us.

What seems to have been happening over the last few years is that the manufacturers are taking hardware shortcuts and trying to compensate with firmware fixes to ‘average’ or ‘predict’ positions. Possibly that is at least part of the explanation why you might think that the new tech seems to be performing worse than the old tech you used to have. (Or you’re getting older and your memory is fading like me!)

Delivering 1m Precision

However GALILEO, in itself, won’t give all the precision gains many would like. A high-quality chip AND electronic design are also needed.

One of the high quality (expensive) chips to watch out for is the XTAL chip.


The constellations have different altitudes and different frequencies of signal, presumably different encryptions as well. Nevertheless, chips already exist that can read all the constellations.

No doubt more battery power will be needed to process 3/4 lots of signals…

Delivering Altitude Precision

So, come 2021 will your Fenix 6 have a quad-constellation receiver (GPS, GLONASS, GALILEO, BDS) and super accuracy to the nearest metre? Maybe! Maybe even more precise if the design is good enough.

Even improving current precision levels by a couple of metres opens up the prospect of notably better GPS-estimated altitude as well as location. The distance from you to the various satellites can also be used to determine altitude – this is already used on many GPS sports devices.

Other sports devices use a barometer to measure pressure changes to estimate altitude/ascent/descent changes.

Yet other devices, TomTom and Suunto for example, combine or FUSE both methods.

These methods are all going to have the potential to get more accurate.

My personal belief is that, just like we currently get a-GPS (assisted GPS), quick satellite fixes by predicting satellite locations then so shall we also get precise altitude figures pre-loaded for known GPS locations. For example, your Suunto might know your home/POI GPS location and ALWAYS will AUTOMATICALLY recalibrate based on that known and fixed altitude at the POI. You could also see how downloading a course/route to your sports watch could also quite easily download various altitudes along the route. [Here’s where you tell me all the wonderful SPORTS devices that can already do that 🙂 ]. You could also see that, as well as storing routes and maps, the space cost of an altitude database on your watch will also add a space and processing cost.

So, the technology is probably ‘there or thereabouts’ right now. The limiter will be the sports watch manufacturers who might determine that not enough of us will be willing to pay for the increased costs of delivering increased accuracy.


Reader-Powered Content

This content is not sponsored. It’s mostly me behind the labour of love which is this site and I appreciate everyone who follows, subscribes or Buys Me A Coffee ❤️ Alternatively please buy the reviewed product from my partners. Thank you! FTC: Affiliate Disclosure: Links pay commission. As an Amazon Associate, I earn from qualifying purchases.

10 thoughts on “GPS Getting 1m Accuracy – Forget GLONASS, What about Galileo and BeiDou / BDS?

  1. It’s not just the satellites but the device design can also be improved. The Garmin Vivoactive HR is more accurate than the Garmin Fenix 3; supposedly because Fenix 3 has metal and sapphire blocking GPS signals. The Ambit3 has a bigger antenna and reception area.

    I also notice that Garmins do not improve their positioning over time. If you do not stay still for ~30sec and do not get GPS lock before you start the activity, there will be false recordings. It would be possible though that the watch determines a better location during a long activity even if you did not have prior GPS lock

  2. In theory it’s all super-cool. In practice everybody turns off GLONASS because once it’s on, positioning accuracy is decreased. At the same time, most people don’t seem to really care much about accuracy (if they did, they wouldn’t have bought latest Garmin’s flawed watches until GPS issues were corrected). So, as of today, it seems just a marketing gimmick.

    1. i’m probably erring towards your opinion on this one Greg. There really ARE some people who want good accuracy tho. I think where it will help will be to provide consistency of a level of accuracy in more demanding conditions as opposed to getting accurate to 50cm.

      1. Still marketing though. If we are not talking about a bad GPS track then what use is accuracy of more than 1m when running/cycling? To track what? To improve to what? GPS is by default bad! Imagine a dog on a rope and you running with it. That is GPS. The worse the accuracy the longer the rope. Now imagine that his dog just bounces around it’s lease while you run. This is how it’s tracked, and how it will be till 2020.

        If you are not marking down hiking paths that are dangerous or a mistake can lead to danger, having <1m accuracy in a run in NYC, well in any point of view it's just a hobby that none except you will see.

        None will go to your Garmin/Strava/Suunto track and say: Oh look the guy passed nearby the garbage bin. None cares. No real athlete has any of these concerns.

        My .50c from all the discussions when a new GPS watch comes out.
        If it's not really bad GPS track and if it shows the running/cycling etc speed ok then you are good to go and train.

      2. yes I think gps is ok on the whole for me. my only real concern is ‘correct’ instant pace. but the counter to that would be that a good athlete would KNOW how fast s/he is running. some people train by distance and want to know the accurate distance, some train by calorie which can also be closely linked to distance.

  3. Don’t expect too much from Galileo, the “increased accuracy” is a side effect of the improved accuracy clock being installed. The improved accuracy clock is partly advances in technology but mostly because GPS is now a part of national infrastructure and is therefore worth investment under both military and civilian budgets. This has literally nothing to do with knowing where you or anything else is and everything to do with financial markets. As a result, the people putting it in really don’t care if you get better accuracy other than because that might make the cost easier to bear for the nations involved.
    The military don’t need or particularly want satelite navigation. It’s probably handy in the trucks, but missiles and other important stuff use celestial navigation, dead reckoning and other clever techniques to navigate because GPS is so insecure and open to attack. It can be blocked with £20 of hardware from Maplin quite easily, and the military regularly do this within the UK – you can find out where by reading NTMs for shipping.
    The real use, and the thing we’re trying to protect is the economy. It’s pretty trivial to not only disable GPS but override it with a fake signal. A hack was once described where the time would be gradually changed to allow a fraction of a second in which to make early stock decisions. Using this, the economy could be made to collapse, or someone might choose to get rich. There is such reliance on GPS that it was unthinkable to have another nation in charge of it, so “we” funded our own. The UK might now need to fund our own Brexit satelites! The new system I believe includes anti-tamper solutions so you’d at least know whether the signal was from fred in his van or from the satelite.
    Radio 4 was due for switch off on certain frequencies and has been kept alive for time keeping for this same reason (3 vote timekeeping with GPS, radio and local timesource such as radium clock) as well as because it’s used in the last resort decisions on Trident to see if the UK is still functioning.

    1. I am actually here with a Garmin Fenix 5s Plus and can confirm I have much better accuracy through Galileo that I had with my Forerunner 935 and Fenix 5 line. Even in cities I see dramatic improvements in accuracy

  4. Polar promised a Vantage update by october 2019 for Galileo/QZSS/BeiDou satellite system support which would improve GPS functionality in specific areas. I noticed that the update page has dropped the BeiDou part (only Galileo/QZSS satellite system support). I´m based in China, so I was looking forward to the BeiDou support (maybe the Japanese QZSS might help a bit). Any reason why BeiDou was dropped out?

Comments are closed.