back to article Europe to upgrade its continental GPS

The European Space Agency has announced plans to upgrade the European Geostationary Navigation Overlay Service (EGNOS). EGNOS uses three geostationary satellites and 40 ground stations to verify GPS signals. Output is deemed accurate to within three metres, rather less than the 17 metres assumed for unrefined GPS. EGNOS …

  1. Anonymous Coward
    Anonymous Coward

    EGNOS on your face?

    Let's hope that after this upgrade EGNOS actually works properly. The last time I had cause to use it (admittedly some years ago but after it was supposed to be fully operational) it actually degraded the accuracy of the GPS fix rather than improved it. Mind you, that was with a u-blox GPS receiver that could do sub 2m accuracy unaided on a good day.

    And before anyone asks, no, it wasn't using L2C or anything exotic, it was just a standard consumer-grade receiver but designed and built damn well. Unlike the current generation of GPS receivers built into a phone's GSM radio module and down-specced to match the US E911 50m accuracy requirement.

    1. Anonymous Coward
      Anonymous Coward

      Re: EGNOS on your face?

      "Unlike the current generation of GPS receivers built into a phone's GSM radio module and down-specced to match the US E911 50m accuracy requirement."

      <<Citation needed>>

      1. Gene Cash Silver badge

        Re: EGNOS on your face?

        Citation: my US Nexus 6P phone where the GPS is crap and the compass is so completely useless I've had to attach a 50¢ bubble compass to my bike.

        The GPS wanders around worse than a drunk after closing time. I've sat there watching it move over a 75m radius.

  2. John Smith 19 Gold badge
    Coat

    Such a waste of a good acronym

    With just a little more effort pilots could be flying across Europe on EGNOG.

    Time to be gone.

  3. DropBear

    I'm having extreme difficulties trying to imagine a scenario in which rescue workers within 17 metres of a person in distress are having major difficulties locating it. Sure, more precision is always a good thing, I just don't see it making much difference in that case (let alone for any aircraft - I don't think any of them is expected to fly within 17m of any treetops...) and frankly I'm kinda tired that (in my experience) no matter what one switches on or off, typical consumer GPS just never seems to get more precise than "somewhere within 5-10m" in practice.

    1. Anonymous Coward
      Anonymous Coward

      Which floor of a high rise flat? Which flat on a floor of a block of flats?

      I don't suppose it's for finding people in an empty field.

      1. phuzz Silver badge

        "Which floor of a high rise flat?"

        GPS is generally about half as accurate in the vertical as it is in horizontal measurements. EGNOGS claim 1-3m horizontal and 2-4m vertical resolution. So you could probably narrow it down to within two or three floors, assuming you were still getting a signal inside the tower block.

    2. Anonymous Coward Silver badge

      If you were in distress on one side of a mountain ridge and the S&R people were diligently searching the other side for you, you wouldn't be too happy. Just 17m away, over this 50m high, sheer, brittle rockface.

    3. Jon 37

      You need that precision for doing instrument landings. You don't want to land with the middle of the plane 17m from the runway center, because you'd be off the edge of the runway (or at least dangerously close to it). 3m, plus an allowance for the rest of the autopilot systems not being perfect, is OK.

      Currently big airports have expensive instrument landing systems that use radio transmitters next to the runway that the plane can use to line up correctly, and smaller airfields just close in bad weather. EGNOS will allow pilots to do instrument landings at smaller airfields that don't have those ground systems, just using GPS co-ordinates pre-programmed into the navigation system on the plane. In the distant future it might allow some of the bigger airports to stop providing/maintaining those ground systems, too.

      Additionally, one major problem with GPS is "what if it goes wrong". Oh sure that's extremely rare, but if you're relying on GPS to separate aircraft, especially in busy parts of the sky such as near airports, then a brief error on one of the satellites could cause a position error on planes that see that satellite, which could cause planes to collide. EGNOS will detect that very rapidly, and tell the planes to ignore the dodgy satellite. People want to use GPS positioning for air traffic control in future, and EGNOS is an important part of doing that safely.

      1. SkippyBing

        'EGNOS will allow pilots to do instrument landings at smaller airfields that don't have those ground systems'

        It already does, it's not that widely spread in the UK yet but there are approved GPS approach procedures here already and other countries are using them quite liberally as once the approach has been designed and approved (~£27K in the UK) it doesn't have any ongoing maintenance costs unlike an ILS. They also have the advantage that you're not relying on line of sight to the ILS transmitter when you're designing your approach, there's at least one for helicopters in Australia that lets you down into a valley which you then follow around a corner to the landing site.

      2. John Brown (no body) Silver badge

        "People want to use GPS positioning for air traffic control in future, and EGNOS is an important part of doing that safely."

        But isn't Galileo supposed to be that much more accurate anyway? And there are a number of developments in Europe, the USA and Japan (at least) bringing in systems where aircraft will use SatNav (Galileo/GPS/GLOSNAS/etc) and allow for free skies route planing, ie the pilot/airline choose the route, not the ATC. Unless EGNOS is already planned as part of that system, then it seems likely to have a very short lifespan for significant investment.

    4. allthecoolshortnamesweretaken

      "I'm having extreme difficulties trying to imagine a scenario in which rescue workers within 17 metres of a person in distress are having major difficulties locating it."

      Happened to a mate of mine*. Fucked up big on his BMW R80. Crashed in the side of an oncoming Audi he had not seen. Curvy country road, cutting corners, way too fast... Anyway, he ripped out the Audi's rear axle and tank. The good thing about riding a boxer was that ge got to keep his leg because the cylinder took most of the brunt. The unfortunate thing about those boxers is that they have a tendency to right themselves and drive on for a little bit after you've fallen off. Which was what had happened. Emergency sevices started looking near the crashed bike; its rider lay maybe 30 m from that. Of course in a ditch under some bushes. They eventually found him when he managed to crawl a bit closer to the road.

      * I know, I know.

  4. Anonymous Coward
    Anonymous Coward

    It seems odd to me that GPS et al aren't accurate to the inch. Does anybody know why they're so inaccurate? Is it hard to keep track of the exact position of the satellite, perhaps?

    1. Peter2 Silver badge

      The handsets don't keep track of the satellites per se. The satellites transmit a timing signal, and the receiving device does some math based on the number of signals received and calculates the position of the handset relative to the position of the satellites.

      With GPS there are two sets of signals, military and civilian. The civilian sets find that the timing signals are slightly (and randomly) offset to deliberately induce a certain level of inaccuracy compared to the military signals. This is a design feature intended to make GPS fit for navigation but useless as an unauthorised guidance tool for throwing missiles around. This was originally about 50M horizontally and similar vertically, but reduced down to about 5M horizontally and 20M vertically for civilian users (inaccurate enough that if somebody built a cruise missile trying to do low level flight using it then it'd probably crash enroute)

      1970's military systems were supposedly accurate to sub meter after decrypting the military signals, I'd guess that it would be possible to do better than this with modern processing power.

      1. Rustbucket

        Selective availability for civilian use was turned off in 2000 by President Clinton, and the horizontal error went up to 100m, not 50m. The article below talks about it and other factors contributing to GPS error.

        https://en.wikipedia.org/wiki/Selective_Availability

    2. John Smith 19 Gold badge
      Boffin

      "It seems odd to me that GPS et al aren't accurate to the inch. "

      No. It's because when the DoD set up the system they thought that was good enough for civilian use. It's actually about 1 arc second around the Earths circumference at the equator. 1/60 of a Nautical Mile.

      The encrypted MilSpec version is accurate down to about 3m. Various things aid in this. For example the civilian signal had a bandwidth of about 1.024MHz, while the mil version is 10.24MHz. SOP is for the military receiver to lock onto the civilian signal first (IIRC its spread spectrum code is 1024 bits, so it repeats 1000x a sec) then recover the time of week so it can lock onto the encrypted signal.

      The civilian signal has a fixed model of atmospheric propagation but the military versions (IIRC) use both frequencies to measure it directly (now you can put a lot of transistors on a chip civilian chips can do this as well). In theory no one can break the encryption on the Mil GPS signal but people have found other ways to use to give near military accuracy, but not in real time, and certainly not in "high dynamics" IE missile applications.

      Cm accuracy is possible by using a network of fixed position transmitters that send out corrections which can be combined with the GPS data stream (50bps about 12000 bits split into 300bit pages repeating some fast changing data IE exact time of day and week and other slower changing data) to give that high resolution. It's called WAIS and commercial aircraft have it. Such receivers are much more expensive.

      1. Andrew 1234

        Re: "It seems odd to me that GPS et al aren't accurate to the inch. "

        John Smith - I think you mean WAAS (Wide Area Augmentation System). That's the US equivalent of EGNOS, just about all consumer GPS systems have it for no extra cost and on a good day it gives you 1 m accuracy. They are both a type of SBAS (Space based augmentation systems).

        A high end RTK GPS will give you 1 cm accuracy. A good RTK system is getting into the >$10,000 price range.

        1. John Brown (no body) Silver badge

          Re: "It seems odd to me that GPS et al aren't accurate to the inch. "

          "A high end RTK GPS will give you 1 cm accuracy. A good RTK system is getting into the >$10,000 price range."

          And that's little more than a rounding error on the price of a new passenger aircraft, fully fitted out.

    3. Anonymous Coward
      Anonymous Coward

      5-10m is inaccurate?

      > It seems odd to me that GPS et al aren't accurate to the inch.

      > Does anybody know why they're so inaccurate?

      GPS is picking up a signal that is about the same strength as background noise.

      It then has to measure the exact time that signal arrived, calculate how long that signal took to reach the user from the satellite and exactly where the satellite was when the signal left it. It then has to calculate the distance traveled in that time (the speed of light varies significantly through the ionosphere) to give it a sphere around the satellite that is must be on. Do this with 4 satellites and you can calculate your location.

      Every nanosecond of error in these measurements and calculations results in ~30 cm of error in the range measurements and so a comparable error in position.

      You get more than 1 inch of error simply due to the errors in the satellites on board atomic clocks, that adds about 2 meters of error. One of the improvements in Galileo over the current GPS system is to use more accurate clocks.

  5. dermotw

    "I'm having extreme difficulties trying to imagine a scenario in which rescue workers within 17 metres of a person in distress are having major difficulties locating it. Sure, more precision is always a good thing, I just don't see it making much difference in that case (let alone for any aircraft - I don't think any of them is expected to fly within 17m of any treetops...) ..."

    Errr well, how about a burning building, person on the other side of a wall? Or ditto & add smoke... High precision is wanted for aircraft as well, because then GPS can be used for navigation (proper, airplane navigation). That is one of the things Galileo is supposed to bring, it should have sub-1M accuracy, allowing you to e.g. land a plane with it.

    Phone GPS chipsets are spec'd basically for US accuracy level GPS, i.e. not very...

    1. Anonymous Coward
      Anonymous Coward

      Phone GPS chipsets are spec'd basically for US accuracy level GPS, i.e. not very.

      true enough however smartphone GPS accuracy can be improved to less then one metre using GNSS corrections. Players within the GNSS industry have for years been enhancing GPS, its now coming to smartphones. Subscription costs for accurate GPS corrections are typically too high for consumers devices like smart phones - that is all changing.

      options include sharing precise location within apps or having stand alone apps. then we have apps that combine indoor and outdoor location information- all to within a metre accuracy and all at a cost suitable for most consumers.

  6. GingerOne

    I know little to nothing about global navigation satellite systems. My phone uses GPS, GLONASS & Galileo. Are people suggesting it uses lowest common denominator (GPS) accuracy, even when receiving signals from GLONASS or Galileo?

    1. Anonymous Coward
      Anonymous Coward

      Are people suggesting it uses lowest common denominator (GPS) accuracy, even when receiving signals from GLONASS or Galileo?

      Only the paranoid or uninformed.

      GNSS (Global Navigation Satellite System, a catch all for GPS, Glonass, Galileo and BeiDou) systems in phones give the best results they can. Their performance is limited by using a crappy antenna that's ofent pointing away from the sky and placed right next to a massive electrical white noise generator called a smartphone.

      The accuracy doesn't sucks because it's been deliberately degraded, it sucks because it's a terrible GPS receiver. The phone manufacturers need to make it work well enough to meet the US E911 requirement but once they reach that point any further improvements are wasted cost. No one buys a phone because it has good GPS.

  7. Allan George Dyer
    Windows

    Reliable but not accurate

    I've found that different phone GPSs in my flat very reliably report a position about 200m away, in a public park. I've guessed this is due to [hand-waving] reflections from nearby buildings and diffraction by reinforced concrete. I wonder whether the proposed upgrade will actually find my home, or merely pinpoint a particular park bench.

    1. John Brown (no body) Silver badge

      Re: Reliable but not accurate

      Is that giving actual co-ordinates which you've than plotted on an actual map or is the nav/map app just placing you at the nearest known point to whatever level of accuracy it's managed to get? For example, my Garmin SatNav is optimised for road use and will often place me as being on a road even when I might be 30 or more metres from where it tells me I am because the side road or car park or whatever isn't on the internal map. If I switch it on while not on a road, I've seen it place me properly on the map as it gets the sat signals, then place me on the road once it realises that the calculated position isn't where it "should" be, ie the nearest road within some margin of error.

  8. anthonyhegedus Silver badge

    Which phones are currently able to pick up galileo signals? And which phones actually use them?

    1. GingerOne

      My new BQ Aquaris X Pro supposedly uses Galileo (it's a Spanish brand) not sure how to check if it actually is though.

    2. S4qFBxkFFg

      "Which phones are currently able to pick up galileo signals? And which phones actually use them?"

      http://www.usegalileo.eu/EN/inner.html#data=smartphone

      edit: to actually check it's working, try this app: https://play.google.com/store/apps/details?id=com.chartcross.gpstest&hl=en_GB (it uses different icons for the different constellations - mine has only ever picked up GPS and GLONASS, and once, an SBAS satellite - it allegedly picks up Beidou, but they probably don't get high enough in the sky here)

  9. Anonymous Coward
    Anonymous Coward

    Its all-out road pricing

    The accuracy and EU ownership is needed to support EU-wide road pricing, which is the next big greeny inspired scam to fleece the poor motorist...

  10. Kevin Johnston

    I recall

    a wonderful story about the Nav system installed on a supertanker which could combine various signals from fixed points with GPS to plot the position of the receiver to within 1metre. Seemed slight overkill to me at the time when it could travel the best part of a mile before it could stop.

    1. John Brown (no body) Silver badge

      Re: I recall

      ...but quite important when manoeuvring through channels where the depth might have an impact on the route taken, especially near harbours or during the actual docking process when you can't see the sides due it being a honking big ship (might save on pilot fees :-))

  11. Blotto Silver badge

    I'm fed up with Strava smartphone users, especially off road, posting great times, kom's etc due to the inaccuracy of the phone GPS that's in their back pack or pocket vs that of say a garmin edge that's mounted on the handlebars with a much better view of the sky.

    I've ridden with both a garmin edge 1000 (with wheel speed sensor) and Strava on my iPhone 6 and the 6 constaly shows quicker times, did a ride with someone with Strava on android (in a pocket) and despite him being behind me all the way and me pulling away from him, he managed to beat my kom by a few seconds on his first try.

    Smartphone gps is good for approximate positioning, but not as good as a dedicated gps unit which is designed for gps and he designers have spent the bom on gps stuff not smartphone stuff like phone specific antennas, speakers , microphones etc.

    1. Anonymous Coward
      Anonymous Coward

      I've noticed this.

      I'm using a Galaxy S7 in my backpack, it gets a great view of the sky and the GPS track is bang-on the middle of the track / trail I'm on. However, there's a couple of other riders, whom I know are slower than me (I have ridden with / flown past them en route) yet somehow they're getting KoMs.

      Some of the these are Android phones and some are iPhones, their GPS tracks go all over the place.

  12. the Jim bloke
    Headmaster

    GNSS accuracy

    Apart from minimally adequate hardware components in consumer smartphones, the main source of perceived inaccuracy using satellite navigation is the same as almost every other flaw in the modern world.

    Reality is failing to comply with the way we want things to be.

    Early computer games notwithstanding, the world is not shaped like a donut or toilet roll (depending on whether you could move directly from south to north edge, or just east-west). Despite the lies-we-tell-to-children, it is not a sphere, or an egg, or a (aussie rules/rugby/american) football, or a pumpkin.

    Initiates of the mysteries know that the true shape of the world... is like a potato.

    As this isnt something that can be easily mathematically modeled, a number of best-fit approximations have been created, appropriate to different regions of the potato. A common trap for novices is using the model for some other part of the world. As we accumulate observations on the real shape of the world, these models are refined, and at intervals put into public use - Australia is about to move onto GDA 2020, (apart from our capital which is still running on AGD 66).

    Even these best fit models need further kludges to get close to the shape of the potato, so there is what is called the N value or separation, which is a database of the local differences between the model and the potato, on an 1.8 km grid over all of Australia - (dont know how others handle it)

    None of this was relevant before we started spinning satellites around, or tried measuring back from them to where we were standing.

  13. M7S

    This thing about emergency services needing to know to the nearest metre or so

    None of the response vehicles I operate, or any with which I am familiar (across several services) in the UK have the kind of kit with this accuracy fitted or manportable (along with all the other gubbins we have to carry when out of the vehicle) and kept in some kind of bracket.

    If we get to a location, probably given as "a mile up the B2345, past the Three Kings Pub" then if we cannot find the incident we are looking for there, we night have a bit of a look around, perhaps peek over a hedge or two to see if someone's vehicle has jumped the verge and is resting in a field out of most people's line of sight. I might even, in an urban area, clamber up on to the odd garden wall (if not topped with cemented-in broken glass) and peer into someone's garden.

    It seems that some commentards think that if the incident I am looking for is not within a metre of the location given, that I will not find it. Do they therefore also assume that if my satnav directs me to drive into a canal, that I might not pause and think "hang on a second, this vehicle is indeed a submersible, but only once"?

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like