What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

gps accuracy

bobnoffs

Well Known Member
this is mostly an academic question., specifically for garmin.
when a gps changes a number in its location, say a latitude ... deg, min., sec. how is the gps programmed to do so? halfway between units, when it reaches the next unit?
 
Pure math, Bob. Values increment based on the rules of math.. Keep in mind there's a fair bit of that math stuff going on in the background, including interpolation of raw GPS data against the WGS84 reference datum (or whatever reference datum is in use in your particular GPS) and a few other things. GPS is what is referred to as an "earth-centered, earth-fixed" positioning system that figures out where it is relative to the center of the earth, then uses that info to try to tell us, in terms we understand, where we are relative to the surface of the earth, since most of us don't care how many metres we are from the center of the earth! :)
 
so the display is reading''.......and 7 seconds. as you move you get closer to ''8 seconds''. my question is when does a garmin display change from ''7 seconds'' to ''8 seconds''?

all this relates to 32 degrees,23 minutes, 7 seconds [for example]
 
My guess - just a guess - is that internally it calculates to 0.1 sec of arc or even 0.01? (outside its typical error bar) and then rounds to the nearest second of arc.
 
GPS has lots of challenges in urban environments (reflections, satellite shadowing, etc) ... anyone in a big city curses their phone when that can't do turn by turn navigation among the big buildings.

However, for we aviators, we have a VERY good view of the sky and we should expect better than 10 meters of error in flight (total error .. latitude, longitude, altitude). Generally, GPS receivers compute Earth-centered-Earth-fixed location (X,Y,Z cartesian coordinates, but easily to LLA ... latitude, longitude, altitude). If you have a WAAS receiver, which compensates for delays due to the various charged layers of the outer atmosphere - ionosphere - you can expect better than 5 m of accuracy.

So, that's the internal accuracy of the solution. From there, the SW will present the solution to whatever level of precision they desire for presentation. 1" of arc at our latitude is ~100 ft (30 m) at our latitudes. Well within the precision of the basic measurements.

Bob asks "how is the gps programmed to do so?" ... it computes a 10m accurate (or better) solution and then presents this to the pilot in a useful form. It seems that 0.1" (3m) would be the best precision without extreme measures (like carrier phase differential GPS, that can get to the cm level, however, that's only used for things like construction projects on terra firma)
 
Last edited:
I doubt that there's anything special going on with the rounding for display. As noted, GPS is computed (using a least-squares solution) for X, Y, Z and t in ECEF coordinates. (X, Y, Z) is then converted to (Lat, Lon, HAE) or (Lat, Lon, MSL) using whatever datum is selected (nominally, WGS-84, but some units let you pick the datum; NAD-27 used to be common, at least for non-aviation units).

That's all done in floating-point, usually double-precision, arithmetic using the math libraries from the RTOS (unless one is doing something more special, like quad-precision or writing your own math functions or whatnot).

Then simple math to take it to Deg-Min-Sec when it's displayed. So in all likelihood, the Garmin (and other boxes) are just using whatever rounding method the compiler/RTOS uses when it displays the seconds as a floating-point number. Symmetric rounding is the most likely candidate.

In other words, it's accurate to within 0.5" (if you are display DD MM SS), or 0.05" if your are displaying DD MM SS.S, and so on.

ETA: BTW, if you want to know accurate the *solution* to the GPS position is, look up how to interpret Dilution of Precision...HDOP, VDOP and DOP.
 
Last edited:
Slight thread drift, but there seems to be some people who know such things around

I once asked why my GPS always cut out when doing aerobatics ?

.... "thats because the GPS system is deigned to prevent use of its signals to guide home made missiles"

True ? - or just an urban myth ?.
 
...

.... "thats because the GPS system is deigned to prevent use of its signals to guide home made missiles"

True ? - or just an urban myth ?.
If it were, it would not be the GPS "system" but the chip on the device you are using. The satellites only broadcast data - how it's interpreted is up to the particular GPS chip in your device. Apologies if I'm stating the obvious. :)
 
Slight thread drift, but there seems to be some people who know such things around

I once asked why my GPS always cut out when doing aerobatics ?

.... "thats because the GPS system is deigned to prevent use of its signals to guide home made missiles"

True ? - or just an urban myth ?.

I doubt missiles would be doing aerobatics. More than likely your antenna was just getting blocked by your fuselage in the unusual attitudes.
 
Definitely urban myth...

Your GPS antenna is likely a "patch" type that looks mostly upwards (in normal flight). When you happen to be upside down that antenna points downwards and it cant't see any satellites. Some GPS systems can use dead reckoning in situations like this for a few seconds to estimate themselves out of the blackout but most don't.

Some GPS systems use helix antennas (they tend to be sort of tubular) - these have a much wider aperture and generally handle aerobatics pretty well - but can sill be blanked by your fuselage.

Rainier
CEO MGL Avionics


Slight thread drift, but there seems to be some people who know such things around

I once asked why my GPS always cut out when doing aerobatics ?

.... "thats because the GPS system is deigned to prevent use of its signals to guide home made missiles"

True ? - or just an urban myth ?.
 
Definitely urban myth...

Your GPS antenna is likely a "patch" type that looks mostly upwards (in normal flight). When you happen to be upside down that antenna points downwards and it cant't see any satellites. Some GPS systems can use dead reckoning in situations like this for a few seconds to estimate themselves out of the blackout but most don't.

Some GPS systems use helix antennas (they tend to be sort of tubular) - these have a much wider aperture and generally handle aerobatics pretty well - but can sill be blanked by your fuselage.

Rainier
CEO MGL Avionics

And some GPS antennae can even work on rotating objects. :)

A Tucson product - https://www.youtube.com/watch?time_continue=5&v=70q4ZXkSgJE
 
There are some export regulations on GPS receivers, however, that probably all manufactures will have to address if they want to export the devices:

15 CFR, Subtitle B, Chapter VII, Subchapter C, Part 774

7A105 Receiving equipment for 'navigation satellite systems' designed or modified for airborne applications and capable of providing navigation information at speeds in excess of 600 m/s (1,165 nautical mph), and “specially designed” “parts” and “components” therefor.

(previously, it also had a 59,000 foot ceiling as well, now removed)

... RVs are not quite up to this level of performance ;-)

As posted earlier, the receiver likely lost lock on the satellites during acro due to antenna shadowing by the fuse/earth or due to flight dynamics that the rx tracking loops can't handle.
 
Back
Top