Jump to content

GPS Accuracy


GSHastings

Recommended Posts

I'm wondering why our hand held GPS devices are only accurate to about +/- 9ft ?

 

In farming, they use GPS tracking that is accurate within a few inches. For instance, they can run the drill (the equipment that plants the seeds with fertilizer), so accurately, that they can offset the rows from year to year, so that each year they are planting between the rows from the previous year. In the case of wheat (I own a small wheat ranch), that is an offset of only about 8 inches.

 

Why are our expensive hand held units so much less accurate? I have a Garmin GPSMAP 64st, and the best is shows is 9 ft accuracy (which I assume is essentially a 9 ft diameter radius).

 

Thanks,

Greg

Link to comment

Maybe I'm not understanding, but sure you aren't confusing GPS with GIS , or (in this case) agricultural mapping ?

 

GPS "accuracy" hasn't changed since Clinton opened it to civilian use in 2000, when selective availability was removed.

"Accuracy" has been around 10 feet on a "perfect day" since. 

 - "Perfect days" are rare, and why most put that GPSr away about 25 feet or so, heading to GZ.    :)

Link to comment
47 minutes ago, GSHastings said:

I'm wondering why our hand held GPS devices are only accurate to about +/- 9ft ?

 

In farming, they use GPS tracking that is accurate within a few inches.

 

The most obvious reason is that a basic handheld GPS receiver is not linked to any local ground references. With a local ground reference, GPS was accurate even before the selective availability was removed.

Edited by arisoft
Link to comment

Those farm units... how much do they cost? I can't imagine that those systems go for the same cost as as consumer grade GPS. Even a survey-grade GPS from Trimble will set you back a couple thousand dollars, not a couple hundred.

Also airsoft's response. I'm under the impression that ground reference stations are set up around farm fields that assist with the precision and accuracy of the equipment being autonomously  deployed. 

Link to comment
2 hours ago, GSHastings said:

In farming, they use GPS tracking that is accurate within a few inches. For instance, they can run the drill (the equipment that plants the seeds with fertilizer), so accurately, that they can offset the rows from year to year, so that each year they are planting between the rows from the previous year. In the case of wheat (I own a small wheat ranch), that is an offset of only about 8 inches.

 

Why are our expensive hand held units so much less accurate? I have a Garmin GPSMAP 64st, and the best is shows is 9 ft accuracy (which I assume is essentially a 9 ft diameter radius).

 

They use special equipment including referenced base stations, and all this comes at a cost.  Here's an example.

There exists other high-end equipment that has improved accuracy, which you could buy (but it doesn't clip to a backback). Even with those, the best case scenario is quoted, but in practice, it varies. When you ask why "expensive" consumer devices aren't as "accurate" as much more expensive systems, the answer is, you get what you pay for. 

 

The accuracy (we tend to mean "precision") of a consumer GPS receiver is good enough for Geocaching.  If I get "within 9 feet" of the cache, now I use the hint, and look for hiding spots.  I can often whittle that down to a single point, once I arrive.

 

If you're waiting for new technology before you go find caches, the good news, technology is always changing:

https://www.theverge.com/circuitbreaker/2017/9/25/16362296/gps-accuracy-improving-one-foot-broadcom

 

Edited by kunarion
Link to comment

With a typical handheld GPSr used for geocaching, I went to the same spot in my driveway and recorded one waypoint a day for twenty days.  I then assumed that the horizontal locations of the points were normally distributed about the calculated center.  Characterizing the precision, or repeatability, as the standard deviation of the collection, the result was ±3.23 ft.  In the context of a two dimensional distribution, 68% of the waypoints would be encompassed by a 6.46 ft. diameter circle.

 

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

 

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

Link to comment

Withi GNSS there is GPS, GLONASS an Galileo satellite  positioning systems. GPS American owned with accuracy to 5m radius, GLONASS Russian owned with accuracy to 2.8m radius and the EU Galileo is 1m radius. However even though the UK has put billions into the EU Galileo system following Brexit we'll not have location access to it. There business GPS systems that can access licenced GPS access to closer radius though not as tight accuracy available to the military.

Edited by kenrw
Link to comment
On 11/19/2018 at 8:51 AM, kenrw said:

Withi GNSS there is GPS, GLONASS an Galileo satellite  positioning systems. GPS American owned with accuracy to 5m radius, GLONASS Russian owned with accuracy to 2.8m radius and the EU Galileo is 1m radius. However even though the UK has put billions into the EU Galileo system following Brexit we'll not have location access to it. There business GPS systems that can access licenced GPS access to closer radius though not as tight accuracy available to the military.

Beware apples and oranges.  The newest breed of satellites still uses ground based stations to improve the accuracy, but manage the additional information independently (no need for special satellites to broadcast the correction information).  You appear to be spec'ing  the American GPS system as though there were no ground based coordination being done.  See http://www.nstb.tc.faa.gov/REPORTS/waaspan17.pdf

 

 

Link to comment

I was basing accuracy on the public view using satellite navigation.

It is true that ALL satellites require ground stations to perform adjustments in the background so that users get their preset accuracy whether it be civilian or military use.

To quote from Gallileo website which most likely be the same for all the systems.

 

"The orbits of the satellites can be slightly nudged by the gravitational tug of Earth’s slight equatorial bulge and by the Moon and Sun. Even the slight but continuous push of sunlight itself can affect satellites in their orbital paths. The quality of signals received on the ground can be affected by their transit through the ever-changing ionosphere, the electrically active outer layer of Earth’s atmosphere.

Galileo sensor stations, with small omnidirectional receiving antennas around just 50 cm high, are on place around the globe to check the accuracy and signal quality of individual satellites in real time, and work together to pinpoint the current satellite orbits.

These measurements are transmitted via secure satellite communications to Fucino, where they serve as the basis of a set of corrections – accounting for timing or orbital slips – to be uplinked to the satellites via a network of 3 m-diameter uplink stations for rebroadcast within navigation messages to users, currently updated every 50 minutes."

Link to comment

It is odd to me but I find that if, under normal terrain circumstances, my GPSr comes in +/- 3-6' of GZ but I typically look at the difference between what the unit says is the distance to the stated coords (GZ) and the Error distance, often the true GZ is pretty darn close

 

However, there has been no mention of the CO's unit used when placement and this has considerable effect on the coordinates as compared to another type of unit. So, as someone noted above, the is a time one just needs to put the unit away and "think"

 

My $.02

Link to comment
On 12/9/2018 at 5:11 AM, Z_Statman said:

It is odd to me but I find that if, under normal terrain circumstances, my GPSr comes in +/- 3-6' of GZ but I typically look at the difference between what the unit says is the distance to the stated coords (GZ) and the Error distance, often the true GZ is pretty darn close

I suspect this is what most people find. I do think it's interesting to keep that 20' error in mind, but it's just a worst case consideration that assumes both you and the CO haven't been able to get a more accurate position via averaging. Thanks to most COs in my area being very careful, I rarely find the cache more than 10' from where my GPSr points, and I start to mention it in my log if it's 15' off.

Edited by dprovan
wrong word
Link to comment

Accuracy is a human failure.  Precision is a machine’s virtue.
 
Precision: GPSr’s are mini-computers and their CPUs are very precise.  In other words, two + two will always equal four, it will not equal five no matter how much the user thinks it should.
 
Accuracy: It all starts with engineers (yes, they’re humans) who are great compromisers seeking a balance between usability, accuracy and cost as they try to tame nature.  It ends with users (most are human) who don’t read instruction manuals, have short attention spans and cry when they can’t get what they want.
 
The User's Options:  Usability, accuracy, or cost - pick the one.

 

  • Upvote 3
Link to comment

Someone correct me if I'm wrong here:

When a GPSR communicates with a single satellite, that satellite doesn't actually know the direction of the GPSR, only the distance. And this distance is calculated based on a ping delay - the GPSR sends a ping to the satellite, the satellite returns a ping and the time between send and receive dictates the distance. The satellite's location is known, so your receiver lies somewhere on a circle with the radius of the distance calculated around that satellite.

So a single satellite won't do. Communication with a second satellite will produce another ring around its location which hopefully intersects with the ring around the first satellite. However when two circles overlap, their circumference will cross each other in two locations. So a third satellite is needed to discern which intersection is your location.

The thing about pings, though, is that they are not a constant. Slight variation in processing times, packet loss, data congestion, radio signals bouncing off of other objects can all make the ping time longer or shorter, generally with some random normal distribution. Thus the distance to each satellite is an estimate, as is the mean overlaping location across all satellites. That's where the position accuracy error comes into play. It's actually a statistical standard error, and so that error should decrease if the variance in ping time to the satellites is small, and/or with more samples. So connecting to 4 satellites will reduce the standard error of the mean around your location estimate compared with 3 satellites. And so forth. Getting a strong lock on all 12 satellites visible in the sky at any given location and time will produce the smallest standard error around the estimated mean position.

Of course accuracy can be thrown off in other ways, such as if a satellite's position isn't properly calibrated. This is could explain why you might have a small error margin, but still have your location reported several meters away from where you actually are. I suspect that when people (incluing myself) see a drop in accurate location reporting with GLONASS turned on, I suspect it's because the Russians are calibrating their satellite locations differently than our own satellites.

http://www.physics.org/article-questions.asp?id=55

Link to comment

The GPSr sends nothing.  It is only a receiver. The satellite signal states a time and location of the satellite.  The receiver knows when the signal was received.  The time difference tells the receiver the distance from the satellite.  The frequency shift in the signal tells it the difference in speed between the two.

 

With three signals, the receiver can calculate its horizontal location by triangulation and its ground speed.  With four signals, it can determine elevation.

 

There are a variety of items that cause error in the reading.

- Raw ability to determine distance from time at the distance covered.

- Signal delay from the atmosphere.

- Bounce of the signal off of the ground or another object.

Edited by Red90
  • Helpful 1
Link to comment

Differential GPS removes the atmospheric error component by having a fix base station nearby that can determine the exact local atmospheric error.  This gets you to the sub meter range.

 

https://racelogic.support/01VBOX_Automotive/01General_Information/Knowledge_Base/How_does_DGPS_(Differential_GPS)_work%3F

 

You can also read about RTK which addresses the ability to measure the time difference enough to get to the centimeter accuracy range.

Link to comment

0?Expires=1544554071&Signature=Tr9MjlQM1

 

Quote

This 24 hour position scatter plot shows a commercial engine in red (Sat Nav's,Mobile Phones, etc.), an un-aided survey grade in blue (VBOX units), aided by SBAS corrections in green, aided by a 20 cm base station in purple, and you can just make out an RTK 2 cm aided system in yellow!

 

Link to comment
On 12/11/2018 at 11:03 AM, Red90 said:

The GPSr sends nothing.  It is only a receiver. The satellite signal states a time and location of the satellite.  The receiver knows when the signal was received.  The time difference tells the receiver the distance from the satellite.  The frequency shift in the signal tells it the difference in speed between the two.

 

With three signals, the receiver can calculate its horizontal location by triangulation and its ground speed.  With four signals, it can determine elevation.

 

There are a variety of items that cause error in the reading.

- Raw ability to determine distance from time at the distance covered.

- Signal delay from the atmosphere.

- Bounce of the signal off of the ground or another object.

 

That's not how I understand it.

 

Satellites (or space vehicles) transmit simply their ID information, the date/time from the onboard clock, and almanac/ephemeris information (basic satellite location information/detailed satellite location information). The satellite does not send it's current location.

 

The GPSr compares the transmitted time from the satellite to the current GPSr time. The GPSr then calculates where that specific satellite was supposed to be at the transmitted time, and uses the received time to determine how far the GPSr was from that satellite at that time. 

 

A minimum of three simultaneously present satellite signals is necessary to perform a basic triangulation of the current GPSr location. This alone is sufficient to determine latitude, longitude and elevation for the GPSr.

 

However, the satellites use very expensive and very precise atomic clocks, which our GPSr are not equipped with. This creates sufficient error to make an accurate determination of the GPSr current location, thus the need for a fourth satellite signal, which then allows the GPSr to compare the 'time' in the received signals of the other three to the 'time' from the fourth satellite (which is far more accurate than the GPSr internal clock) to calculate the accurate 3D position we desire.

 

WAAS and EGNOS are forms of SBAS which are also used by the GPSr to further refine its location accuracy.

 

See FAQ 14 at GPSrChive > GPSMAP 66 for more information about these systems and how they work.

Edited by Atlas Cached
typo
Link to comment
7 hours ago, Mineral2 said:

Someone correct me if I'm wrong here:

When a GPSR communicates with a single satellite, that satellite doesn't actually know the direction of the GPSR, only the distance. And this distance is calculated based on a ping delay - the GPSR sends a ping to the satellite, the satellite returns a ping and the time between send and receive dictates the distance. The satellite's location is known, so your receiver lies somewhere on a circle with the radius of the distance calculated around that satellite.

 

Satellites and GPSr do not have two way communication, or any communication for that matter.

 

The GPSr simply listens to the signal being transmitted from any given satellite within range. The satellites know nothing of your GPSr, nor do they care.

 

The satellite is a transmitter.

 

The GPSr is a receiver. (thus: Global Positioning System receiver).

Link to comment
14 hours ago, Atlas Cached said:

That's not how I understand it.

 

You then go on to repeat exactly what I stated. You do not need to argue against everyone in every post.

 

To clarify...the ephemeris  is NOT a "basic outline of where and when the satellite is supposed to be".  It is the precise information of where the satellite is located at all times.  If it was not precise, the location calculations would not work.

 

  • Upvote 1
Link to comment
3 hours ago, Red90 said:

You then go on to repeat exactly what I stated. You do not need to argue against everyone in every post. 

 

On 12/11/2018 at 8:03 PM, Red90 said:

With three signals, the receiver can calculate its horizontal location by triangulation and its ground speed.  With four signals, it can determine elevation.

 

This was not correct and not the same.

 

Four satellites are needed to get the fix. Without a fix no coordinate information can be calculated at any plane.

Edited by arisoft
Link to comment
35 minutes ago, arisoft said:

This was not correct and not the same.

 

Four satellites are needed to get the fix. Without a fix no coordinate information can be calculated at any plane.

 

Nope.  The fourth in that case is use to fix for any timing error, so some receiver will not show a fix until four are seen.  This, though is not mandatory and some, especially old unit will fix with three satellites. I stand by what I state.  You guys don't need to be pedantic nitpickers.  We are just trying to give a layman's understand of the basics.

 

You see, I've probably been around longer than you guys and quite clearly remember waiting for a location fix while watching the satellite pages.

 

http://retro-gps.info/Manuals/downloads/files/Garmin GPS 38.pdf

Quote

In order to determine a position fix, your GPS needs to continuously "see" at least three satellites.

 

See page 13

http://static.garmin.com/pumac/GPS12_OwnersManual.pdf

 

Page 38

https://static.garmincdn.com/pumac/GPSMAP60CSx_OwnersManual.pdf

Edited by Red90
  • Upvote 1
Link to comment
1 hour ago, Red90 said:

The fourth in that case is use to fix for any timing error, so some receiver will not show a fix until four are seen.

 

Yes, if you have an atomic clock with you, you do not need the fourth satellite. But your commercial grade receiver unfortinately have only a commercial grade chrystal oscillator which is as accurate as a wristwatch. No handheld GPS receiver have an integrated atomic clock.

Link to comment

This thread is a very interesting read for someone without a ton of knowledge about this.  I do wonder what people think is the best setting to use on the newer GPSrs that have GPS, Glonass, and Galileo settings?

I have seen posts that say GPS +Galileo is not as good as GPS + Glonass because the Galileo system is newer and not all the eventual satellites are online yet.  That makes sense.  BUT, I have also seen lots of people post that they get the best accuracy on their tracks, etc. by setting their units to GPS only.  Why could that be?

How could taking Glonass out of the equation improve accuracy?  Isn't it the more satellites, the better?  Except that I also read that using only GPS results in better battery life?

Edited by Cheminer Will
Link to comment
12 minutes ago, Cheminer Will said:

I have also seen lots of people post that they get the best accuracy on their tracks, etc. by setting their units to GPS only.  Why could that be?

 

One explanation is that there is a problem with the device itself - a bug. Theoretically more satellites means more accuracy.

Link to comment

That is what I thought but I see many saying otherwise.  Like this for example saying too many satellites can cause issues with accuracy.

 

 

Quote

 

From I have read the best is one single source. So for example if you have a good connection to GPS go with GPS only and don’t turn on GLONASS or Galileo.

Basically if there are a lot of satellites sourcing into your chipsets and if the satellites are closer together it can cause...

 

Quote

 

 And I read that in higher latitudes (north or south - think Russia or Australia) go with GPS+GLONASS since GLONASS covers a higher latitude.

While in low latitudes, GPS performs better and gives much better accuracy than the GLONASS, because of its high coverage over the globe.

 

 

That info just came off one site, but I have seen similar statements elsewhere.  I guess testing in  your area is best to determine what is best.  But that could change day to day I think.  So maybe the best approach is to see what might be best overall for your area and just leave it set that way rather than changing back and forth?

Edited by Cheminer Will
Link to comment
12 minutes ago, Cheminer Will said:

That is what I thought but I see many saying otherwise.  Like this for example saying too many satellites can cause issues with accuracy.

 

 

All problems can be tracked down into the receiver software and device capabilities. For example... some receivers may have very limited number of satellites they can receive. My first GPS receiver could receive only 4 satellites simultaneously. It was a worthless unit for that reason. If your unit can receive  only 12 satellites at once, it is better to use GPS satellites as they most propably spreads evenly. Accuracy suffers if most satellites are in one direction which is possible if the receiver can not handle all of them simultaneously.

Edited by arisoft
Link to comment
9 minutes ago, Cheminer Will said:

I see, so you think with a newer unit like the 66 series, there would be no reason to set it to GPS only, but just keep it on GPS + Glonass?

 

Only it there was a bug in the design which have been fixed in the newer device. It is easy to test this hypothesis if you have opportunity to use these devices. You need a tape measure. Mark at lest two spots on the ground and use the receiver to measure the distance from one spot to the another and compare in which mode the results are better.

Edited by arisoft
Link to comment
38 minutes ago, Cheminer Will said:

I have also seen lots of people post that they get the best accuracy on their tracks, etc. by setting their units to GPS only.  Why could that be?


Many smartphones use a Wi-Fi Positioning System (WPS) in combination with GPS to approximate their location in cities, where the GPS signal bounces a lot on buildings.

Wikipedia:

Quote

Wi-Fi positioning system (WPS) or WiPS/WFPS is a geolocation system that uses the characteristics of nearby Wi-Fi hotspots and other wireless access points to discover where a device is located. It is used where satellite navigation such as GPS is inadequate due to various causes including multipath and signal blockage indoors, or where acquiring a satellite fix would take too long. Such systems include indoor positioning systems. Wi-Fi positioning takes advantage of the rapid growth in the early 21st century of wireless access points in urban areas.


In regions where no accurate mapping of Wi-FI hotspots is available, WPS only introduces noise. Garbage In, garbage out. It's best, in those situations to use GPS exclusively.

Link to comment
1 hour ago, arisoft said:

 

Yes, if you have an atomic clock with you, you do not need the fourth satellite. But your commercial grade receiver unfortinately have only a commercial grade chrystal oscillator which is as accurate as a wristwatch. No handheld GPS receiver have an integrated atomic clock.

 

Good lord buddy.  I provided link to the Garmin manuals showing the provide a fix with three......

Link to comment

Mathematically, 3 readings are enough to pinpoint a single location. It's the physics of GPS that requires the 4th to hone the final location to high accuracy.

 

Locating in 3D space, basically:

1 source distance reading = sphere

2 sources = circle

3 sources = single point (lat/lon and alt) +/- margin of error

4+ sources = reduced margin of error

 

Link to comment
59 minutes ago, thebruce0 said:

Mathematically, 3 readings are enough to pinpoint a single location. It's the physics of GPS that requires the 4th to hone the final location to high accuracy.

 

Locating in 3D space, basically:

1 source distance reading = sphere

2 sources = circle

3 sources = single point (lat/lon and alt) +/- margin of error

4+ sources = reduced margin of error

 

Well... in 3 dimensions,  3 sources = two points along the circle from 2 sources, and 4 sources = single point. But, the earth acts as the 4th sphere with the alternate point being somewhere other than the earth's surface. So we can get away with 3 sources to plot our position. I think 3 satellite sources will get you a decent latitude and longitude, but the 4th source will give you acceptable altitude. And of course, 4+ reduces the margin of error. 

Link to comment

True, yes the Earth can be considered the 4th sphere; basically the two intersecting points of 3 spheres only has 1 reasonable point at/near the Earth sphere, but that point has X/Y/Z attributes, so it technically provides altitude along with lat and lon, even if it's not that accurate. Well, it provides a point, and the altitude data referenced at the lat and lon provide the data needed to calculate relative altitude from sea level :P

A 4th satellite sharpens up that accuracy, basically taking the average centerpoint of the collision of each of the spheres. The more satellites provide data, the sharper that calculation will be.

Link to comment
10 hours ago, Red90 said:

 

You then go on to repeat exactly what I stated. You do not need to argue against everyone in every post.

 

 

 

No, sir. The devil is in the details, and I did not repeat your details.

 

This is an excellent conversation to be having, I think a lot of readers are going to learn a great deal here!

 

8^)

Link to comment
10 hours ago, Red90 said:

Good lord buddy.  I provided link to the Garmin manuals showing the provide a fix with three......

 

When you start to solve puzzle caches, the first fact you will learn is not to trust any source of data because there is always a better and more reliable source of data available to be found.

 

In this case a better source is a patent applied for the GPS system.

 

Edited by arisoft
Link to comment
On 12/13/2018 at 6:26 AM, Red90 said:

 

My degree in mathematics says otherwise.  3 satellites provide a hoziontal (2D) location.  Four to give altitude (3D).  With 3 satellites, they assume altitude. 

 

Come on now.

 

Altitude is used when you are above (off) the ground. (Airplane pilot cruising at xxxxx altitude)

 

Elevation is the ground level above mean sea level. (Road signs on mountain roads read "Elevation: yyyyy")

Edited by Atlas Cached
Link to comment

I'll add my weisenheimer comment as the final word. :D

Normally only three signal scorches are needed to determine a position in 3-space.  However, there is a time element that must be considered because only the atomic clocks onboard the satellites are synchronized. The earthbound receiver has a less accurate quartz timer and it must be in sync with the satellites to achieve accurate results.  A 1 µSec time error will produce a 300M position error.  Thus there are four variables that require four independent equations which are derived from four separate satellites.

Longitude (X), Latitude (Y), Height (Z) and Time error (T)

Despite the inaccuracy on the part of the GPSr clock, an accurate position can be calculated with only four satellites.

Link to comment
1 hour ago, Red90 said:

 

My degree in mathematics says otherwise.  3 satellites provide a hoziontal (2D) location.  Four to give altitude (3D).  With 3 satellites, they assume altitude. 

 

Then you can solve the given mathematical problem using only 3 of 4 references. Did you already solve it?

Edited by arisoft
Link to comment
42 minutes ago, Capt. Bob said:

I'll add my weisenheimer comment as the final word. :D

Normally only three signal scorches are needed to determine a position in 3-space.  However, there is a time element that must be considered because only the atomic clocks onboard the satellites are synchronized. The earthbound receiver has a less accurate quartz timer and it must be in sync with the satellites to achieve accurate results.  A 1 µSec time error will produce a 300M position error.  Thus there are four variables that require four independent equations which are derived from four separate satellites.

Longitude (X), Latitude (Y), Height (Z) and Time error (T)

Despite the inaccuracy on the part of the GPSr clock, an accurate position can be calculated with only four satellites.

 

Yes.  I already stated that earlier.  However, as also mentioned....manufacturers such as Garmin, as referenced....will provide a location fix with three satellites.  The error in the clock is only whatever change happened since it last had a signal.  Thus the timing error becomes dependent on the time since last use and the accuracy of the internal clock.

 

In today's world. none of this matters as modern units wait for four satellites and normally lock on many more within seconds and you never see this.  Perhaps, I'll video the GPS38 as it acquires signal as it takes 10 to 15 minutes...  You can watch it show the location with three satellites.

Link to comment
Just now, arisoft said:

 

I know that I can solve it because I have already found the cache but can you solve it with only 2 references of 4? That is the question.

 

You need three as stated to have one definitive lines in space.  With two, you have a choice of two lines.  You can solve it, but you get two possible solutions.

Link to comment
7 minutes ago, Red90 said:

You need three as stated to have one definitive lines in space.  With two, you have a choice of two lines.  You can solve it, but you get two possible solutions.

 

You don't have to explain your solution. Just find the coordinates for the cache. There is a GeoChecker on the cache page you can use to check that you have the correct result.

Link to comment
29 minutes ago, Red90 said:

The error in the clock is only whatever change happened since it last had a signal.  Thus the timing error becomes dependent on the time since last use and the accuracy of the internal clock.

 

We know the accuracy of the internal quartz clock. It is about +-20ppm. In this case, the effective drift is something between -300m*20 to +300m*20 per second. In the worst case the position will move about 6000 meters (over 3 miles) in one second, if there is no fix to at least four satellites to compensate this error.

Link to comment

The accuracy of a quartz oscillator can be far better than +- 20ppm. Inexpensive temperature compensated quartz crystal modules can be accurate to around +- 2 ppm when exposed to the typical ranges of temperature one would experience outdoors, and these consumer very little additional energy than a simple quartz crystal. To get even better accuracy from a quartz oscillator you can use a oven-ized quartz oscillator module, but these are larger and far more energy intensive.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...