Aaaaaaargh! QueTrip...here we go again.

[Bokkie]

Cast your mind back a few months where I had an ongoing session with QueTrip also known as Dances With Satellites? I'm happy to report that I have the same problem with the cfQue. The screenshot proves it. I was standing still, honestly, I never moved an inch, not a millimeter, not a step in any cardinal position on the compass.

Yet, QueTrip continues to show that I am still moving. I have a theory though. Those aliens that have been causing such mayhem (especially with the outbreak of Swedish Silliness)? Could it be they have levitated me slightly above terra firma and are in fact nudging the world under my feet, or maybe they planted a nanorobot in my cfQue?

Answers again, not on post cards, but this time emails to the one-eyed aliens in the spaceship...

[Bokkie]

Darn! Those aliens are trying to suppress the truth and they nuked my screenshot. Maybe this time...

[apersson850]

That it detects movement, in spite of you standing still, is of course due to errors in measurements. Happens to any kind of insturment.
How it can get the idea that you have a speed, but not during any time move, is more tricky to explain.
Still, silly or not, I can think of two plausible causes for such a report:

It's as we've been told. Position change is determined by the PVT algorithm, basing its information on the correlation, and thus locking, of the Pseudo Random codes in the signals sent from the satellites. In this case, it's move so little, that the iQue has filtered it out as noise.
But speed is determined by the Dopplereffect on these signals. Admittedly, it's derived from mathematic operations on the very same signals, but perhaps that calculation did get a resulting speed, that was high enough to register in the average speed in your QueTrip?

The other theory is that there has been a position change, very quickly away, and then back home again. All in error, due to measurements, of course, but fast enough to contribute to the average speed, but short enough not to be listed as moving time.
But since the resolution of the movement timer is one second, and the iQue is supposed to update the position once per second, it doesn't seem likely, does it? Unless it does use one of these GPS position announcements, which are caused by a transition from 2D to 3D, vice versa, or (in the case with the aliens) from 3D to 4D.

How's that for silliness?

[Bokkie]

If anyone cares to decode that and describe it to me in terms of monsyllables I'd appreciate that. All I know about Doppler is that if a man stands on a train and farts, then it has a different frequency coming towards you compared to when it goes past. The only constant factor is the smell.:eek:

[stephanpls]

Originally posted by apersson850
...But speed is determined by the Dopplereffect on these signals...

Do you mean the iQue can measure frequency changes (the concept of Doppler measurements) in the satellite signal and process them into a speed? I hardly can believe this, or are you kidding us?

I think the iQue (and all down to earth GPS receivers) derives the speed from position changes during a certain number of seconds. This way the speed is derived precisely enough due to the fact that subsequent positions calculated by a receiver have a high degree of correlation. I mean if position A has a say 3 m deviation, in a certain direction, from the real position, position B will have similar deviations. Calculating the speed between A and B will therefore be almost without an error.

[apersson850]

I thought just like you, Stephan, that this Doppler thing can hardly be done by such a receiver. But I've seen, somewhere on the knows-it-all internet, the explanation of how it's done. Not that I remember it precisely enough to repeat the explanation, but these values came out, more or less as a side-effect of the other calculations, required to compute the PVT anyway.

So, yes, it certainly can measure the Doppler effect, but not in the most obvious way a physician would think of.
Garmin themselves claim that speed is computed as a combination of change in position over time, and this Doppler effect.

[stephanpls]

Wow! So the iQue measures satellite signal frequency changes.

This makes me admire that little gadget called iQue even more.

[Bokkie]

Can I remind you reprobates, that this is my thread. I started, so please remain on topic - just like I always do.

[apersson850]

Here is a link, for those interested (aka geeks):

http://www.aprs.net/vm/gps_cs.htm

There's much more on the net, but it's all rather mathematical.
Try google on things like Kalman, gps and interferometry, for example.

Basically, the doppler effect is something you get out of the Kalman filtering, which is applied to the GPS signals, as part of the correlation process, required to determine the distance to the satellites.

[stephanpls]

Originally posted by apersson850
... the doppler effect is something you get out of the Kalman filtering ...

Hmmmm. Really? A Kalman filter gives a position estimate based on A. measurements (be it GPS, Loran-C etc) and B. the previous position, where A. and B. are weighted in a way. I do not see how a frequency shift can be the outcome of a Kalman filtering. I rather think it is input in the Kalman process.

But, Anders, you read all that stuff, I don't (time nor guts) and I only refer to my knowledge gathered as a student quite some time ago, knowledge that's getting more and more rusty.

Maybe I slept those days while the professor talked about Doppler shifts and Kalman :D

[apersson850]

From the link I posted above:
More modern receivers take all the PR+PRR data from all the N satellites in view for the past T seconds and feeds the 2*N*T PR+PRR samples it into a single mathematical "black box" (BB) (usually a Kalman filter) to produce an over- determined estimate of the same 8 parameters. So in modern receivers, this BB is using both the combination of past & present PRs and PRRs from many satellites to improve the Position, Velocity & Time (PVT) estimate. So Paul's statement about velocities being determined by changes in position is sorta, partially correct, but (when you look at the equations inside the BB), the measured "apparent Doppler" frequencies are even more important.
Could be they are lying. If they are, I'm too.
But it could also be that the Kalman filter is one part of the "box", and the frequency shift another.

[Bokkie]

I'll go away from my thread, before anyone realises I'm around. I'd hate to get in the way of you two.:D

[apersson850]

Please don't. We need someone to get us started.

[stephanpls]

Sorry Bokkie, one final phrase only!

From

"... velocities being determined by changes in position is sorta, partially correct, but (when you look at the equations inside the BB), the measured "apparent Doppler" frequencies are even more important...."

my interpretation is that the Doppler frequencies are used in the Kalman filter (what I expected) rather than being an outcome.

[jonasolof]

Bokkie, I believe the reason for the result you get is that your GPS (antenna) is not working well. So the positioning calculation is quite insecure.

LAst easter I tried QBuddy speed indications wjhile driving down to southern France. I was amazed of how sensitive the speed indications ewere when slowly, slowly moving ahead in a line before a toll road pay boot. Still, no speed could be measured as soon as the car was brought to a stand stil.

So where can one see the actual positioning change in high resolution? In GPS acreage, on auto. The author had also made a version for me with extra high resolution, i.e. 32 meters accross the 160 pixels of the screen. 1 dm - one pixel.

Now, trying this on an open field movements where quite small and qBuddy did not indicate any speed (Gilsson connected).

Next test was in a forest under canopy. Swings were much larger and QBuddy did at times indicate (a low) speed).

I posted about this last spring.

Having a badly functioning antenna with signals under 40 dB (three lines down in bar chart) appears to give a result similar to standing under canopy with a good antenna. Thus, you get movement where there is none.

On doppler: This has been up a zillion times in scigeosatnav. To what extent doppler measurement is included in speed measurement is hidden in the proprietary GPS software of various brands. The discussion usually starts with somebody refusing to believe it and then some recognized pundit coming in and pointing at the hard facts.
Try a search for doppler at groups.google.com

Bokkie, Garmin decided that you don't need more sensitive reception than you got. You could take a better GPS, such as a GPS 60C or an iQue with an external antenna and compare it to the fQue. Do they all show movement at standstill??

[apersson850]

So now I can add "pundit" to my business card! :D

[Curious Cat]

Here's a link where I clearly explained, more then I knew ;) , the Doppler Effect within the context of GPS position location.

Here (http://www.pdastreet.com/forums//showthread.php?threadid=39171)

[stephanpls]

Originally posted by apersson850
So now I can add "pundit" to my business card! :D

Which pundit do you opt for? According to a Dutch vocabulary it has two meanings:

1. Expert.
2. An arrogant person who always knows better.

:D :D

[Bokkie]

Originally posted by apersson850
So now I can add "pundit" to my business card! :D

If you two don't shaddup you can add something else to your business card.:mad: :D

Thanks Jonas for the update on this. I don't know why but I can't find anything on the cfQue that lets me disable the accuracy circle. This might be useful or not but it seemed to show a greater potential to indicate movement in QueTrip when the circle was much larger. Of course, by the time I switched to the GPS info screen and back to the map the circle had changed and I suspect reduced in size as more accurate computations kicked in?

I tried another fix a little while ago, and I got a 2D fix in 11 seconds from the time I enabled the GPS acquisition, and a 3D fix in about 23 seconds. An interesting thing about the cfQue is that you cannot switch it off like the iQue by simply folding it away. You have to disable on the GPS screen. Otherwise you can be in a building for a few hours and it will still try to lock on. I suppose the GPS 10 will work the same way with Bluetooth.

[rkevwill]

ummm waitjustadamnminute here. Let me propose a rather simple minded foolish reason. The earth rotates. The satellite is in orbit and moving. Isn't it reasonable to assume one or the other might be moving JUST a touch out of sync from second to second, minute to minute, along with the aforementioned variances?

Just an observation from a NON scientific mind.

[Sho-Bud]

Bokkie,

It's a proof of your age: your hands are trembling :)

[apersson850]

Regarding pundit:
I'm not sure which of the definitions that fit best, but perhaps I qualify for both?

Regarding accuracy circle:
If you have lock-on-road enabled and the unit has found that you are indeed travelling along a road, then the accuracy circle will disappear. The iQue is a bit special, in that you can select to show or not to show the accuracy circle. My old Vista doesn't have such an option either, but displays it on its own discretion, when it hasn't locked on to a road.
Note that it may not lock on to any road, even if that option is enabled, if you aren't near any road, or moving in a directions that's perpendicular to the road.

Regarding rotating earths and such:
Yes, that happens every so often, but very precise measurements of these variations go on continously. Data is then transmitted to the receiver, to inform it about the current facts, not just the theories. Corrections to the satellite's orbits are also done, if necessary, to get them back on track.

Regarding Bokkies age:
I'll put my comments in brackets.
[ ]

[stephanpls]

Originally posted by apersson850
Regarding pundit:
I'm not sure which of the definitions that fit best, but perhaps I qualify for both?


Joke, joke, Anders.
Of course you are an big expert in this field, and not someone who always knows better and certainly not an arrogant person.