Do GPSr's automatically search for the best satellite geometry?

[+Cachetrotters]

After visiting this page at China Lake and seeing the capability to predict satellite coverage, and giving the resultant four best satellites by number, GDOP, HDOP, etc., I was wondering if there would be an advantage to forcing the receiver to use only those birds. Or do the current commercial(consumer) receivers do this automatically now by tracking all that they see and calculating the 4 best themselves?

 

don

[+EraSeek]

I'd have to search around for the info, but yes, I do believe your unit's programing looks for the best geometry.

 

[+apersson850]

As far as I've understood, they use an overdetermined solution, when possible, which automatically gives you the lowest DOP that can be attained at the moment.

 

Anders

[+paul_stratton]

There's a cool program that shows your GPS data as present when in NMEA mode. There's a page that shows the satellite info that you're receiving. There's even a WINCE version for PPC...

 

Visual GPS

 

paul_stratton

And to think that I once had trouble finding my own "***" with both hands...

[+apersson850]

Since that program, VisualGPS, also plots both position and GPS altitude, it's quite educational. Especially for those who think that a position is a position and nothing else.

 

Let a computer run for a day, with this program running and your GPS connected. Sometimes you can see hilarious moves performed by your house, or wherever you are, at least as they are reported by the GPS.

 

To return to the original question: On my Garmin Vista, the signal strength bars have different colors (shades), depending upon if they are just received, or received and used in the PVT calculation. The same goes for the satellite icons in the sky view. Hence, you can see for yourself if the satellites used are spread all over the sky (good) or just in one corner (bad). This doesn't give any precise figure, of course, but is a good indicator, at least.

 

Anders

[+Cachetrotters]

I have and use SA Watch, and just downloaded the VisualGPS. I'll check it out.

 

don

[+Renegade Knight]

Not that I know of. A GPSr uses the geomotrey available whatiever it is. Buy using as many satalites as it's rated for (most are 12) it gets a better location. I don't know how it picks which 12 if 13 are availabe. Since I've never had a 12 lock (11 max so far while I've been looking) I've had nothing to observe.

 

Wherever you go there you are.

[+apersson850]

Although I've seen it lock onto all twelve channels my Vista has, at the same time, I have no way of knowing if there was a thirteenth satellite over the horizon at that time.

 

The only satellites I know about are those displayed by the GPS, and that's never more than twelve.

 

However, if you enable WAAS/EGNOS, one or two channels are dedicated to the SBAS satellites. In case that more than ten satellites are available at that instance, one or two has to be sacrificed to allow for the SBAS channels. There ought to be some strategy for the unit to determine which to remove, even if it's as simple as it will take those with the smallest ID number.

I assume it will replace one or two sats that are lowest on the horizon, but I don't know.

 

Anyone who knows, or have seen a pattern?

 

Anders

[Kerry.]

GPSr's are "reasonably" clever in what they use and what they don't and has all to do with the logic, which is based on many factors, both broadcast and dynamically computed.

 

Actually the 12 channel receiver was sort of a "mistake" as it was basically meant to be an 11+1 (somebody got the specs a little wrong) in which there was 1 channel, which cycled thru ALL the remaining satellites much like the 4+1 receiver did. Up until a short time ago 14 visible sats was a regular occurance (but not for any length of time).

 

Just because a receiver has 12 channels doesn't necessarily mean it actually processes all data to all sats.

 

Cheers, Kerry.

 

I never get lost everybody keeps telling me where to go

[+apersson850]

One could perhaps imagine that different manufacturers and even different models from the same manufacturer don't use exactly the same strategy either, when it comes to selecting which one to omit?

 

Anders

[+Renegade Knight]

To answer the original question.

 

The geometry changes all the time. To force it to only use certain satalites that were good now, would force it to use them when they were not so good.

 

Wherever you go there you are.

[+EraSeek]

When using WAAS, two slots are dedicated to those WAAS sats. WAAS sats transmit info to correct for signal delay and such, but the can also be used for navigation like other satellites. When locked onto a WAAS sat, you are gathering the correction info, but I don't think this means you are nessasarily using the WAAS sat for navigation. You may or may not be depending on configuration. Does anyone know for sure?

 

[Kerry.]

If the WAAS Geo is transmitting a "use for navigation" message then it can basically be used as part of the position determination just like the "real " GPS sats.

 

I tend to get the impression that the Geo's at present can be a little unstable at times and probably don't get flagged for positional use all that much?.

 

Cheers, Kerry.

 

I never get lost everybody keeps telling me where to go

[+apersson850]

When they did a test run of the EGNOS system at November 25-26, the satellite icon for AOR-E (Garmin #33) became black after a while. Black icon is supposed to mean "part of the PVT solution".

Long before that, the "D" signs showed up at the other satellites' signal strength bars, to indicate that corrections for those were indeed received. Also, a "D" may show up for the SBAS satellite itself.

 

Anders

 

[This message was edited by Anders on January 24, 2003 at 12:00 AM.]

 

[This message was edited by Anders on January 24, 2003 at 12:01 AM.]

[+EraSeek]

Yes, of course, I guess that would answer it. I have seen ''D's'' without the bars, and also with black bars on both 35 and 47 indicating being used for navagation.