Re: how to forward RTCM data to gpsd?

[Greg Troxel]

"Gary E. Miller" <gem@rellim.com> writes:

>> A note concerning the accuracy:
>> In the tests I conducted, GPS + dgps resulted in the roughly the same 
>> accuracy as GPS + Galileo + glonass + egnos.
>
> As expected.  That result is why not a lot of effort has gone into
> improving that code.

@kristoff: Note that if true, this is showing that the RTCM2 corrections
are good, because a single constellation and DGPS is even with 3
constellations with SBAS.

I wonder why you are only getting GPS from the local station, but
presumabbly it's old and not sending Galileo and GLONASS corrections.

> RTCM3 can perform better.

If by that, you mean "receiver does RTK based on carrier phase data
carried in RTCM3", then agreed.

Repeating something I said earlier my local RTCM3 provider is now
sending data on all 4 constellations.

>> An interesting test - I think- would be to compare GPS + EGNOS with
>> GPS 
>> + DGPS.
>
> EGNOS is also basically RTCM2, so the limitations of the protocol
> are similar.  One would think local corrections are better than area
> corrections, but the real limitation is the receiver itself.

There are two things going on at once: the error in pseudorange
measurement, which is the same, and the errors due to pseudorange
correction information or the standard model vs the actual delays.  As I
understand it SBAS has a fairly coarse grid so local DGPS if quality
should be pretty good.  But it may be also that with 3 constellations
and the resulting high number of satellites, the SBAS approach is
already good enough this doesn't matter.

>> (**) Something strange:
>> When not using dgps (so all three 3 + sbas enabled), xgps did say "3D 
>> FIX DGPS", but all three egnos satellites where marked as "used: no".
>
> GPS sats orbit a lot lower than SBAS ones.  The accuracy of their
> signals is also lower.  Thus they are only used in a fix when there
> are not enough GPS in sight.  Some modern (u-blox 9) dont even bother
> listening to SBAS at all.

It would be nice if receivers were clear about "using corrections from
this SBAS" vs "using this SBAS's signal to observe pseudoranges and
including those in the position estimate".  It's very natural for
someone to expect the using: field of an SBAS satellite to mean that
corrections are used.

>> So what 'DGPS' does xgps refer to when all the egnos/sbas satellites  
>> are marked as 'not in use'?
>
> Every chip/firmware is different.  Broadly DGPS means some corrections
> have been used.  The RTCM2 from a SBAS may be used to improve the fix
> even when ranging from the SBAS is not.

Operationally there is a mode in the data from the receiver which has a
small number of values and gpsd makes it available.

See gpsd_json(5).

signature.asc
Description: PGP signature