Bookmark this page:
If you were Registered and logged in, you could reply and use other advanced thread options
New GPS Satellite's Problems Indicated
http://sidt.gpsworld.com/gpssidt/content/printContentPopup.jsp?id=3D59=
6465
May 4, 2009
Lt Col David Goldstein, chief engineer for the GPS Wing, told the plenary=
session at the=20
European Navigation Conference in Naples, Italy, that the Wing is experie=
ncing some =E2=80=9Cout=20
of family=E2=80=9D measurements from the recently launched IIRM (20) sate=
llite. This appears to=20
corroborate some unofficial rumors that have circulated recently about pr=
oblems with=20
=E2=80=9Clegacy signals=E2=80=9D from the satellite, that is, L1 and L2. =
The April 10 broadcast of the=20
first L5 signal secured that frequency for the U.S. GPS program; since th=
at signal=20
contains no navigation message at present, it is presumably not affected =
by these problems.
Goldstein told the ENC opening session, Monday May 4, that the Air Force =
will not launch=20
any further satellites until this issue is resolved. IIR(M) 21, the last =
of the IIR(M)=20
series, is currently scheduled to rise sometime in August, with the first=
of the IIF=20
generation to follow in late 2009 or early 2010.
Normally, a satellite is set healthy within 28 days of launch, after exte=
nsive testing,=20
but this has not occurred with the satellite launched on March 24. The U.=
S. Air Force has=20
formed a response team and is working =E2=80=9Cnearly round the clock=E2=80=
=9D to resolve the problem, but=20
according to Goldstein is not rushing the issue, seeking a thorough solut=
ion since the=20
overall constellation is robust at 30 satellites.
=E2=80=9CWe are currently examining data from the satellite that is not c=
onsistent with data from=20
the other IIR(M)s,=E2=80=9D he stated, characterizing the variances as =E2=
=80=9Cmeasurements with larger=20
than expected pseudorange errors that are elevation-dependent, and that w=
e have not seen=20
before. We have experimented with a few fixes and it looks very promising=
=2E=E2=80=9D
He described the response team=E2=80=99s approach as making a =E2=80=9Cfi=
shbone diagram=E2=80=9D of all potential=20
failure mechanisms, and working through them methodically. =E2=80=9CWe th=
ink we have identified=20
the failure but it may be several more months before the analysis is comp=
lete, and the=20
situation is fully resolved.=E2=80=9D
Have a look at my BLOG at http://gnss.servolux.nl/gnss_blog.html to see a
detailed look at the problems with this satellite.
It is clear that there is a small problem with the code observations when
used simultaneously with the phase observations. This problem is only at the
few meter level and thus not really important. The cause is most likely an
issue with the so called group delay of the code observations.
What is much more problematic is that the broadcast ephemerides are very
poor. When compared with an orbit determination, like the IGS is doing, the
differences seem to be at the 60 km level!! The reason for this is unclear
at present for "normal" observers like me. So there must be more at hand
then the view meters we see in the code residuals....
I will update my BLOG as soon as I learn more about this problem.
Cheers,
Tim
http://gnss.servolux.nl/gnss_blog.html
> http://sidt.gpsworld.com/gpssidt/content/printContentPopup.jsp?id=596465
> May 4, 2009
> Lt Col David Goldstein, chief engineer for the GPS Wing, told the plenary
> session at the European Navigation Conference in Naples, Italy, that the
> Wing is experiencing some “out of family” measurements from the recently
> launched IIRM (20) satellite. This appears to corroborate some unofficial
> rumors that have circulated recently about problems with “legacy signals”
> from the satellite, that is, L1 and L2. The April 10 broadcast of the
> first L5 signal secured that frequency for the U.S. GPS program; since
> that signal contains no navigation message at present, it is presumably
> not affected by these problems.
> Goldstein told the ENC opening session, Monday May 4, that the Air Force
> will not launch any further satellites until this issue is resolved.
> IIR(M) 21, the last of the IIR(M) series, is currently scheduled to rise
> sometime in August, with the first of the IIF generation to follow in late
> 2009 or early 2010.
> Normally, a satellite is set healthy within 28 days of launch, after
> extensive testing, but this has not occurred with the satellite launched
> on March 24. The U.S. Air Force has formed a response team and is working
> “nearly round the clock” to resolve the problem, but according to
> Goldstein is not rushing the issue, seeking a thorough solution since the
> overall constellation is robust at 30 satellites.
> “We are currently examining data from the satellite that is not consistent
> with data from the other IIR(M)s,” he stated, characterizing the variances
> as “measurements with larger than expected pseudorange errors that are
> elevation-dependent, and that we have not seen before. We have
> experimented with a few fixes and it looks very promising.”
> He described the response team’s approach as making a “fishbone diagram”
> of all potential failure mechanisms, and working through them
> methodically. “We think we have identified the failure but it may be
> several more months before the analysis is complete, and the situation is
> fully resolved.”