Navigation: Main Page » Computer / Recreational / Science / Society / Television
 
Web N-N-A.com
         
Science Groups Forum Index  »  Geology - Satellite Navigation  »  SVN49 Tip of the Iceberg?
Page 1 of 1    
Author Message
Tim Springer
Posted: Sun Jul 19, 2009 10:54 pm
Guest
After looking into the details of the SVN49 anomaly we have now discovered
that more GPS Block IIR and IIR-M satellites may be suffering from the same
root cause. Have a look at the latest information on my blog or on
InsideGNSS

My blog: http://gnss.servolux.nl/gnss_blog.html

InsideGNSS: http://www.insidegnss.com/node/1599
HIPAR
Posted: Tue Jul 21, 2009 2:39 pm
Guest
On Jul 19, 2:54 pm, "Tim Springer" <Tim-Springer-nos...@t-online.de>
wrote:
Quote:
After looking into the details of the SVN49 anomaly we have now discovered
that more GPS Block IIR and IIR-M satellites may be suffering from the same
root cause. Have a look at the latest information on my blog or on
InsideGNSS

My blog:http://gnss.servolux.nl/gnss_blog.html

InsideGNSS:http://www.insidegnss.com/node/1599

Now that we have discovered NAVSTAR is orbiting space junk, can
someone direct me to an xTex class hand held GLONASS receiver ?

--- CHAS
GPSGUY
Posted: Tue Jul 21, 2009 6:11 pm
Guest
Could be and probably worthy of some investigation, but no one will
argue about the performance of the current constellation. But they
will debate whether SVN49 should be included in the current
constellation.

Quote:
After looking into the details of the SVN49 anomaly we have now discovered
that more GPS Block IIR and IIR-M satellites may be suffering from the same
root cause. Have a look at the latest information on my blog or on
InsideGNSS

My blog:http://gnss.servolux.nl/gnss_blog.html

InsideGNSS:http://www.insidegnss.com/node/1599
HIPAR
Posted: Tue Jul 21, 2009 6:38 pm
Guest
On Jul 21, 2:11 pm, GPSGUY <sat_ali...@yahoo.com> wrote:
Quote:
Could be and probably worthy of some investigation, but no one will
argue about the performance of the current constellation. But they
will debate whether SVN49 should be included in the current
constellation.


Certainly Tim's investigation should put the Air Force on notice that
someone is independently checking their work. Hopefully, the GPS IIF
and GPS III satellites will be scrutinized for these kinds of issues.


--- CHAS
GPSGUY
Posted: Wed Jul 22, 2009 1:18 am
Guest
Between this and the GAO report, the heat in the kitchen has certainly
risen at the GPS Wing.

Quote:
Could be and probably worthy of some investigation, but no one will
argue about the performance of the current constellation. But they
will debate whether SVN49 should be included in the current
constellation.

Certainly Tim's investigation should put the Air Force on notice that
someone is independently checking their work.  Hopefully, the GPS IIF
and GPS III satellites will be scrutinized for these kinds of issues.

---  CHAS
Mike Jr
Posted: Wed Jul 22, 2009 2:24 am
Guest
On Jul 21, 9:15 pm, "Marty Ryba" <martin.ryba.nos...@verizon.net>
wrote:
Quote:
"HIPAR" <captc...@verizon.net> wrote in message

news:d922fa82-cf37-4b84-a6ac-06246690c9a3@h18g2000yqj.googlegroups.com...

Certainly Tim's investigation should put the Air Force on notice that
someone is independently checking their work.  Hopefully, the GPS IIF
and GPS III satellites will be scrutinized for these kinds of issues.

[snip]
On the other hand, since he sees the effect only on certain PRNs, I'm fairly
sure he's drawing the correct conclusions.

Marty,

Lots of people inside and outside the wing are constantly analyzing
the data and they didn't spot it till they began to look specifically
for an elevation dependent effect. The effects are tiny but they are
there.
Mike Jr
Posted: Wed Jul 22, 2009 2:28 am
Guest
On Jul 21, 10:39 am, HIPAR <captc...@verizon.net> wrote:
Quote:
On Jul 19, 2:54 pm, "Tim Springer" <Tim-Springer-nos...@t-online.de
wrote:

After looking into the details of the SVN49 anomaly we have now discovered
that more GPS Block IIR and IIR-M satellites may be suffering from the same
root cause. Have a look at the latest information on my blog or on
InsideGNSS

My blog:http://gnss.servolux.nl/gnss_blog.html

InsideGNSS:http://www.insidegnss.com/node/1599

Now that we have discovered NAVSTAR is orbiting space junk, can
someone direct me to an xTex class hand held GLONASS receiver ?

---  CHAS

[grin]

The elevation dependent effects reported by Inside GNSS for other IIR
SVs are very, very tiny. So not space junk yet. Even "Sputnik" is
serving a useful purpose in securing the L5 bandwidth. Just don't
turn that puppy loose on the poor GPS system users.
Marty Ryba
Posted: Wed Jul 22, 2009 5:15 am
Guest
"HIPAR" <captchas@verizon.net> wrote in message
news:d922fa82-cf37-4b84-a6ac-06246690c9a3@h18g2000yqj.googlegroups.com...
Quote:
Certainly Tim's investigation should put the Air Force on notice that
someone is independently checking their work. Hopefully, the GPS IIF
and GPS III satellites will be scrutinized for these kinds of issues.

Agreed, although I remember the big ballyhoo several years ago about "Evil
Waveforms" and "More Evil Waveforms". The Stanford group wrote a whole pile
of papers on these failures that best I understand turn out to not be
realistic failure modes of the IIA/IIR/IIRM modulators and filters. There
are a couple requirements now in the draft SS-SS-800 specification that
governs the development of the GPS IIIA birds that specifically address
testing to ensure that they can't happen over certain time scales. Remember,
every added requirement (and there are hundreds in SS-SS-800) costs
significant money in analysis, tracking, validation and verification,
testing, etc.

I could forsee a (reasonable) requirement on the maximum amplitude of echo
components of the signal as a function of time delay, and some ways we could
test to exclude them, although it's harder than it sounds. For instance, an
echo at 30 nanoseconds delay translates to a bandwidth of 33 MHz to detect
it reliably (unless you use some fancy "superresolution" processing). That's
about the bandwidth of the L1 filter (it's in the neighborhood of 30 MHz to
accommodate M-Code). The echos that Tim is now looking for in the other
birds are on the order of 10 nanoseconds, which is 100 MHz. High bandwidth
correlators that impose a certain signal model trade SNR for resolution
(that's the key to superresolution processing...if you have the SNR you can
extrapolate bandwidth if you have a reliable model of the modulation). On
the other hand, *any* bandpass filter will have an impulse response with
time domain ripples in it. I'm sure Tim's receivers do, so some of his
errors could even be from the receiver or antenna LNA itself, although those
should not be a function of nadir angle but more fixed. They could be SNR
dependent though, which *does* correlate with nadir/elevation angle. Hmmmm.
On the other hand, since he sees the effect only on certain PRNs, I'm fairly
sure he's drawing the correct conclusions.

-Marty
s_anode@comcast.net
Posted: Wed Jul 22, 2009 4:06 pm
Guest
On Jul 20, 11:30 am, "s_an...@comcast.net" <s_an...@comcast.net>
wrote:
Quote:
On Jul 19, 2:54 pm, "Tim Springer" <Tim-Springer-nos...@t-online.de
wrote:

After looking into the details of the SVN49 anomaly we have now discovered
that more GPS Block IIR and IIR-M satellites may be suffering from the same
root cause. Have a look at the latest information on my blog or on
InsideGNSS

My blog:http://gnss.servolux.nl/gnss_blog.html

InsideGNSS:http://www.insidegnss.com/node/1599

SVN49 problem was caused by impedance miss match with the L5
demonstration payload.  It is a unique configuration.  The elevation
dependent pseudorange error is not seen on other satellites.

                 Tom

I don't disagree that each SV has its own variations, as does each
receiving antenna/filter/receiver set, my point is SVN-49 has a
problem due to the installation of the L5 demonstration payload. The
other SVs do not have the same problem. SV transmit amplifier and
filter characteristics, the characteristics of the satellite antenna
array, the PRN codes themselves, the type of correlator used, the RF
section of the receiver, the antenna, the radome, and LNA all
introduce small (usually) biases. Antennas, radomes, and AGC
introduce a lot of elevation dependence. Heck, there is not even a
universally accepted definition of the exact point of the pseudorange
measurement (i.e. which is right? a full chip correlator or a narrow
correlator?) Whether or not the carrier windup correction is applied
will also make a elevation like dependent difference in high accuracy
applications.


Tom
Mike Jr
Posted: Thu Jul 23, 2009 3:09 am
Guest
On Jul 21, 9:18 pm, GPSGUY <sat_ali...@yahoo.com> wrote:
Quote:
Between this and the GAO report, the heat in the kitchen has certainly
risen at the GPS Wing.
[snip]


History has taught us that the best way to handle a mistake is:
1. Admit it
2. Fix it
3. Don't repeat it

You can't fix a problem that you don't admit exists.

The wing seems to treat engineering as a commodity, like electricity.
"GPS is settled science/engineering; I can plug any component, any
product, and any company's engineering talent in and it will work."
Nothing could be a bigger mistake. Until the wing admits and fixes
this mistake they will continue to be plagued by these kinds of
problems.
Tim Springer
Posted: Mon Jul 27, 2009 4:05 am
Guest
"Mike Jr" <n00spam@comcast.net> schrieb im Newsbeitrag
news:31434325-6c65-4014-89da-785ec56d8079@q35g2000vbi.googlegroups.com...
Quote:
On Jul 21, 9:18 pm, GPSGUY <sat_ali...@yahoo.com> wrote:
Between this and the GAO report, the heat in the kitchen has certainly
risen at the GPS Wing.
[snip]

History has taught us that the best way to handle a mistake is:
1. Admit it
2. Fix it
3. Don't repeat it

You can't fix a problem that you don't admit exists.

The wing seems to treat engineering as a commodity, like electricity.
"GPS is settled science/engineering; I can plug any component, any
product, and any company's engineering talent in and it will work."
Nothing could be a bigger mistake. Until the wing admits and fixes
this mistake they will continue to be plagued by these kinds of
problems.

Personally I find the behavior of the GPS wing with respect to the SVN49
issue quite professional. They have admitted to the problem very quickly and
even managed to find the cause and explain it to the public. This is much
better and much more "open" communications then I experience with the
European counterpart of GPS named Galileo. Despite the fact that Galileo is
NOT a military system everything is treated as "top secret" and no one is
allowed to publish anything without prior approval of the EU and/or ESA.
Really a very surprising and very disappointing and frustrating situation
for a "public" system build with "public" money.

So from me a big compliment to the GPS wing! I really love the GPS system
and the performance of it is truly great. Of course the SVN49 is a bit of a
"hick-up" but at least it has secured the L5 frequency. Europe had to launch
two test satellites (Giove-A and Giove-B) to secure its frequencies and the
usefulness of those two is really limited.

For the users of the GPS system SVN49 really does not pose much of a
problem. A simple polynomial model can remove the issue and make the
satellite fully usable. Whether all cheap GPS chips as used in mobile phones
and such will implement such a correction is an open question. But at least
all "post processing" software can easily correct the issue.
HIPAR
Posted: Mon Jul 27, 2009 2:22 pm
Guest
'For the users of the GPS system SVN49 really does not pose much of a
problem. A simple polynomial model can remove the issue and make the
satellite fully usable. Whether all cheap GPS chips as used in mobile
phones
and such will implement such a correction is an open question. But at
least
all "post processing" software can easily correct the issue'.

I'd guess adding SVN49 to the constellation as is would cause no
noticeable problems to everyday Standard Positioning Service users.
What would a comparative analysis based upon historical errors for
each satellite show when SVN49 is added 'as is' versus adding SVN49
exhibiting the proper 'in family' error (i.e. 0.6 meter) for a Block
IIRM satellite? Let's say the results would be computed at one minute
intervals over the period of a two orbit day. Of course, this is not
a simple task but I'm thinking tools already exist to do it.

--- CHAS
Mike Jr
Posted: Tue Jul 28, 2009 9:35 pm
Guest
On Jul 26, 8:05 pm, "Tim Springer" <Tim-Springer-nos...@t-online.de>
wrote:
Quote:
"Mike Jr" <n00s...@comcast.net> schrieb im Newsbeitragnews:31434325-6c65-4014-89da-785ec56d8079@q35g2000vbi.googlegroups.com...



On Jul 21, 9:18 pm, GPSGUY <sat_ali...@yahoo.com> wrote:
Between this and the GAO report, the heat in the kitchen has certainly
risen at the GPS Wing.
[snip]


Quote:

Personally I find the behavior of the GPS wing with respect to the SVN49
issue quite professional. They have admitted to the problem very quickly and
even managed to find the cause and explain it to the public. This is much
better and much more "open" communications then I experience with the
European counterpart of GPS named Galileo. Despite the fact that Galileo is
NOT a military system everything is treated as "top secret" and no one is
allowed to publish anything without prior approval of the EU and/or ESA.
Really a very surprising and very disappointing and frustrating situation
for a "public" system build with "public" money.

Sorry to hear about life under Galileo. Hadley CRU in Great Britain
seems to have a similar problem in making public something as simple
as weather data.

Quote:

So from me a big compliment to the GPS wing! I really love the GPS system
and the performance of it is truly great. Of course the SVN49 is a bit of a
"hick-up" but at least it has secured the L5 frequency. Europe had to launch
two test satellites (Giove-A and Giove-B) to secure its frequencies and the
usefulness of those two is really limited.

For the users of the GPS system SVN49 really does not pose much of a
problem. A simple polynomial model can remove the issue and make the
satellite fully usable. Whether all cheap GPS chips as used in mobile phones
and such will implement such a correction is an open question. But at least
all "post processing" software can easily correct the issue.

It does look like punting the problem to the receiver manufactures is
the likest answer if one is hell bent on making SVN49 healthy to all
users. However I doubt that more than a small minority of existing
receivers can be fixed with a firmware download. Future generations
of receivers could benefit from better algorithms for handling
multipath. Of course, this does nothing for the ten million GPS
receivers already out there. OTOH, it would help sales of new
receivers; sounds like a faustian bargain in the making. I can see a
new government program now. "Turn in that clunker of a GPS receiver
for a new, more multipath efficient model and get a government voucher
worth ten dollars off your federal taxes." I can see the GDP rising
as I write this.

BTW, Jonathan Swift is my favorite author. http://en.wikipedia.org/wiki/A_modest_proposal

Mike Jr
 
Page 1 of 1       All times are GMT
The time now is Fri Jul 30, 2010 10:24 am