Slightly OT - GPS-Based Accurate Direction Finding
On Thu, 26 Aug 2010 06:58:01 -0700
jimlux jimlux@earthlink.net wrote:
If you get a bit "closer to the metal" you could use two GPS L1 samplers
running off a common clock, and do the PN code acq and track, which
would give you carrier phase. If you do the nav solution, you know the
"look angle" to the various SVs, which would tell you the phase
differential vs azimuth.
I believe that there are open source codes out there to do the
processing. The data rate isn't all that high.. the GPS samplers are 1
bit. There's certainly lots of papers from grad students on this kind
of thing.
Yes, but for this you have either to build a GPS module yourself,
which is not trivial (neither is it to get GPS chips in single
quantities) or you have to get one of those GPS experimentation
boards which are not really cheap.
Not to mention that you are then running your own GPS code
which is probably more than the OP was asking for.
IMHO it's simpler to use available precision modules that provide
phase data and use that for the caluculation. If additional accuracy
is required, both modules can be modified to run from the same OCXO
instead of the TXCO they usually use.
Attila Kinali
--
If you want to walk fast, walk alone.
If you want to walk far, walk together.
-- African proverb
Hi
Most fire towers have a pretty good view of the surroundings. How about
simply picking an adequately remote point and using that (based on map data)
as your direction reference? If you get lucky the fire tower is actually a
survey tower that got converted to fire use and there's a very well known
location for the tower.
Bob
-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Attila Kinali
Sent: Thursday, August 26, 2010 11:20 AM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Slightly OT - GPS-Based Accurate Direction Finding
On Fri, 27 Aug 2010 00:09:31 +1000
David Smith david@smithfamily.net.au wrote:
Some background - I'm needing an accuracy of 1 degree or better. The
experiments are using digital communication modes and sometimes aircraft
scatter so signals are regularly inaudible and often non-existent, so
peaking "by ear" is not usually an option.
1 Degree isnt that hard. Even a standard magnetic compas reaches that.
I've quite an old one at home with prismas for aiming with which it
is possible to get better readings than 0.5°. Of course, you'd have
to take the deviations from true north into account. But magnetic maps
should be readily available (at least they were when i was a scout).
As Atilla says, the software is probably not that fundamentally
complicated. However, the devil is possibly in the detail of aligning
sample timing, positioning ...
You don't have to allign sample timing. It is helpfull though, to get
better precision/accuracy. It is enough that you use the relative
phase differences of the SV on each GPS module. This way you can get
rid of the unknown sample phase difference of the GPS modules.
(the principle is the same like normal GPS measurement using a 4th
satelite to get rid of the unknown "time").
I just had a look at the LEA6-T protocol specs. The RXM-RAW message
provides you with a phase difference (in L1 cycles) and a frequency
offset (doppler, in Hz) and pseudorange (in meter) for each SV that
is being tracked. With this you should be able to eliminate the
phase and frequency difference/drift between the two modules.
Using a long enough measurement time, you should be able to get
to well below a degree of heading accuracy. I know that the ETH
in Zürich uses LEA6-T modules to get positioning resolutions
better than 2mm with single GPS modules, though they have to collect
more than an hour of phase data for this.
Attila Kinali
--
If you want to walk fast, walk alone.
If you want to walk far, walk together.
-- African proverb
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
On 08/26/2010 05:24 PM, Attila Kinali wrote:
On Thu, 26 Aug 2010 06:58:01 -0700
jimluxjimlux@earthlink.net wrote:
If you get a bit "closer to the metal" you could use two GPS L1 samplers
running off a common clock, and do the PN code acq and track, which
would give you carrier phase. If you do the nav solution, you know the
"look angle" to the various SVs, which would tell you the phase
differential vs azimuth.
I believe that there are open source codes out there to do the
processing. The data rate isn't all that high.. the GPS samplers are 1
bit. There's certainly lots of papers from grad students on this kind
of thing.
Yes, but for this you have either to build a GPS module yourself,
which is not trivial (neither is it to get GPS chips in single
quantities) or you have to get one of those GPS experimentation
boards which are not really cheap.
GNSS samplers is off the shelf stuff for reasonable money.
Not to mention that you are then running your own GPS code
which is probably more than the OP was asking for.
Part of the fun. Not too hard really... got pretty far on my code already.
IMHO it's simpler to use available precision modules that provide
phase data and use that for the caluculation. If additional accuracy
is required, both modules can be modified to run from the same OCXO
instead of the TXCO they usually use.
Just using the same TCXO is a step forward. Off the shelf boards have
been hacked for this very purpose before.
Systems for heading, tilt etc. is commercially available for both
airplanes and boats under various names. The main system usually have
one main antenna for positioning and additional antennas for orientation.
The critical aspect is carrier phase detection, if one has carrier phase
reports, the rest can be done in post-processing.
Cheers,
Magnus
Does anyone know how laser gyroscopes are developing?
Laser gyroscopes - as in Ring Laser Gyroscopes or as in Fiber Optic
Gyroscopes?
RLGs are a standard commercial product. Several years back I was
walking through the Honeywell plant in St Paul, MN, and they had a
display case of at least a dozen RLGs that they've made over the past
few decades.
Commercial?
US RLGs are all ITAR.
"All types of gyros usable in the systems in Item 1, with a rated drift
rate stability of less than 0.5 degree (1 sigma or rms) per hour"
http://www.fas.org/spp/starwars/offdocs/itar/p121.htm
Honeywell has about 2 different RLGs. Only one (gg1320) of which you can
make a north sensing out of. Litton (now NGC) used to do RLGs (their "zero
lock gyros") but I think they were on the loosing side of a patent war
with Honeywell.
French Sagem do some for high end military systems. Have I missed a RLG
manufacturer? Almost as few vendors as in the Cesium oscillator market...
No new RLG sensors has been announced during the last decade or two.
--
Björn
I would suggest you buy an old surveying transit.
This is what largely they were intended for.
With the industry going electronic, really nice ones can be found
fairly inexpensively (considering the quality of the optics on the
good ones).
I picked up a nice K+E with the tripod for $300.
Tom Frank
On Aug 26, 2010, at 10:09 AM, David Smith wrote:
Thanks for all the interesting responses.
Some background - I'm needing an accuracy of 1 degree or better.
The experiments are using digital communication modes and sometimes
aircraft scatter so signals are regularly inaudible and often non-
existent, so peaking "by ear" is not usually an option.
I've paced out direction using a handheld GPS (GPSMap 60CSX) and
this gives reasonable results if there's a reasonable baseline.
It's a bit impractical when operating from a firetower though!
Using Sun/Moon/Stars is difficult when there's cloud. We've tried
using Sun RF Noise, but accuracy declines significantly when the
sun is high in the sky.
VOR is an interesting suggestion, but a very sharp (and large)
antenna would be needed and multi-pathing may cause problems.
So, my interest turns back to a GPS-based solution and the military
units suggested by Brooke look perfect ... except that they are
most likely a restricted export and unavailable to us Down Under.
Other links on Brooke's site have lead me to many papers
researching GPS-based attitude systems. I note that the Uni of
Calgary have developed a package called HEADRT+ that can take raw
measurements from several GPS mounted on a small baseline and
produce attitude information. This is the sort of thing I'm after,
but I get the impression that licensing costs are high.
As Atilla says, the software is probably not that fundamentally
complicated. However, the devil is possibly in the detail of
aligning sample timing, positioning ...
Any other suggestions?
Regards,
Dave
Thomas A. Frank wrote:
I would suggest you buy an old surveying transit.
This is what largely they were intended for.
ANd with a transit, you can easily see Polaris in the day time.
Here's an article discussing how to do it
http://www.cadastral.com/cad-polr.htm
skip down to the section on Observation Procedure
It's a bit trickier than it says...
-- you won't be able to do this the first time
small field of view on the telescope means you have to be pointed
pretty close to find it. What I wound up doing is marking the place
where I'd set the theodlite up, going out at night and finding Polaris,
and finding a suitable object on the horizon to reference to. Then the
next day, I set up the tripod and used that object to get the azimuth
approximate, tipped up to the right elevation, and, et voila, there it
was in the scope. The comment about focus is right on though.
Back in the early 1980's when attending college I worked on a single
axis multi-mode fiber optic
rate gyro project that used GRIN fiber. Back then a military three axis
unit based on single mode
fiber was alleged to be a little larger than a one inch cube and cost
slightly less than a million dollars.
We used a three inch spool for the fiber and put everything in a six
inch cube for a housing.
The NASA contract was part of the NASP program.
The company that we worked with wanted to produce a product for the
commercial "private
pilot" aviation market. I will have to ask what happened...
I think the patent issue may have had something to do with it since the
company had a
relationship with Litton.
Jim Cotton
n8qoh
bg@lysator.liu.se wrote:
Does anyone know how laser gyroscopes are developing?
Laser gyroscopes - as in Ring Laser Gyroscopes or as in Fiber Optic
Gyroscopes?
RLGs are a standard commercial product. Several years back I was
walking through the Honeywell plant in St Paul, MN, and they had a
display case of at least a dozen RLGs that they've made over the past
few decades.
Commercial?
US RLGs are all ITAR.
"All types of gyros usable in the systems in Item 1, with a rated drift
rate stability of less than 0.5 degree (1 sigma or rms) per hour"
http://www.fas.org/spp/starwars/offdocs/itar/p121.htm
Honeywell has about 2 different RLGs. Only one (gg1320) of which you can
make a north sensing out of. Litton (now NGC) used to do RLGs (their "zero
lock gyros") but I think they were on the loosing side of a patent war
with Honeywell.
French Sagem do some for high end military systems. Have I missed a RLG
manufacturer? Almost as few vendors as in the Cesium oscillator market...
No new RLG sensors has been announced during the last decade or two.
--
Björn
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Hi:
An interesting fact about gyros, independent of how they work, is that
their Long Term Drift Rate is proportional to some power of the inverse
of their volume. As the sensor gets smaller the LTDR goes up as 10 ^ 4
http://www.prc68.com/I/Sensors.shtml#Gyroscopic
Ring Laser Gyros at 7E-4 deg/hour to MEMS at 1000 deg/hr.
I find the AG8 North Finding Gyro theodolite system very interesting,
but have heard that's it's very easy to break the gyro suspension and so
buying one used is very risky if you want to actually use it. It uses
a classical spinning mass gyro to find True North. I expect it's still
being used in places where GPS is not available, like in caves.
An old web page on North Finding:
http://www.prc68.com/I/North.shtml
Have Fun,
Brooke Clarke
http://www.PRC68.com
Jim Cotton wrote:
Back in the early 1980's when attending college I worked on a single
axis multi-mode fiber optic
rate gyro project that used GRIN fiber. Back then a military three
axis unit based on single mode
fiber was alleged to be a little larger than a one inch cube and cost
slightly less than a million dollars.
We used a three inch spool for the fiber and put everything in a six
inch cube for a housing.
The NASA contract was part of the NASP program.
The company that we worked with wanted to produce a product for the
commercial "private
pilot" aviation market. I will have to ask what happened...
I think the patent issue may have had something to do with it since
the company had a
relationship with Litton.
Jim Cotton
n8qoh
bg@lysator.liu.se wrote:
Does anyone know how laser gyroscopes are developing?
Laser gyroscopes - as in Ring Laser Gyroscopes or as in Fiber Optic
Gyroscopes?
RLGs are a standard commercial product. Several years back I was
walking through the Honeywell plant in St Paul, MN, and they had a
display case of at least a dozen RLGs that they've made over the past
few decades.
Commercial?
US RLGs are all ITAR.
"All types of gyros usable in the systems in Item 1, with a rated drift
rate stability of less than 0.5 degree (1 sigma or rms) per hour"
http://www.fas.org/spp/starwars/offdocs/itar/p121.htm
Honeywell has about 2 different RLGs. Only one (gg1320) of which you can
make a north sensing out of. Litton (now NGC) used to do RLGs (their
"zero
lock gyros") but I think they were on the loosing side of a patent war
with Honeywell.
French Sagem do some for high end military systems. Have I missed a RLG
manufacturer? Almost as few vendors as in the Cesium oscillator
market...
No new RLG sensors has been announced during the last decade or two.
--
Björn
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
--
Have Fun,
Brooke Clarke
http://www.PRC68.com