Tom, Let's do a separate thread on that one. Cheers, Magnus On 04/18/2013 08:07 PM, Tom Knox wrote: > > > I hope I am not steering the thread to much, I have an Allen Osborne TurboRogue SMR12 RM L1 L2 GPS rec that came from JPL that appears specifically designed for Common View use. Does anyone know the history of these ,I can find almost nothing. > Email me directly if you have info, or I can start a new thread. > Tom Knox actast@hotmail.com > >> Date: Thu, 18 Apr 2013 19:58:22 +0200 >> From: magnus@rubidium.dyndns.org >> To: time-nuts@febo.com >> Subject: Re: [time-nuts] antennas was Re: Common-View GPS Network >> >> On 04/18/2013 04:00 PM, Jim Lux wrote: >>> On 4/18/13 12:01 AM, Chris Albertson wrote: >>>> If I read the paper correctly you can skip the choke ring if you mount >>>> the >>>> antenna on top of a 2 meter or longer mast. Iron pipe comes on 10 foot >>>> lengths. The choke ring is for portable survey antenna that can't be >>>> placed on tall rooftop masts. I think a 2 meter pole on a roof pretty >>>> much meets their criteria of multi path difference being over 10 meters. >>> >>> >>> >>> yes.. multipath that is more than a chip away is generally filtered out >>> by the PN tracking loop, so all you really worry about is multipath >>> signals within a chip. For C/A code at 1 Mchip/sec, the chips are 300 >>> meters long. If you're doing P/Y code, it's a tenth of that. >>> >>> In reality, if the multipath signal is lower, and it's "far" away (a >>> good fraction of a chip) it doesn't contribute much to the output of the >>> correlator. So their 10 meter thing is probably a good number for a >>> "typical" receiver they make. >> >> It's an interesting mix of sample-rate/bandwidth, code you track and >> distance between early-late detectors (normal distance is one chip) >> comes in when analyzing and combat the multi-path. I recall there is >> some subtle points with some of the C/A codes. >> >> Cheers, >> Magnus >> _______________________________________________ >> 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.

Magnus, Jim, > On 04/18/2013 04:00 PM, Jim Lux wrote: >> On 4/18/13 12:01 AM, Chris Albertson wrote: >>> If I read the paper correctly you can skip the choke ring if you mount >>> the >>> antenna on top of a 2 meter or longer mast. Iron pipe comes on 10 foot >>> lengths. The choke ring is for portable survey antenna that can't be >>> placed on tall rooftop masts. I think a 2 meter pole on a roof pretty >>> much meets their criteria of multi path difference being over 10 >>> meters. >> >> >> >> yes.. multipath that is more than a chip away is generally filtered out >> by the PN tracking loop, so all you really worry about is multipath >> signals within a chip. For C/A code at 1 Mchip/sec, the chips are 300 >> meters long. If you're doing P/Y code, it's a tenth of that. >> >> In reality, if the multipath signal is lower, and it's "far" away (a >> good fraction of a chip) it doesn't contribute much to the output of the >> correlator. So their 10 meter thing is probably a good number for a >> "typical" receiver they make. > > It's an interesting mix of sample-rate/bandwidth, code you track and > distance between early-late detectors (normal distance is one chip) > comes in when analyzing and combat the multi-path. I recall there is > some subtle points with some of the C/A codes. All of the "high quality" GNSS receiver manufacturers have their own version of correlator that try to mitigate multipath. See for example this Ashtech-document (for a ca 10 year old L1 only receiver (DG14/16)). ftp://ftp.ashtech.com/OEM,%20Sensor%20&%20ADU/DG16%20&%20DG14/Reference%20Material/Correlator.doc /Björn

On 4/18/13 1:40 PM, bg@lysator.liu.se wrote: > All of the "high quality" GNSS receiver manufacturers have their own > version of correlator that try to mitigate multipath. See for example > this Ashtech-document (for a ca 10 year old L1 only receiver (DG14/16)). > > ftp://ftp.ashtech.com/OEM,%20Sensor%20&%20ADU/DG16%20&%20DG14/Reference%20Material/Correlator.doc > I'm sure.. I wonder, though, if you have a GPS receiver for which you do not know how it works internally, can one come up with a guideline for how much multipath suppression you want. At some level, all those fancy algorithms are trying to build an adaptive equalizer and/or filter for the multipath, helped by the knowledge that the "true" path is the shortest one. In theory, one should be able to deconvolve arbitrary multipath (if you collect all signals from all directions of the sky, etc.), but I think the idea of the chokering (and other clever antenna designs) is to reduce the work for the back end processing.