If I'm hearing you correctly you're trying to do post-process DGPS using consumer grade GPS devices?
If so, this will not work because there is not sufficient data output in the NMEA string. Specifically, you need to have the raw data (pseudo ranges) for each satellite, because the correction for each satellite will be different. Some of the recreational devices may have the ability to turn this stream on but I would not rely upon it for any serious purpose unless I had rigorously verified the accuracy and repeatability. /r - bri On Sun, Feb 13, 2011 at 1:01 AM, Stefan Keller <[email protected]> wrote: > Sorry for asking again. > I'm just wondering about a fundamental configuration decision. > My configuration idea is to have two low cost GPS receivers, where one > of them is on a known position. Both issue just GPS position and log > time data > Is it enough to process just position and log time data (NMEA) from > base station as input to a tool like RTKlib (online or > postprocessing)? > Or is DGPS really only possible with GPS receivers (at base station) > which output ephemerides etc.? > > Yours, Stefan > > P.S. To Dan: Which inertia motion sensor and software did you (and and > typical Autopilots) use to do this sort of dead reckoning? > > 2011/2/4 Dirk-Willem van Gulik <[email protected]>: >> >> On 4 Feb 2011, at 15:52, Dan Lyke wrote: >> >>> On Fri, 4 Feb 2011 16:17:09 +0100 Stefan Keller <[email protected]> wrote: >> >>>> Now given that mobile phones can be connected to low-cost GPS (e.g. >>>> via bluetooth) it should be possible to implement at a "consumer DGPS" >>>> which gets an accurracy of 2m or better. >>> .. >>> Some of those low-cost consumer grade devices already integrate SBAS >>> (Space-Based Augmentations Services) correction like WAAS, EGNOS and >>> MSAS, which can put them close to 2m. These systems work by having a >>> ... >>> However, a good portion of the additional error you're seeing with >>> cheap GPS receivers with tiny little antennas is probably multi-path >>> error, and DGPS won't help you with that. >> >> I got stuck on exactly that as well. And when I improved that - things where >> far to large and heavy. >> >> Subsequently found that using a simple inertia motion sensor (this was for a >> http://paparazzi.enac.fr/wiki/Autopilots board) to do relative stuff at < >> 10cm combined with some very robust points from a GPS - though much less >> accurate - was sort of a good compromise for periods of time under a few >> hours. >> >> Thanks, >> >> Dw. >> _______________________________________________ >> Geowanking mailing list >> [email protected] >> http://geowanking.org/mailman/listinfo/geowanking_geowanking.org >> > > _______________________________________________ > Geowanking mailing list > [email protected] > http://geowanking.org/mailman/listinfo/geowanking_geowanking.org > -- Brian Russo / (808) 271 4166 _______________________________________________ Geowanking mailing list [email protected] http://geowanking.org/mailman/listinfo/geowanking_geowanking.org
