Please excuse the spelling errors, I'm using webmail today. Right now it looks like my software does to the demodulation and calculation of position into xyz coordinates, I'm still looking for enough raw data to be able to test the full position calculation.
Major todos: -Integration with GNURadio, I think most of the low level portions of the GPS reciever will have to be one block, because all of the correlators have to be syncronized. There is a question as to what this section should output. The major problem is that GNURadio blocks can only output at one data rate and the correlator outputs are not syncronized. So there will need to be a stream of two types of packets: unsyncronized C/A symbols from each correlator and syncronized snapshots of correlator counters to calculate pseudo ranges. -Integration with GPStk, I think my code is good enough for basic testing, but there is so much effort that has been put into the GPStk that it doesn't make sense to duplicate it. -Threshold calculation to speedup search. -Selection of PLL and DLL loop filters, -High level receiver functionality, including tracking only satellites that are in view, almanac managment (ie. cold start vs. warm start), ... Quoting David Bengtson <[EMAIL PROTECTED]>: > Hi: > > I'm still working on this slowly. I've selected the Maxim 2745 as the > best part available that is reasonably accessible. I've ordered and > received samples of the part from Maxim, and am starting to work on a > block diagram, followed by a schematic. This is looking like a RX module > to plug into the USRP board, and feed data into the FPGA for down > sampling and filtering. > > Question for folks on the mailing list. I'm a hardware engineer, and am > more adept in programming in Matlab and solder than Python. Are there > enough folks out there that would be interested in helping with the > software to make this feasible, as I'm certainly not the guy to write > the code. I do feel competent to do the block diagram for the software, > although Krys is probably more up to speed on that. My guess is that we > would need ~ 2 to 3 folks writing code to demodulate the GPS signal and > turn it into a location. We can talk about generating hardware, as it > looks like the GPS board by itself will be pretty cheap, although costs > will have to wait until I have BOM > > > Dave > > > > Krzysztof Kamieniecki wrote: > > > I love replying to myself. > > > > I updated my website with a plot showing the results of a correlator > > doing a brute force search thru frequency and delay on the real raw data. > > > > Krzysztof Kamieniecki wrote: > > > >> I am, slowly but surely. I plan on using the USRP. > >> I recently got some raw data from a person associated with GPStk if > >> you are interested. > >> > >> Right now David Bengtson is working on a daughter card. > >> http://www.keystoneradio.com/GnuGPS.html > >> > >> Other links of interest > >> My page: http://www.kamieniecki.com/krys/gps/ > >> GnuRadio GPS page: http://comsec.com/wiki?GlobalPositioningSystem > >> > >> Robert McGwier wrote: > >> > >>> Who is working on a GPS receiver for the USRP or other hardware > >>> in this project? I would like not to do duplication of effort. > >>> > >>> Bob > >>> > >>> > >>> > >>> > >>> _______________________________________________ > >>> Discuss-gnuradio mailing list > >>> [email protected] > >>> http://lists.gnu.org/mailman/listinfo/discuss-gnuradio > >>> > >> > > > > > > _______________________________________________ > Discuss-gnuradio mailing list > [email protected] > http://lists.gnu.org/mailman/listinfo/discuss-gnuradio > _______________________________________________ Discuss-gnuradio mailing list [email protected] http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
