6 ms·
Seems fairly useless to me, for a few dollars I can sample the L1 frequency with a dumb device which has no idea of speed or altitude, and do the calcs with FOS
by awesomeusername 2mo ago
Seems fairly useless to me, for a few dollars I can sample the L1 frequency with a dumb device which has no idea of speed or altitude, and do the calcs with FOSS which is in the wild.
Basically a rule which inconveniences the honest and has zero impact on the bad dudes, whoever they are
- inigyou 2mo agoHave you done that?
- londons_explore 2mo agoI have. It works. Used a huge amount of compute though and takes a long time to get a fix.
- inigyou 2mo agoMaybe that's why nobody else does
- rescbr 2mo agoYou can use a FPGA, like this 2013 project: http://www.aholme.co.uk/GPS/Main.htm http://www.aholme.co.uk/GPS/Main.htm Or go old school (1991-1992): https://lea.hamradio.si/~s53mv/navsats/theory.html https://lea.hamradio.si/~s53mv/navsats/theory.html So yeah, this regulation is absolutely an inconvenient measure.
- minetest2048 2mo agoThere's a strong indication that SpaceX does use software receiver in Falcon 9 and Starlink: when they didn't encrypt the downlink telemetry someone captured the signal and found some plain text: https://www.rtl-sdr.com/receiving-space-x-falcon-9-telemetry-with-a-hackrf-and-1-2m-satellite-dish/ https://www.rtl-sdr.com/receiving-space-x-falcon-9-telemetry... That plain text looks like what a software GNSS receiver outputs as its very verbose, and this paper: https://radionavlab.ae.utexas.edu/wp-content/uploads/2022/10/Nichols_ION_GNSS_2022.pdf https://radionavlab.ae.utexas.edu/wp-content/uploads/2022/10... , a paper about a software receiver mentioned this: > PpRx has been licensed through the Radionavigation Lab to multiple commercial companies, but notably a major aerospace company that uses the technology across their suite of advanced spacecraft and satellites. The SDR is deployed across the company’s mega-constellation of satellites used for broadband Internet
- minetest2048 2mo agoFor L1 I think a raspberry pi can have a reasonably fast time to first fix, for offline processing you can go faster than real time. L2/L5 need large sampling rate and a pi is probably not fast enough.. unless if you ditch float32 processing and do 2 bit signal processing, a uni commercialized that: https://radionavlab.ae.utexas.edu/wp-content/uploads/2022/10/Nichols_ION_GNSS_2022.pdf https://radionavlab.ae.utexas.edu/wp-content/uploads/2022/10...
- londons_explore 2mo agoNearly all commercial GPS receivers use a 1 bit ADC. Ie. Just a comparator. The sample rate isn't high either - 16Mhz IIRC.