Added an active GPS antenna to the RTL 4000 tuner/spectrum analyzer

I once wrote: Added an active GPS antenna to the RTL 4000 tuner/spectrum analyzer. Got a little active GPS antenna from ebay (item 150748428125) only 6 $, and made a little 'T' section to power it, while connected to the 4000 DVB-T tuner:

formatting link

And then today: First read this blog:

formatting link
I am using the mathlab files in Linux octave to process the IQ data recorded from the tuner on my laptop, then did send the files via the LAN to this PC, that has octave installed.. Hanged the GPS receiver out of the window to get a betetr signal, now it was pointing horizontal.

Acquired with: rtl_sdr -f 1575420000 -g 42 -s 2728000 filename.dat Works!

formatting link
formatting link

Acquired at a lower sample rate (2048000), seems to make little difference:

formatting link
formatting link

The lower sample rate is prefered as that promises a more reliable streaming via the USB tuner, making long sessions and extremely precise position calculation possible.

These were about 12 second recordings each..

As you can see it takes octave almost 5 minutes to process the 12 seconds data on a 2.8 GHz dual core Athlon, with one core 100% dedicated to octave (with taskset). This is mainly due to search space for unknown frequency offset caused by the Elonics 4000 USB tuner local oscillator offset, it shows up here as about 30 kHz! Now we know it is 30 kHz the search space (for Doppler from the sats) can probably be reduced in the math, should not normally be more than a few hunded Hz for the sats (due to motion).

Actually 30kHz offset for that LO is very good at 1.5 GHz!

This is what was in the air according to my real GPS module: id 13 * elevation 69 azimuth 68 snr 27 id 10 elevation 63 azimuth 286 snr 0 id 4 * elevation 48 azimuth 221 snr 23 id 2 elevation 43 azimuth 287 snr 0 id 23 elevation 43 azimuth 69 snr 0 id 7 * elevation 40 azimuth 166 snr 32 id 16 elevation 15 azimuth 65 snr 0 id 30 elevation 14 azimuth 38 snr 0 id 8 * elevation 14 azimuth 182 snr 19 id 20 * elevation 10 azimuth 120 snr 23 id 5 elevation 10 azimuth 294 snr 19 id 29 elevation 6 azimuth 345 snr 0

Updated diagram, changed the 10 pF coupling to 1nF:

active 1nF GPS -0-------||-------0- E4000 DVB-T stick connected to PC USB antenna | with xpsa spectrum analyzer software SMA ) LM317 ) .3u _____________ | | | | +4.75

+9V top + 15V ----- |in adj out|----------------- | |_____________| | | === | [ ] 220 | | 1u | | | /// ------------| === iu | | | [ ] 6k8 [ ] 680 /// | [1] | /// ///

:-)

Reply to
Jan Panteltje
Loading thread data ...

Added offset for tuner VCO shift of 30kHz, and reduced search space to 10kH, now 17.4 seconds for a solution in octave!

formatting link
QED

Reply to
Jan Panteltje

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.