VORs and SDRs part 3 supplemental materials


I recently posted part 3 of the VORs and SDRs series. In this video, I look into the instrument landing system (ILS), and how it works. I just uploaded the raw baseband data and GNU Radio companion document for anyone that wants to play around with it:

GNURadio Companion GRC file
Baseband data

, , , , , , , , ,

  1. #1 by Kevin Reid on March 4, 2015 - 8:24 pm

    Would you care to upload the simultaneous VOR data (together, not baseband)? I have something it’d be useful test data for.

  2. #2 by hpux735 on March 4, 2015 - 8:30 pm

    Do you mean one file with all of the signals in it? I can’t do that because only two were within one SDRs worth of band. I can upload that one file if that’s close enough. The other option is I have the baseband of each in separate files that are all sampled at the same rate.

  3. #3 by Kevin Reid on March 4, 2015 - 8:37 pm

    The single file with just two signals would still be useful β€” I’m looking for guaranteed time alignment of the multiple VOR signals. The third one (in baseband or not, whichever) might also have some value since it still came from the same flight and I could glue it together close-enough.

  4. #4 by hpux735 on March 4, 2015 - 9:05 pm

    Ok, I’ll try to remember to upload them all tomorrow. They were captured in the same gnu radio flow graph, so they should be at least close…

  5. #5 by hpux735 on March 5, 2015 - 4:35 pm

    @ Kevin Reid
    I’ve uploaded them to a server at work. The files are 17GB each, so you’ve been warned! πŸ™‚

    http://ramone.coas.oregonstate.edu/tuner_0.raw
    http://ramone.coas.oregonstate.edu/tuner_1.raw

  6. #6 by Kevin Reid on March 6, 2015 - 7:16 am

    Got them, thanks! I think I’ll attempt to work out the sample rate and center frequency solely based on the signal content…

  7. #7 by hpux735 on March 6, 2015 - 7:36 am

    @ Kevin Reid
    LOL. Ok, well if you have any trouble with that let me know.

  8. #8 by Joe Dubner on March 13, 2015 - 4:12 pm

    Enjoyed the YouTube video of your LOC demodulation experiment. Right away I recognized taking off from Corvallis and the URL for your raw data confirms that. I’m just north of you at Independence and have an airplane so I’m always looking flying activities. Also have a ham ticket (K7JD) and a B.S.E.E. that expired long ago πŸ™‚ I’d be glad to haul you and your gear around to sniff some RF if you tire of renting Cessnas. Give me a shout if I can help.

    73,
    Joe

  9. #9 by hpux735 on March 13, 2015 - 5:57 pm

    Hi Joe!

    Wow, this comment could not have come at a better time! I am almost finished with the programming for a triangulation experiment and video, and that was going to be the end of the series. I wanted to do a real-time experiment, but I couldn’t think of any way to do that while also flying! πŸ™‚ Can I email you in a week or so to discuss details? (If your email from the comment form is correct, I’ve got it.)

    Thanks again for the offer, I hope you’re serious because I’ll take you up on it! πŸ˜‰

  10. #10 by Joe Dubner on March 13, 2015 - 6:56 pm

    Absolutely, William; I would be pleased if you would contact me when you’re able. Yes, my email is correct and the phone works too (REDACTED). Looking forward to hearing from you.

  11. #11 by hpux735 on March 13, 2015 - 7:23 pm

    Great, Joe! I’ve saved your phone number into my contacts (and removed it to keep it safer). I’ll be in touch! πŸ™‚

  12. #12 by Kevin Reid on March 28, 2015 - 4:26 pm

    Re sample rate and signal frequency: My first strategy was to measure the apparent frequency of the FM sidebands and compare to the standard value (9960 Hz). That was low-precision and unnecessary, as I was reminded when I actually demodulated and heard a (wrong-pitch) voice on one of them.

    Instead, I listened to all three of them, copied down their Morse IDs, and looked them up on airnav.com (EUG, CVO, ONP). This gave me the frequency separation between the two which were in the same file, tuner_0 (1700 kHz), which is sufficient information to set the sample rate correctly (and I’ll assume tuner_1 is the same rate as tuner_0).

    Then determining the center frequency is just a matter of subtracting the observed relative frequency of the signal from the published transmit frequency.

    Sample rate: 2,048,000 Hz
    tuner_0 center frequency: 116,310,000 Hz
    tuner_1 center frequency: 112,405,000 Hz

    Now that I have a file with simultaneously received real VORs (which should intersect at a point which moves at a reasonable rate), I can calibrate against them…

  13. #13 by hpux735 on March 30, 2015 - 12:23 pm

    Well done! πŸ™‚

  14. #14 by Flo Behner on April 9, 2015 - 11:21 am

    @ hpux735
    We tried to download the baseband files (tuner_0 and tuner_1) but the first failed at approx. 8.8 GB and the second after 226MB. We tried different systems and networks but the problem remains. We would like to analyze VOR signals in the signal processing lab at our University.

    Do you know of any problem with the server or can you provide the data through any other service?

  15. #15 by hpux735 on April 9, 2015 - 12:10 pm

    Thanks for letting me know! You’ve discovered a disk failure on my server, and I’ve since remedied it. The files should be back up in about 4 minutes. Make sure to let me know how the analysis goes. I’m very curious.

  16. #16 by Matt on March 9, 2017 - 3:35 am

    Hey, thanks for the files!

    Please correct me if I’m wrong: To calculate your position using the ILS you have to calculate the DDM (Difference in Depth of Modulation), which you get from subtracting the Modulation Index of one signal from the other. As I can see in your GRC file, you are subtracting the signals directly, without obtaining the Modulation Index of each signal first (even though the result shows the position of the plane anyway).

    So here comes my question: What does come out at the output of your diagram? Can we consider that as the real DDM or it’s just an illustrative value? If so, which is the scale of the “Counts” axis (mA, percentage)?

    Thanks again!

  17. #17 by hpux735 on April 11, 2017 - 6:52 am

    This one was a bit of a hack job, to be honest.

    If the counts axes has any meaning, I’d say it’s an uncalibrated proportion of the planes position across the localizer. So, roughly percentage, and an illustrative value. πŸ™‚

(will not be published)

Please complete this capcha. I get almost 1000 spam comments a day! * Time limit is exhausted. Please reload CAPTCHA.