from Hacker News

Signal Identification Wiki

by hhthrowaway1230 on 10/13/23, 11:50 AM with 28 comments

• by curiousfab on 10/13/23, 12:35 PM

  If you find "wave hunting" this interesting but don't have a SDR / receiver of your own, there are plenty of (free to use) WebSDRs to get started: http://websdr.org/

• by MandieD on 10/13/23, 5:23 PM

  For those of you already set up to receive or send WSPR, FT8, or CW (Morse code), make sure you’re receiving and sending reports tomorrow during and around the eclipse (1200-2200 UTC)

  More info here: https://hamsci.org/eclipse

• by kps on 10/13/23, 2:33 PM

  All the recent ones sound like noise, of course, but this one is musical: https://www.sigidwiki.com/wiki/Inmarsat-D(D%2B)_Downlink

• by invokestatic on 10/13/23, 1:17 PM

  Depending on the frequency and source maprad.io may be more useful for identifying signals.

• by freedomben on 10/13/23, 1:54 PM

  The HN hug strikes again! What I saw before it went down was pretty amazing though, and definitely something I will be returning to later when the heat dies down.

• by eternityforest on 10/14/23, 1:21 AM

  Super cool!

  Looks like all the data is in a fairly decently scrapable format. Could one, in theory at least, add auto-identificaton to an SDR app?

  Is there an (possibly ML) algorithm that says "take this single example picture and tell me if there is a match(Which could be scaled differently) in this other picture"?

• by rollulus on 10/13/23, 1:05 PM

  Are there any ML approaches to identifying signals? Since using a receiver that produces sound given a FM/SSB demodulation of whatever true modulation is used, or visually inspecting a waterfall certainly has limitations.