Loading ...
Sorry, an error occurred while loading the content.

JT44 vs CW simulation

Expand Messages
  • Roger Rehr W3SZ
    Hello all. This is just a brief note to say that at the request of several EMEers I have added two CW files to my JT44 Simulation webpage, which is at
    Message 1 of 1 , Apr 27, 2002
    • 0 Attachment
      Hello all.

      This is just a brief note to say that at the request of several EMEers I
      have added two CW files to my JT44 Simulation webpage, which is at
      http://www.qsl.net/w3sz/JT44.htm . They are near the bottom of the page,
      just after the list of JT44 test files.

      What I did was to create 2 CW-in-the-noise files with the signal-to-noise of
      the cw the same as the signal-to-noise where JT44 had good copy. One file,
      test52cw.wav has cw at -52dB with the noise at -54 dB, and the other
      test48cw.wav has cw at -48 dB with the noise at -54 dB. These are the first
      two levels where JT-44 achieved single-exchange complete copy (although JT44
      did complete copy at lower signal levels with averaging).

      I find I can see the cw signals at these levels with Linrad, spectran, and
      dsp-blaster, as well as with fftdsp, but so far neither I nor anyone else
      has copied the cw in the files correctly.

      Please note that these files are long- 2 to 2.5 minutes in length, as
      shorter file durations made trying to copy the messages more difficult.
      Because of their length, both files are in zipped format on my website.

      Any comments appreciated, especially if you think you've broken the code on
      one or the other of the files. Remember-assumptions can be dangerous ;)

      If you think you've broken the code, please reply to me directly and not to
      the list, so as not to 'spoil the fun'.

      73,

      Roger Rehr
      W3SZ FN20ah
      2 Merrymount Road
      Reading, PA 19609-1718
      http://www.qsl.net/w3sz
    Your message has been successfully submitted and would be delivered to recipients shortly.