JT65B spurious callsigns?

[G3LVP]

Hearing the JT65B tones on GB3VHF (144.300) some time back I was encouraged to load the software to see how it worked.

What I can't understand it that whilst the GB3VHF callsign and locator are decoded OK I also see other callsigns like DA0NRW, IK0BZY & VE4DK apparently on the same frequency. Obviously this can't be true partcularly in the case of the VE. How and why is the program generating these callsigns?

Going back to read the user manual again....

73....

Ken
G3LVP

[G7CNF]

Ken.

You will not necessarily find an answer that fully satisfies within the manual.

There has long been a debate over whether JT65 really consitiutes a mode. Unlike JT6m and FSK441, both of which rely completely on the actual tones detected, the JT65 decoder contains algorythms which process the data received and depending upon the user settings, will make 'best guesses' or compare to a list of calls contained in the call3.txt file.

There is a valid argument which I will not reproduce here, which questions how, when using the deep search mode, it is possible to obtain two calls and a report in just 14 bits of data! Of course, it isn't!

The more scrupulous users of JT65, most of whom are EME enthusiasts, will delete the call3.txt file and not enable the aggressive or deep search modes and rely only on the tones received. This is closer to the spirit of the old days of EME with a tape recorder!!


So Ken, if you go to the Decode menu > JT65 > select "No Deep Search" and then go the wsjt installation directory and delete call3.txt or rename it and try decoding the beacon again, see what results you get....

Don't get me wrong, I am not against the mode, I am after all one of JT6m's biggest fans, and I also concede that JT65 is the most sensitive of digital modes ever to be released into the public arena, but it should come with a warning, as with the wrong settings, novices could be easily duped.

73 es GL

Nige.

[G3LVP]

Nige,

Thanks,

That confirms my suspicion that the program was 'making' calls up.
I'll do as you suggest and delete that .txt file, however if you call a station does the program then 'learn' that call and construct it out of the noise?


73....

Ken

[G7CNF]

There isn't a simple yes or no to your closing question, as various combinations of settings can yield differing results.

If you are afraid of 'call-construction' as you so succinctly put it, then do NOT check the Sked box, just above the GenStdMsgs button (on 5.9.6.309). Any call entered in the "To radio:" box will be favoured as the decoder churns over the data.

This option will utilise the algorithms that the more scrutinous may wish to avoid...

If said option is not enabled then one is working in "random" mode which enjoys the least tinkering from the software and more closely allies itself to conventional FSK modes.

Hope this heps and would you like a JT65b QSO?

Regards,

Nige.

[G3LVP]

Thanks Nige,

I probably won't spend much more time playing with this mode now that I know a bit more about it, if I hadn't heard a signal with my own ears I don't think that I would claim to have had a QSO. However it would be interesting from a propagation point of view if more beacons also used JT65 of FSK 441 etc.
There wouldn't be much point in using any of these modes between Cheltenham & somerset unless I disconnected the antenna!


73....

Ken