LF: Re: Interesting False WSPR Decode

["James Moritz" <james.moritz@btopenworld.com>]

Dear Andy, Eddie, LF Group,

Actually these "double false decodes" are not particularly rare, although 
not usually correctly formatted as a beacon signal, so they don't make it 
into the database. For instance, O4V1FS appeared twice at 1640utc. I think 
they must be an artefact of the demodulation/decoding algorithms used by 
WSPR - don't ask me how! There have actually been some fairly impressive 
bursts of noise this evening.
I have a feeling that this kind of behaviour is linked to times when WSPR 
takes a very long time to complete decoding. In fact, sometimes the blue 
"decoding" button in the WSPR window remains active for a whole 2 minute TX 
period, and overruns into the next period - this may be when these decodes 
appear. This behaviour seems to be associated with bursts of noise. Or at 
least, so it seems, but it happens quite rarely, and so life is probably too 
short for me to check! So it might be some sort of bug.
The false decode of an "extra" G3ZJO on 503.989kHz is curious too, 
presumably it must be related. The DT time offset figure in the WSPR readout 
is -1.4s for the real signal, and +2.9s for the imposter - the two "7D0LX"s 
were also offset in time with DT of  +5.7 and +6.4 seconds. A long delay 
echo seems to be stretching credibility a bit!
Of course, there also have been genuine double decodes, where a station has 
had hum sidebands and so on.
Cheers, Jim Moritz
73 de M0BMU



> Jim -
> How did you manage this? Never seen one like it before.
>
> A false WSPR decode, but the identical message appearing twice, 124 Hz
> apart.
> Must have been an interesting burst of pulsed WB interference to generate
> in.