Wolfgang Büscher wrote:
I do confirm that there is a blocking effect in the WSPR decoder, but
I am not really sure about the details...
......
But drawing any conclusion may be premature - some neighbour may
simply have switched off his local QRM generator.
My impression is that some processing stages within the WSPR decoder
lack dynamic range, or the FFT has significant leakage, or some part
of the calculation is done with short integer values (instead of
double precision floating point values). The decoder is open source,
but lacking any knowledge of the Fortran programming language, I won't
try to understand the algorithms.
.......
73,
Wolf DL4YHF ...-.-
Hi Wolfgang, LF
i also can confirm some sort of the "blocking" effect but indeed it is
to early to draw serious conclusions . when i red your mail and you
mentioned about the FFT
as a "side product" of the ongoing discussion i just realized why
wspr is using this , say, strange, 12000 Hz sampling rate ( there was
also some fuss about this
particular sampling rate due to its non-integer downconverting from
hardware established 44100 Hz and therefore - it is guessing - causing
some problems with decoding the signal properly) . the wspr is using
bin size of FFT =8192. In other words , the 162 data bits are audio
sampled 8192 times each and therefore when using 12000 Hz sampling rate
the wspr transmission lasts 162*8192/12000 = 110.59 seconds - less
than 2 minutes - it composes nicely into 2 minutes periods of TX/RX
etc.. providing also a room for time adjustments t/r switching quick
decisions what # of tx message to pick up when being in the wsjt/qso
mode ..
if the 11025 sampling rate was used than we have 162*8192/11025 =
120.37 seconds i.e. JUST OVER 2 minutes period with all the
'consequences' ...
73, peter, sq7mpj
qth: lodz, poland/jo91rs/
|
