At 12:55 PM 3/24/01 -0800, Stewart Nelson wrote:
Hi Uwe and all,
The present implementation of WOLF, unfortunately, has very poor
dynamic range. The BPSK signal must be between -20 dB and -50 dB,
relative to full scale.
If you are receiving your own transmission, or one from a nearby
station, you may need to reduce the record level by 20 dB to prevent
internal overload when it starts averaging.
If you are trying to receive a signal buried in noise, you should
set the record level so that frequent peaks reach -6 dB or higher,
even if it means that occasional static crashes overload.
This will ensure that the -50 dB minimum is not a limiting factor.
If your Rx output is dominated by a strong interfering carrier, it may
not be possible to set the record level high enough. In this case,
try to reduce the interference with a narrower IF filter, with a notch
filter, or by preprocessing the recording through an audio editor
with filtering features.
Although 8-bit recordings normally degrade the effective S/N only
slightly, recording in 16 bits can be a big help if the signal is
mostly QRM, rather than noise.
You will have to live with the dynamic range limitations for quite
a while, because my priorities are for improving usability and weak
signal performance. Sorry for the trouble it has caused.
73,
Stewart KK7KA
I think we'd all agree that no apologies are necessary. This software is
remarkable, and the first
practical application of something that developers have "talked" about for
years. You are to be
congratulated. Many have mentioned features they would like to see,
including GUI interface,
real time processing, etc...but I think your priority for development
should rightly be
weak signal performance.
Thank you again for your efforts that have benefited all of us!
Les Rayburn, N1LF
Les Rayburn, N1LF
4919 Cox Cove
Helena, AL 35080
XMGR 184.8988khz WOLF Mode
1LF 187.300khz
Looking for a challenge? Try the ultimate weak signal mode, on 1750 Meters!
Visit the Noise Floor!
http://www.highnoonfilm.com/xmgr
|
