> My opinion on the narrow bandwidth/noise/weak signal phenomenon is that
> when trying to copy a weak sig out of the noise, narrower is worse(er).
> The reason being, the noise is limited to a frequency bandwidth that is
> *really* close to the frequency [say, 600 Hz] that you are trying to copy.
> It's harder for your brain to distinguish that.
Wider filters increase noise power outside the "passband" of your
focus. Using a wider filter does not move the noise to a new
frequency and improve S/N.
While S/N ratio in your "mental area of focus" remains exactly the
same, the wider noise window simply increases noise power and
eventually saturates your ears with so much noise you can't dig the
signal out. Ears and even your brain, just like electronic equipment
used to process signals, has limited dynamic range and IMD
performance.
With non-white or non-smooth level noise, bandwidth restrictions
increase duration of noise pulses or peaks. That's why noise
blankers MUST be ahead of any selectivity in order to function
correctly, and why DSP systems (even in the IF or in the AGC
loop) do not work well on impulse noise when they follow selective
filters.
Filter "ringing" is an unavoidable part of high selectivity, since any
level change requires bandwidth to accomplish the change. When
the bandwidth restricts the rate, we get what most of us call "filter
ringing".
Some filters have peaks in overall response or internal selective
elements that cause excessive ringing for a given bandwidth and
shape factor.
73, Tom W8JI
w8ji@contesting.com
--
FAQ on WWW: http://www.contesting.com/tentecfaq.htm
Submissions: tentec@contesting.com
Administrative requests: tentec-REQUEST@contesting.com
Problems: owner-tentec@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|