Topband: Noise, DSP and Receivers

[Tom Rauch]

> 2)  Computer DSP can only impact Gaussian noise, not
> impulse noise.  So it doesn't work well with receiver
> NR schemes.  In general, the reason for this is, DSP
> samples signals over some sample period, and determines 
> how to treat the NEXT signal event when it arrives.
>  If that NEXT signal is significantly different
> from the preceeding events, the treatment will be inappropriate.

DSP can work very well with impulse noise, just as well as a 
conventional blanker system, if bandwidth of the system before the 
DSP processing is wide and does not overload.

The same is true with any impulse noise blanker. They all need 
considerable bandwidth before the blanker. 

Most rigs do not provide sufficient bandwidth to allow impulse 
blanking through DSP systems because they use filters narrower than 
ten or so kHz bandwidth before the DSP systems. Even a 6kHz bandwidth 
can degrade impulse noise removal.

The major problem with blanking lightning is that lightning crashes 
are not impulse noise. They are a broadband noise with random steps, 
peaks, and valleys. They also, unfortunately, last a long time 
compared  the tick of an ignition pulse or pulse from a power line 
arc. There really isn't a reasonably simple and effective way to 
remove lightning, other than antenna directivity or nulling the 
noise.

Unlike low average power high peak power noise like ignitions and the 
power lines, the duration of each lightning noise "crash" is so 
long that average energy is very high and smothers the desired 
signals. If we disable or blank the receiver during the pulse, we are 
left with no noise *and* no signal.

That's why the brain and no AGC or very fast AGC works so well 
compared to other systems. There are some computer programs that are 
supposed to work pretty well also, but as far as I know they require 
stereo (phase locked) receivers, wide bandwidth, and complex software 
and processing systems.73, Tom W8JI
W8JI@contesting.com