> Unfortunately, audio-stage DSP's, including the popular outboard
> varieties, can do nothing to eliminate the AGC action caused by this kind
> of noise. That's where the noise blanker does its magic (but, admittedly,
> also causes degraded handling of adjacent strong signals)... IF DSP's, if
> inside the AGC loop, should be a better approach, tho I've not yet used a
> rig with one to see how well they work.
AGC is a worry, but a minor one. If the blanker follows the AGC
system, the receiver will lose gain from AGC being held up by the
noise.
Overload of preceding stages is another lesser problem, although it
has little to do with the AGC loop position.
The major problem is selectivity ahead of the blanker.
Any "blanking" or "limiting" noise reduction system, DSP included,
suffers performance degradation when installed after selectivity,
unless the bandwidth is so wide that the pulses are not lengthened
by the selectivity. That's because filtering that limits system
bandwidth, no matter how good, always extends the pulse length of
sharp noise pulses.
That means the narrow "noise duration window" when the noise
pulse is on gets lengthened or stretched, and the blanker has to
punch a larger "hole" in the desired signal to remove the noise.
Eventually the noise becomes so extended that it resembles a
desired signal, and there is nothing the blanker can do....whether
DSP or conventional.
Don't expect magic just because the DSP is run at "almost audio"
(called IF DSP, but really at a few dozen kHz) and included in the
AGC loop.
The only "blanker" that works when pulse duration is high, or
perhaps if narrow selectivity is required on a crowded band to
prevent blanker overload, is a cancellation-type system that is
phase derived. Noise cancellers only work when you have a
separate sense antenna that picks up the same noise that is
causing problems. The system works by shifting phase and level of
that noise until it is 180 degrees from the noise on the main
antenna, and then summing the outputs so the noise total is zero
while the signal total remains essentially unchanged.
There are a few systems available like that. One has no phase
rotation (even though it has a phase shift knob), and it doesn't work
at all except by blind luck. Another shifts phase, but also shifts
amplitude severely and so has limited effectiveness and is hard to
use. And one system shifts phase almost perfectly and has little to
no amplitude variation as phase is shifted.
Canceller systems that work properly can totally remove any type
of noise without harming the weakest signal IF the noise comes
from one source that is not moving and IF the sense antenna
clearly "hears" the noise source.
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
|