OpenBCM V1.07b12 (Linux)

Packet Radio Mailbox


[JN59NK Nuernberg]

 Login: GUEST

KB2VXA > DSP      11.07.03 23:51l 62 Lines 2763 Bytes #999 (0) @ WW
BID : 10163_WT3V
Subj: Re: M0PZT Re: G0FTD > DSP & pulse noise
Sent: 030711/0123Z @:WT3V.#CNJ.NJ.USA.NA #:10163 [Lakehurst] $:10163_WT3V
To  : DSP@WW

Hello Charlie and Andy,

While eliminating impulse noise via the computer sound card may look easy 
on paper it's quite a bit more difficult in practice. Impulse noise 
eliminators using the receiver's IF has been common practice for many 
years so I shall explain the theory that started it all.

A pulse detector similar to a TV sync separator is used to separate the 
pulses from the signal ahead of the controlled stage. The pulses are 
shaped and timed, inverted and injected into the controlled stage at 
equal or greater amplitude than the original. Since they are phase 
inverted and of equla or greater amplitude, phase cancellation takes 
place and the stage is blanked for the duration of the pulse. Since the 
pulses and resulting blanking interval is quite short, the holes are 
barely noticable or not heard at all. I have used receivers of this type, 
the latest being a GE Mastr2 Lo Band FM mobile unit converted to 6M 
operation. Under the very worst conditions there was absolutely no noise 
and no disruption of received signals whatsoever.

It seems to me that while such blanking may be accomplised either by a 
computer sound card or digital sound editing software, the timing, 
inversion, shaping and insertion of the blanking pulses could be jus a 
bit tricky at best. I see no reason why it can't be done in theory, but 
working it out in fact could prove to be an impossible feat for the 
average Ham or someone outside some company's engineering department. If 
not done perfecly it doesn't work at all, so working it all out can be a 
very frustrating experience.

For that particular reason I suggest checking out a suitable source for 
information from someone who has already done it a much simpler way, the 
hardware method. I expect that the RSGB like the ARRL has a construction 
article somewhere in the archives describing in detail a "little black 
box" that will do exactly what you want without all the fuss. Since 
impulse noise rather than a broad spectrum of various noises is the 
focus, the circuit should be simple and straightforward for that type of 
audio DSP.

Then you could do it the RIGHT way there Charlie and buy a rig with IFDSP 
which does a far superior job than any sort of audio processing digital 
or analog. (knowing grin, really big one too)
Then maybe you just want to download the PROPER software written just for 
the purpose???

(OH! You should see the grin NOW!)


73 de Warren, KB2VXA@WT3V.#CNJ.NJ.USA.NOAM


Preserve the integrity of our network,
stop Internet forwarding!

Message timed: 01:30 on 2003-Jul-11 GMT

Read previous mail | Read next mail

 30.11.2020 12:09:36lGo back Go up