Linrad Noise Blanker

[Earl Shaffer]

Hello.

 

I have an older version of Linrad running, perhaps a year old.

I would love to try the version with the squelch, but I have forgotten more

than I learned about the install process so I am struggling through an older version

for now. I think it was the version released just before portaudio was implemented.

 

I have the noise blanker partially working, but only on weak signal CW.

I do not have a noise blanker at all in the SSB or AM mode.

I would very much like the to run the noise blanker in those modes, but don't

remember how to add it to those modes. In the weak signal CW mode

it seems about the maximum bandwidth I can run is around 500 hz which

is not useful for SSB or AM. I am able to adjust the dumb blanker so the

yellow numbers are below 20 by sliding the horizontal slider. I am able to

set both smart and dumb blankers on manual. In automatic mode, the

noise level pulses between two states, so I use manual. I am not able

to make any adjustment to read anything but zero on the blue number

when the smart blanker is turned on.  I am feeding a 160 meter signal to

it which has substantial power line noise. It's about 40db above normal

SDR noise floor. Either noise blanker alone reduces noise (per S-meter) by 10db.

Sometimes by dumb luck, both blankers together will reduce the noise by

20db, but it is not reproducible. I built a relay box that switches RX to the

SDR and TX to my Icom 706MKII which has a noise blanker that does not

work at all on AM and barely works at times on SSB. I use the rig on 160 AM

and would like to test Linrad for that.

Is there a tutorial somewhere from installing to using the noise blankers?

I think I already have the smart blanker calibrated and I have the calibration

files. Last year I built and used the noise generator to make those files.

 

73, Jim Shaffer, WB9UWA.

--

Earl Shaffer,  WB9UWA

 

My EME array photos

http://www.gm4jjj.co.uk/WB9UWA/array_files/page33-1001-full.html

Detailed array photos

http://www.gm4jjj.co.uk/WB9UWA/

Facebook

http://www.facebook.com/earl.shaffer

Linkedin
http://www.linkedin.com/pub/earl-shaffer/12/881/735

[Earl Shaffer]

Hi All.

 

After considerable research I have concluded the following on the Linrad noise blanker.

On 160 meters, with the smart noise blanker, I see 6db of noise reduction. With only

the dumb or both together, I see 10 to 11db of blanking. After turning on the blanker

I see a difference in the wide waterfall, but no difference in the wide spectrum.

I can readily adjust the blanking level of the dumb blanker. I find no adjustment

at all for the smart blanker (except for the limiter which turns parts of the wide waterfall red).

I have never seen the Blue number in the upper right hand corner read anything above 0.

 

I also still do not know how to enable the noise blanker on any other modes than weak signal CW.

I think it might be related to turning on the high resolution spectrum, but I don't know how to do that.

I have been following K1JT's notes on Linrad as well as Leifs notes and example on the linrad home page.

 

There is a broadband hiss that is present on all bands from 160 meters to 2 meters. I think it comes

from an unfriendly neighbors garage where possibly he charges his boat battery with a smart charger.

I have tracked it down on 2 meters. This may be the hiss I am left with after the obvious line noise is

blanked. The 160 meter antenna raises the noise floor of the wide spectrum display by 40db with or

without the noise blanker on during daylight hours. This same hiss can be heard on 40 meters where

higher antennas receive less noise and lower antennas receive more noise (backwards from typical).

My neighbors garage is only about 10 meters from my tower and about 4 meters from my garage.

 

 I know where most of the power line noise is coming from. There are two different poles in

two different directions that I tracked down using both 2 meters and 432Mhz. One of these sources

is strong enough to add 10db of noise to signals on my 440 Mhz FM repeater. Both sources are

about 2 city blocks away. One of the poles can go intermittant when I kick the pole.

 

73, Jim Shaffer, WB9UWA.

[Leif Asbrink]

Hi Jim,

> I have an older version of Linrad running, perhaps a year old.
> I would love to try the version with the squelch, but I have forgotten more
> than I learned about the install process so I am struggling through an older
> version
> for now. I think it was the version released just before portaudio was
> implemented.

I suggest you install the latest version in a new linrad directory.
That is the default if you use installers under Windows.
Mor recent versions are more user friendly (i hope) and I think
you will find that setup is finished in a couple of minutes at most.

Anything that you do will not affect your old installation.
Linrad does not write anything in the registry or elsewhere
outside the linrad directory with an exception for the EME
database which has a fixed place in the file system.

> I have the noise blanker partially working, but only on weak signal CW.
> I do not have a noise blanker at all in the SSB or AM mode.
> I would very much like the to run the noise blanker in those modes, but
> don't remember how to add it to those modes.

"Linrad for newcomers" is the reference page for basic info:
http://www.sm5bsz.com/linuxdsp/usage/newco/newcomer.htm#BLANKER

It says: "The second FFT must be enabled to allow the noise blanker"

73

Leif

[Leif Asbrink]

Hi Jim,

> After considerable research I have concluded the following on the Linrad
> noise blanker.
> On 160 meters, with the smart noise blanker, I see 6db of noise reduction.
> With only
> the dumb or both together, I see 10 to 11db of blanking. After turning on
> the blanker
> I see a difference in the wide waterfall, but no difference in the wide
> spectrum.

Correct. See die block diagram:
http://www.sm5bsz.com/linuxdsp/usage/newco/newcomer.htm#BLANKER
wide graph is before the blanker, waterfall is after.

> I can readily adjust the blanking level of the dumb blanker. I find no
> adjustment
> at all for the smart blanker (except for the limiter which turns parts of
> the wide waterfall red).

The two control bars for routing signals through or at the side of
the blanker are common to both blankers.

The clever blanker requires correct calibration and that the pulses have
a much larger bandwidth than your system so they actually are determined
by the pulse response of your system.

> I have never seen the Blue number in the upper right hand corner read
> anything above 0.

Have a look here:
http://www.sm5bsz.com/linuxdsp/blanker/leonids.htm
Fig 8 for example. The clever blanker removes 2%. This is bad
powerline noise with groups of about 10 pulses repeating at 100 Hz.
That is 1000 pulses per second.

73

Leif / SM5BSZ

[Earl Shaffer]

Hi Leif.

 

The wide graph on figure 2 seems to show a substantial improvment for only

having the noise blanker on while other later figures do not show the same

improvement. Figure 2 is what confused me. It would seem that I do not have

enough line noise to read anything higher than 0 for the smart blanker.

 

73, Jim Shaffer, WB9UWA

[Earl Shaffer]

Hi All.

 

I looked at some archeives and found out that I had previously figured out to enable

the noise blanker. What is not obvious is how to turn on FFT2. When I start Linrad

I have no way to see or adjust it. I found later (much later) that if I "X" back a level, I then have

the opportunity to enable FFT2. This was the missing key information that I read endlessly

looking for.

Newcomer mode is a good start, but it does not lead one well to a more advanced operation.

Now that I have some things figured out again, I'll install the latest version.

The right and left arrows still give me fits at the top of the wide waterfall. Currently one

of my linrad copies has a very narrow waterfall with no numbers at all on it. If I leave

linrad and come back, it remembers my error :(

I will try to keep one of my earlier Linrads useable. As I recall some of the parameters

are hard to figure out and it just doesn't work if one is set wrong.

I have done considerable reading on your web site and the information organization is

greatly improved. Thank you.

 

I studied the Linrad folder and remembered how I had renamed the IQ and calibration

files so they would work in other modes such as SSB and AM. The noise blanker

reduces noise well in the AM mode, but it does not seem to handle strong signals

at all. A strong signal reads minimal signal strength when the noise blanker is on.

 

73, Jim Shaffer, WB9UWA.

[Leif Asbrink]

Hi Jim,

> The wide graph on figure 2 seems to show a substantial improvment for only
> having the noise blanker on while other later figures do not show the same
> improvement. Figure 2 is what confused me. It would seem that I do not have
> enough line noise to read anything higher than 0 for the smart blanker.

http://www.sm5bsz.com/linuxdsp/blanker/leonids.htm
To understand anything you have to read all the text. I suggest
you download the recording and practise on it.

Figure 2 is during a short period without any powerline noise
in the recording. It is intended as a reference for "perfect
blanker operation" when you compare to figure 3 where the interference
is present and the blankers running.
Without blankers it would look as figure 1.

Note the noise floor level level in the wide spectrum.
47 dB in fig 1 and 3 but 23 dB in fig 2.

Further down the page I have tried to explain how the blanker
works - and why. It is not much automated,

The things that affect the blanker are:
The threshold (horizontal bar(s) in the high resolution graph.
The strong signal controls (vertical bars in the high resolution graph)
The number of averages in the wide waterfall.
The number of averages in the wide spectrum.
The selected bin resolution for fft1 and fft2.
Whether second fft uses 16 bit integer or 32 bit float.

73

Leif / SM5BSZ