Re: [Flexradio] Noise Blanker and large S9+ signals outside of passband
I have had excellent results with NB2 for the line noise around here (its bad enough to keep me off the radio if I cannot use it, the regular NB doesn't help). It seems to me that the problem Kurt is describing has gotten worse over the past 6 weeks or so. Just an observation in case anyone sees something that changed to exacerbate the problem. It does make NB2 almost unusable in contests with very loud signals. Neal Campbell www.abrohamnealsoftware.com AIM:nealk3nc On Dec 19, 2007, at 10:47 AM, KX5KW wrote: > Kurt, > You're not the only one. It makes the noise blanker almost useless > for me since there are always other signals in the 96kc if range. > > Bob McGwier, N4HY, has some excellent ideas for noise cancelling in > the code, but he has a LOT of other projects going on. > > I use a modified version of MFJ-1025 here and although it is very > finicky to get set up and must be reset any time you change > frequency, I would not be able to hear some signals without it. > PM me if you would like any particular info on my setup. > > Kent/KX5KW > > > On Wednesday 19 December 2007 08:21:44 am Kurt Vangsness wrote: >> I've noticed that when NB or NB2 are engaged, large (S9+) signals >> that are outside of the filter end up interacting with the noise >> blanker causing the noise floor to vary in proportion to the >> amplitude of the large signal. For example, if clicking the NB >> results in a 6 dBm decrease in the noise floor, the amplitude >> changes in the large signal outside of the passband will cause the >> noise floor within the passband to increase by as much as 6 dBm. It >> would be nice if the noise blanker could be restricted to only >> operate within the passband. This wouldn't be an issue if the noise >> blanker was so darn good on PowerSDR! ;-) Anyone else see this same >> behavior or do I have something configured wrong? >> >>73, >> Kurt KC9FOL > > ___ > FlexRadio mailing list > FlexRadio@flex-radio.biz > http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz > Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ > FlexRadio Knowledge Base: http://kb.flex-radio.com/ > FlexRadio Homepage: http://www.flex-radio.com/ > ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Knowledge Base: http://kb.flex-radio.com/ FlexRadio Homepage: http://www.flex-radio.com/
Re: [Flexradio] Noise Blanker and large S9+ signals outside of passband
KX5KW wrote: > Kurt, > You're not the only one. It makes the noise blanker almost useless > for me since there are always other signals in the 96kc if range. > > Bob McGwier, N4HY, has some excellent ideas for noise cancelling in > the code, but he has a LOT of other projects going on. > > I use a modified version of MFJ-1025 here and although it is very > finicky to get set up and must be reset any time you change > frequency, I would not be able to hear some signals without it. > PM me if you would like any particular info on my setup. > > Kent/KX5KW > All noise blankers operate on wideband signals. They do this because the filtering does pulse stretching making it very hard to determine with a good probability what is and isn't a pulse. ALL blankers modulated the wideband signal because they gate the receiver off. NB2 does this less than NB because it attempts to take an insufficient (sic) statistic on the signal and replace what it sees as a pulse with a reasonable candidate for what the signal should be. This is based on find the rank order mean of the signal "in the vicinity". Neither of these is perfect but if the pulse noise is periodic you will modulate the entire passband since the gating on and off of the receiver is like a mixer. If a strong signal (such as CW) is inside the passband and it pass the stupid test for pulse, it will be gated off. This can and SHOULD be replaced with a test for "widebandedness" so strong CW signals do not pass the test. I will work on this. The ultimate tool will be a pulse noise "subtracter". It is not a blanker at all (gating the receiver off). A pulse is detected and then its amplitude and phase are determined. An inverse of this is subtracted from the incoming signal. This works like "noise cancelling headphones" by injecting an opposite, negative copy of the signal to null it. This has the advantage that it is a linear operation. It cannot cause mixing. Furthermore, single pulses can be excised. How can this go wrong? A false positive detection will remove actual signal by subtracting the false negative of the pulse. A fail to detect will not remove the pulse. Another plus is that this will operate over much more of the pulse than the blanker. There are quicker and not quick things in this we can do. All new development is behind a) wideband image rejection b) downsampling 192000 and 96000 internally to get the same shape factor on the IF filters that we have at 48000. c) IMPROVE the shape factor by downsampling to all divisors of 48000 d) finalize the correct settings and operation of the LMS filtering to work on complex signals with the correct parameters. More later Bob -- AMSAT Director and VP Engineering. Member: ARRL, AMSAT-DL, TAPR, Packrats, NJQRP, QRP ARCI, QCWA, FRC. ARRL SDR WG Chair “An optimist may see a light where there is none, but why must the pessimist always run to blow it out?” Descartes ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Knowledge Base: http://kb.flex-radio.com/ FlexRadio Homepage: http://www.flex-radio.com/
Re: [Flexradio] Noise Blanker and large S9+ signals outside of passband
Kurt, You're not the only one. It makes the noise blanker almost useless for me since there are always other signals in the 96kc if range. Bob McGwier, N4HY, has some excellent ideas for noise cancelling in the code, but he has a LOT of other projects going on. I use a modified version of MFJ-1025 here and although it is very finicky to get set up and must be reset any time you change frequency, I would not be able to hear some signals without it. PM me if you would like any particular info on my setup. Kent/KX5KW On Wednesday 19 December 2007 08:21:44 am Kurt Vangsness wrote: > I've noticed that when NB or NB2 are engaged, large (S9+) signals > that are outside of the filter end up interacting with the noise > blanker causing the noise floor to vary in proportion to the > amplitude of the large signal. For example, if clicking the NB > results in a 6 dBm decrease in the noise floor, the amplitude > changes in the large signal outside of the passband will cause the > noise floor within the passband to increase by as much as 6 dBm. It > would be nice if the noise blanker could be restricted to only > operate within the passband. This wouldn't be an issue if the noise > blanker was so darn good on PowerSDR! ;-) Anyone else see this same > behavior or do I have something configured wrong? > > 73, > Kurt KC9FOL ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Knowledge Base: http://kb.flex-radio.com/ FlexRadio Homepage: http://www.flex-radio.com/
Re: [Flexradio] Noise Blanker and large S9+ signals outside of passband
And greetings to my cousin Kurt! Yes I have seen this behavior in all versions since 1.10.2. very difficult to use the NB (NB2 doesn't seem to do much) at any setting more aggressive than 14. Problem is any setting higher than 14 has very little effect. 73, Carl WCØV Kurt Vangsness wrote: > I've noticed that when NB or NB2 are engaged, large (S9+) signals that > are outside of the filter end up interacting with the noise blanker > causing the noise floor to vary in proportion to the amplitude of the > large signal. For example, if clicking the NB results in a 6 dBm > decrease in the noise floor, the amplitude changes in the large signal > outside of the passband will cause the noise floor within the passband > to increase by as much as 6 dBm. It would be nice if the noise blanker > could be restricted to only operate within the passband. This wouldn't > be an issue if the noise blanker was so darn good on PowerSDR! ;-) > Anyone else see this same behavior or do I have something configured wrong? > > 73, > Kurt KC9FOL > > > ___ > FlexRadio mailing list > FlexRadio@flex-radio.biz > http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz > Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ > FlexRadio Knowledge Base: http://kb.flex-radio.com/ > FlexRadio Homepage: http://www.flex-radio.com/ > > ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Knowledge Base: http://kb.flex-radio.com/ FlexRadio Homepage: http://www.flex-radio.com/
[Flexradio] Noise Blanker and large S9+ signals outside of passband
I've noticed that when NB or NB2 are engaged, large (S9+) signals that are outside of the filter end up interacting with the noise blanker causing the noise floor to vary in proportion to the amplitude of the large signal. For example, if clicking the NB results in a 6 dBm decrease in the noise floor, the amplitude changes in the large signal outside of the passband will cause the noise floor within the passband to increase by as much as 6 dBm. It would be nice if the noise blanker could be restricted to only operate within the passband. This wouldn't be an issue if the noise blanker was so darn good on PowerSDR! ;-) Anyone else see this same behavior or do I have something configured wrong? 73, Kurt KC9FOL ___ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Knowledge Base: http://kb.flex-radio.com/ FlexRadio Homepage: http://www.flex-radio.com/
