DSP - ANF Performance Specification

Lionel

The community has many discussions about the SSDR DSP functions and many of these are qualitative posts saying that ANF, for instance, needs to be fixed. How about tightening focus?

So, lets take ANF. For me it seems to work okay and not having anything to compare I’m satisfied. But then I’m mostly on CW. Nevertheless, what exactly is the spec for the ANF? We have a sense of what it is supposed to do, but within what limits? Its a software subsystem tightly integrated into SSDR so Flex must have some specs to test their design. That spec they must hold close to not reveal intent and there isn’t much detail in paragraph 15.4 of the users guide nor specific performance specs.

That said, you can’t fix that which you cannot measure. We all know this.

Here’s an attempt at a provisional ANF performance spec, ultimately, in some fashion a “users’” desired specification:

1. Modes of operation: SSB, AM

2. Attenuation of interfering audio tone: 30 – 50dB

3. Effect on desired signal in passband: ? %THD?

4. Number of simultaneous tones notched in the passband: 5

5. Passband effectiveness: 1.8 to 5 khz

6. Number of slices using ANL: 1 to 8

7. Attack time: less than 100 msec

8. Effect on slice noise floor: none seen on spectrum

9. Effect on slice IMD: none (how to measure)

10. ?

Then, measure SSDR with with our “users’” specification? These are sort of in order, my order, of importance. Users might consider more or less spec items. Users might consider different required values.  Quantitative results might be a challenge.

All of the above is my idea of what to measure, I’m certain it can be improved.

By looking closely at the ANF’s desired spec a technical discussion can be had that may in fact be useful to Flex engineering. There are no hardware limitations as far as I know.

My thought is to move away from Flex flogging and into the tech specifics that we hams can debate. Field performance of the ANF (also for any comparable rig) can be measured using user accepted provisional specs as a grading rubric. Getting to this point is not easy in a public forum, but, perhaps it needs to be done to move forward.

Similar quantitative specs and answers for NB. NR and WNB could be considered.

So, for ANF, what exactly is it that we want?

ctate243

Lionel--

First-- IMO, Spot on.  Keeping issues on point, specific and with supporting information either about the way it acts differently then desired, or specifically how it is not perceived as working correctly and how to duplicate the issue is the best way to get the visibility of the team.  It speeds re-creation of the issue both in alpha testing and in the programming lab, and is our best chance to get certain issues corrected.  Keeping on point while being respectful is simply the right way to do it and a win win for everyone in the flex community. 

Jim Gilliam

I think he misses the boat completely. The problem with the ANF is when someone talks the notch depth decidedly reduces. The ANF should respond to a notch exactly like the TNF. The TNF works superbly.


Jim, K6QE

ctate243

Jim. What he is saying is there is a perceived issue with ANF, and the best way to help identify the issue specifically is to lay out in technical detail, what specifically is perceived not working, show instructions on how to duplicate the issue, and finally what the desired behavior would be, and leave it at that, without any criticism.  My comment was on the presentation strategy, not the specific issue.  

 

Bill -VA3WTB

Chris, I understand what you just said and I understand the thread, But bare with me a bit please.

If I understand things that Gerald said in the past, is that Flex is doing the DSP funtions very none conventional. They were developing something no one else was doing, as an example, wide band DSP coding. I would imagine Flex is keeping their work guarded?

So then, how can we be sientific and staying within the focus here without really understanding just how Flex is even aproaching this?

Robert Lonn

Keep in mind Flex cant talk about this WBN and other features of their radios!! , it is a Secret. Flex is working with the newly created Space Force.. The Wide Band filter is actually designed to eliminate noise generated from the Inter Steller Warp Drive engines created from Alien Space Craft..This must be true, since I heard this on "Coast To Coast" late night AM radio!!! LOL

Robert

Geoff AB6BT

I don't believe we need to know the algorithms or topology used internally. Just consider it a black box function. Define what you want it to do and leave the engineering of how to accomplish that function up to Flex. 

ctate243

I couldn't have said it better myself.

Bill -VA3WTB

Sorry I must have misunderstood you Chris

Lionel

Notch depth decreasing with desired signal variation (increasing-the same in both directions?) seems undesirable, depending on magnitude.  That said, the effect may not be brought to zero, like the TNF, and if that be the case, perhaps some acceptable limit needs to be established.  This particular issue I think has been reported in the past.

Is it possible to determine at what, in passband signal strength does the effect become apparent? Is it influenced by bandwidth?  What is the variation? What other features, NR, NB, WNB are in use? Do they have any effect on ANF performance?  One would expect not and therefore could be dropped out of the ANF discussion, if the answer is "no." If yes, well, oh boy.....engineering would probably want to disentangle. 

It seems a "notch depth variation with signal strength" is an expectation that could reasonably be included in specifying ANF performance.  Or, as posted, perhaps the intended limit is zero, as is the case with TNF's.  I suggest the effect is also related to, or identifiable with, receive signal distortion or degradation that I seem to recall reading in the community. 

There isn't much field data to capture the effect nor to agree upon limits, except to desire "zero.". Some field data is going to be needed, a significant challenge.

ANF, I think can, as someone mentioned, be considered a black box, in general.

I would be interested in technical details of the ANF performance disappointments.  Is notch depth variation as a function of signal strength the central issue? The only issue?

So,  reported ANF performance issues, at this moment:

1. Notch depth varies with signal strength.   (Performance Spec would be how much variation is acceptable and under what conditions). Someone implied that how engineerring achieves the spec is not relevant. I agree.

2.  ??

What else is wanted for ANF performance?

James Whiteway

Chris, one sure way to test the ANF is to listen to some of the roundtables on 75 and 40 meters on Friday and Saturday nights! Plenty of jamming tuner uppers to test against. And lots of kilowatts too! :-) Seriously, the issue can be reproduced tuning an amp into a dummy load with one radio on the same frequency as a qso in progress on another radio hooked to an external antenna. Or even using a signal generator in place of the second radio and amp(and dummy load) That will reproduce the blowby that happens with a strong carrier tuning on top of a qso. No setting I have tried to date with the ANF will stop the offending signal from popping thru the top of a conversation one is trying to monitor. The closest thing to helping is dropping the AGC-T all the way to zero. But, even then blowby can happen. The Icom 7300 and several other radios I have or had, handle the situation with ease. That's what frustrates me. A $5,000.00 radio cannot do as well as a $850 radio on such a situation as this. I would have thought it would be just the opposite, but it's not. Other than that and a few other issues, the 6600M is a great radio. But, it's a shame this issue has continued from v1.x til v3.x. My sincere hope is the basics are fixed in all three versions of SSDR. Had I not traded up to the 6600M, I would, in all likelihood, would have still been on v1.x till multi client(aka MultiFlex) came along. James WD5GWY

Lawrence Gray

I agree with the need for a specification, but this is work that FlexRadio should be doing.  They should benchmark Noise Reduction, Noise Blanking, and Notch Filtering features in other ham transceivers.

It is actually kind of easy to do.  For example, I purchased ANAN radios specifically to check what other hams told me regarding the superiority of ANAN's recovered audio.  The ANAN NR and NR2 features basically eliminate noise without having a noticeable effect on the audio.  The NB and SNB features remove various  periodic noise without having a noticeable effect on the audio.  The ANF feature removes carriers, period--no effect on the audio and carriers are completely removed.   Any honest comparison of the recovered audio, in real operating conditions, of a Flex to that of an ANAN quickly reveals the superiority of ANAN's DSP functionality.

In the company I operated, we regularly benchmarked ourselves against our competitors--this is a key method for product improvement.  My suggestion is that FlexRadio honestly benchmark themselves against the competition and use that information to develop the specifications for improved DSP functionality.

When FlexRadio introduced the "knob" versions of the 6400 and 6600, Flex entered the mainstream transceiver market.  Hams in the "knob" radio market space are used to noise mitigation features that work well and expect a transceiver in the FlexRadio price range to have these features.

The relatively poor noise mitigation features of the Flex 6000 series have been repeatedly discussed at length over several years without any substantial improvement.   I don't don't think it is unreasonable to ask FlexRadio to address these issues.

Larry, W1IZZ

HCampbell WB4IVF

“The ANAN NR and NR2 features basically eliminate noise without having a noticeable effect on the audio.  The NB and SNB features remove various  periodic noise without having a noticeable effect on the audio.”

But another ANAN user says:

“it amazes me how there's so much difference in the results of noise testing. Scott swears by NR2 whereas I have only once found it to be effective - generally it won't work unless the signal is strong and there isn't random noise like storm crashes, in which case I wonder why use it?  SNB also does nothing for me other than change the pitch ... i've never seen a case where it improved S/N. I guess everyone has different results ... of course I'd think that would not be the case, that sometimes, once in a while, we'd all see some benefit if anyone of us is seeing a benefit. go figure.”

https://groups.yahoo.com/neo/groups/apache-labs/conversations/topics/36995

Yep, go figure.  (-:

Dan KG0AQ

"So, for ANF, what exactly is it that we want?"

I want it to work as good or better than my 20 year old Icom 706 MKIIG. I don't think that is too much to ask. Almost any knob and button radio made in the last 20 years has a better ANF than my 2018 Flex 6400.

 Go to 40, 80 or 160 SSB any night and turn it on. Tell me what you think? You can't miss it failing miserably with +S9 signals..

Flex doesn't need specific engineering input. They know it's broke and should know how to fix or they wouldn't be good software based electrical engineers.. It's not like a bug that just happened. Every hardware DSP chip made in last 20 years works decent.  Sad, but fixable...someday. I can wait.

Bill -VA3WTB

We need to be carful in comparing the Flex DSP to other radios. Flex is doing the DSP in a much different way then other radios. Most modern radios do not do DSP 100% in software, they combine hardware filters with software.

The Flex DSP is wide band filters, the Anan is not. You can't do wide band in the PSDR platform.

The problem for Flex is, they are inventing every time they touch the DSP filters, they have nothing to really compare with SSDR.

Robert Lonn

So I was on 20 meters about an hour ago.. While listening to this W2 in New York here in California,, a strong carrier popped up within the 2.3 KHz Bandpass, looked to be about 1.1 KHz away, someone tuning up his TX.. I hit the ANF and , boom, it was gone while the W2 was still readable.. I think that tells me everything is working fine!!! The W2 was a S5, the tune up was a S-9,,  Robert

Dan KG0AQ

Why do we need to be careful? I understand the electrical differences between SDR and IF based radios  One works with ANF and the other one doesn't. Why do some people around here tiptoe through the tulips? It's a hobby and a radio company. I can wait for a fix.

Bill -VA3WTB

Because we are not comparing the same things. yes it may be true, black and white, it does not work as well as this or that radio. But the DSP is created much different then another radio lets say, that should be considered.

Even some of the other SDR radios do not do DSP 100% in software, so then what would we compare?

Dan KG0AQ

On weak signals and low noise floors the ANF works...ok. Go above S9 with a SSB signal and it falls apart.  Try listening to 40-160M tonight and post your thoughts.  73

Lionel

Dan, I'll try it on 75 this evening. Robert's experience above indicates it works when the intended signal is much lower than the carrier, as you also point out.  So is it reasonable to say that at some point, unknown, the ratio, signal to carrier, "breaks" the ANF; I think it is but at what point and if we reversed the ratio, does it still fail?

I have good signal generator and if I can find a combiner around here it will be an interesting experiment.  Maybe someone else in the community can try this.  I suspect Flex has but maybe in the lab and not in the adverse conditions we routinely encounter; not much value in replicating a lab test.

Has anyone ever used the ANF on multiple carriers? Does it work? Is it supposed to work?  

Lionel

testing ANF on 75 a bust here, way too much atmos.  But on 40m, with S6 atoms noise, and lightning crashes:

AGC on slow througout.

ex:1 signal, LSB 7233, S9, carrier 7230 S9. BW 3.3Khz. ANF seems to have no effect at any setting.
ex: 2 tuned to 7230.950 LSB BW 2.4 kHz, many overlapping SSB. carrier at S9 on 7230, ANF setting has an effect but very small and carrier still audible. 
ex: 3 tuning around 7230 to 7233 BW at 2.4, location of the offending carrier has an effect on the ANF performance. As the carrier is closer to the the tuned frequency the ANF performance is better but still does not knock out the carrier.
ex: 4 moved to 7300 SW station, BW 2.4khz, LSB, carrier at +20db over, tuned to 7301, so 1 Khz tone, ANF knocks the heterodyne completely out with some variation on modulation peaks.  Variation is minor. 
ex: 5 moved to 7315 SW station with carrier at +30 to +40 QSB, tuned to 7316 and ANF eliminates the heterodyne- AT ANY SETTING.   Once again, modulation peaks cause ANF performance to vary, but again, minor. 
ex: 6 on 7270.7 an S6 carrier, slightly above background noise, at 7269.7, ANF had no effect at any setting. BW 2.4Khz.

It's just data and a few scenarios.  But I would say that ANF performance is a function of the intended signal, and it's modulation peaks, AND the background atmos noise.  

Can anyone replicate and/or add to the data?

As a one off, I tried the ANF on CW signals, LSB mode two or three signals in the passband.  The ANF takes out one but not all and notch effect varies with keying, that is there is a time to attack the heterodyne and time to release. No real measurement here, but the release time seems shorter than the attack time.   ANF can take out one heterdyne only based on this informal test.

I feel this is a minor step in characterizing the ANF in real world scenarios and the results, though far from definitive, do seem to substantiate the comments posted by others in the thread.

Craig Williams

I agree but I was flamed in the community for my comments.

ctate243

So based on what I am hearing here, it sounds like it works in certain conditions worse or not at all in high noise floor conditions, and usually on only 1 heterodyne, with improvement of success the closer into the tx qrg you get.  thats a sentence I can wrap my head around and a good message to the dev team.

Bill -VA3WTB

Chris, why I said what I said yesterday is because it is interesting how Flex is doing this.

In PSDR for the ANF there are settings we can custom. Taps, Gain, Delay, leak.

I wonder if they are making the same changes in the SSDR code mush as the same way in PSDR?

Or is there much more involved.

Lionel

I decided on a more analytic approach so I brought up Spectrum Lab - DAX to SL. 

Interesting data.

NB. NR, WNB off, AGC slow, AGCT set at approx knee.

Radio China just came up on 7285 so tuned to 7286 LSB and RC carrier showing about -70 dBm and I have a 1,000 hz heterodyne.

Using Spectrum Lab and ANF OFF the 1,000hz showed about -26 dB.
ANF On at 50% -45 dB
ANF On at 100% -57 dB. 

All of these are +/- due to QSB.

But interestingly, the Radio China modulation did not seem to have a major "audible" effect.  That is there was very little change that resulted in me hearing the tone.  On SL I could see the amplitude of the 1000Hz tone moving up and down with modulation, as expected, but the attenuation from the ANF stayed about the same.

I did a similar run using WWV at 10Mhz and offset by 1000 hz and even with the various voice announcements and time ticks and tones I was still seeing 35 - 40 dB notch on the 1000 hz tone.  It never was audible.

This seems to contradict what I found last night, but conditions are different at 2000z versus 0100z with a lot more noise,, signals, and QSB. 

I'm going to try Spectrum Labs on the later evening 40m conditions and try to get some meaningful data.

So, yes, the ANF works, and it is really pretty darn good with nulls in some cases of 40 dB.  But it seems affected by numerous factors, and using a haphazard search for test cases leaves a lot to be desired in identifying those factors.  My guess, the very high level of noise. QRM in the passband is too much data for the ANF to process.  That said, I have zero knowledge of how they handle the ANF process; so maybe I'm way off. 

The ANF zeroes in, apparently, on the strongest heterodyne and perhaps with multiple heterodynes (40m at 0100z + for example) at near the same amplitude it loses the ability to process the data.

I guess information can be good with a proper dose of skepticism. 

If anyone else wants to try Spectrum Lab, it's free, and the input can easily be selected from DAX. It has a bit of a learning curve - it's sort of a Swiss army knife of audio analysis.

It would be very interesting to have more data from users reporting significantly poor ANF performance to see if their scenario can be replicated.

Craig Williams

Finally an actual measurement. What a relief.

Gary Schulz

Is ANF supposed to work on AM or not?  My issues are far from subtle; The actual button is not clickable when on AM for me...  Also WNB just causes the spectrum to jump up and down and it acts like the AGC just starts pumping/oscillating.  Debating the relative performance of the DSP NR type functions is basically a moot point because non of the functions seem to perform well at all, at least on my 6700...My Elecraft KX3 seems to do a better job in many circumstances though I know it is not done in SW.

Burch - K4QXX

I have issues with the ANF but the WNB works great for me.  If you are getting pumping with the WNB, you have the level to high.  On some bands I can run the WNB at 90% or more and on other bands I can't run it more than about 70%.  Some days I have S9 noise on 20 meters and the WNB takes it down to S5.  If you run the WNB to high you can get all sorts of artifacts and pumping on the panadapter.

Joe N3HEE

Have you identified the actual noise source that your WNB works with ?  I have not seen any improvement at all with WNB turned on.  I suspect it only works on a very limited number of noise types ?

Burch - K4QXX

My issue is with powerline noise.  I have a buzz sound on most HF bands.  It gets really bad when it's dry outside.  Once the summer rains come and clean off the lines here in Florida, my noise gets better.  Before flex added the WNB, some days I couldn't use my flex and had to use another radio.  The WNB has done wonders to my situation.  That said, the two NBs on my old Flex 5000 worked better than the WNB on the 6000.  But the WNB was a huge improvement for my type of noise.

Mark K1LSB

Lionel,

My main problem with the performance of the Flex ANF is not that it's not 100% effective in all cases, but rather that it so severely degrades the quality of the audio any time it's activated.

The TNF function displays the same objectionable behavior. It's very aggravating any time I need to use either feature -- I'm torn between deciding whether to grit my teeth and put up with the degraded audio or just shrug and put up with the squeal of the birdie.