Rx auto-tune

SteveM

Sorry if there is already a plan for this or if this is a duplicated idea.

I find that I like to see the entire band that I am working in the panadapter. This makes accurate click-tuning more difficult to do. So, I would like to see a feature that can be turned off/on that reduces the mouse wheel work associated with tuning in the pan-adapter. For example, when the feature is enabled, I double-click to a frequency in the spectrum display, and if I land on or very-near a strong-enough signal, the feature kicks-in and quickly finishes the job or gets me to within a couple of wheel-clicks from top-dead-center. It would probably be fairly easy to develop an accurate algorithm for modes that include a carrier, maybe not as easy to make it accurate for modes such as SSB. But any help to make it quicker/easier to click-tune would be much appreciated.

After this feature is implemented, I think it would naturally lead to a band-scanning feature with all kinds of settings to control the scanning mode. Sometimes I like to multitask.

Thanks.

Al_NN4ZZ

Steve,
Take a look at this idea for auto zero beat on click tune and see if it does what you are looking for.

https://community.flexradio.com/flexradio/topics/ssdr_autozerobeat_on_cw_click_tune

It has 22 votes so far......please add a vote if you like it.

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com

SteveM

Thanks, Al. I added my vote. I'm new to Flex radios so I didn't know what the feature would be named. Maybe this idea should be nixed. However, I don't like that your idea indicates that this feature is exclusive to CW. Does the PowerSDR feature work for all Rx modes?

Also, as I mentioned, the zero-beat feature would be the building-block for a band-scanning feature which is missing in SSDR, right now.

ai6re

Steve, I currently leave the Step size at 500 hz so if I left click on a signal it will tune to the nearest 500 hz. Usually this is right on the signal I want. It also provides pretty quick mouse tuning across the band. I would like to see the right mouse button held down to cause 10 hz mouse tuning for quick fine tuning with the mouse. Don

SteveM

Don,

I usually have my step set high, too. Why? To lessen the repetitive scrolling motion that is required to tune. But during contests or while DX searching, the high step is no good. My whole point is that it would be a lot quicker if I need only give SSDR an approximate indication of the target signal. SmartSDR should be "smart" enough to figure out my intent ;-)

SteveM

Here is the way I would see this feature through to the end.

1. The radio preprocesses the waterfall data to determine tunable positions. That is, the radio finds all of the possible contact signals within the displayed spectrum, filtering out the noise spurs from the rest of the valid signals. The list of contacts should be continuously updated.

2. The contact list is sent to SSDR and displayed across the top of the spectrum panel as click-able tuning positions. Perhaps, as the mouse pointer hovers over each of the contact points, the spectrum and waterfall area are highlighted for that contact point. One left-click and you're there.

3. The scanning feature, based upon user-tailored settings, rotates through those contact points high-lighting the spectrum and waterfall area at its current position to show its progression, and low-lighting all other contact points to show where it will go.

Q.E.D.

SteveM

Whoa - tough crowd!

Never would have thought I was backward for proposing 1-click tuning and skip-scanning. Perhaps I should recondition my thinking more towards the big-knob-on-a-little-plastic-box as the ultimate SDR experience.

Oh well. /thought

Charles - K5UA

Probaby not so (Q)uite (E)asily (D)one. Hi

SteveM

Maybe not trivial. But if we only ask for the trivial, we shouldn't be surprised to get just that. As for the difficulty, it may be worth the reward (Contesters - help me out here!) If the operator can so easily see and recognize a signal in the spectral plot and discern it from the noise, then I'm sure a clever person could accomplish the same (with reasonable accuracy) in software. I have accomplished this in a similar situation with a tele-conferencing application. The conference bridge must determine which of the potential speakers is actually speaking so that those in a (potentially very) noisy environment are excluded from the mix when they are not active.

FYI - the "D" stands for "Demonstrated".

SteveM

To finish my nod to the contesters. Imagine the top of the spectral plot lit up with red markers. Each one represents a search/pounce point. You start from the left and work right using the nifty one-click tuning feature. After breaking the pile-up, you right-click on the marker and select "Worked" which changes the color of the marker to yellow. The marker stays yellow until it disappears (i.e., the algorithm in the radio determines the contact is gone). New contacts appear as red markers interspersed among the other red/yellow markers.

Yeah, I know - I'm a dreamer.

KY6LA_Howard

Steve

As a contester - what you are suggesting requires too much hand movement.

I could see it more likely as follows

Use the tab key to get to the net spot.. have the spot show up automatically in your log
Work them - log them - tab to the next spot.. - system automatically changes the last spot to worked.

Serious contesting is about RATE.. targeting 200- 300 Q per hour or better, you just don't have that much time to take your hands off your keyboard--

SteveM

Yes, Howard, good points. To add to what you said, since the system knows you have worked a particular spot, it would automatically tab-past, or skip over, it as you rotate around the band to pick up the new stations.

km9r.mike

After scanning the ideas section, this thread is the closest to what I would like to see. Which keyboard assignment is kinda not too important, but I think either left or right arrow or up and down arrow would be logical and accessing these keystrokes would jump to the next "mode dependent" signal displayed on the waterfall either up or down the band depending on which arrow was selected. For a next signal that is currently not being displayed an adjustable history buffer could be incorporated say variable from 0 to 30 sec. In other words, if a signal was displayed on the waterfall within the past 30 secs but it is currently not being displayed, the jump to feature would still go to that spot regardless. It will then be up to the user to decide if you should continue to wait there for the return of the signal or simply jump to the next adjacent signal either active or historical. A selectable filter for strength of signal might be useful as well. In other words, only jump to next adjacent signal if it is of this magnitude or greater. 

Using mouse click and wheel is just as easy as using old skool knob for me at this point however both take focus away from keyboard and all testing logging functions are focused on the keyboard so why not have a paradigm shift and incorporate a game changing "jump to" next adjacent signal feature via the keyboard. I am not looking for a "smart" jump to next feature. In other words, only jump to signals that I have not worked, but I am strongly interested in a feature that would reliably jump to the next signal and the only focus I have to do to accomplish this is via a keystroke.

This feature would also be highly desired for those who incorporate SO2R in their contesting. Basically while cqing on one radio (running a frequency), the op is manually tuning (search and pounce) the second radio  to work another contact during the short down times of cqing. This manual tuning workload on the second radio would be greatly reduced if a reliable "jump to" feature was available.

I think the jump to next adjacent signal feature also has great potential for decreasing the workload for those contesting with just a single radio  and for those who are just scanning the bands for everyday amateur activity.

 Not only is big knob so yesterday but for certain aspects big mouse/mouse wheel should be so yesterday as well.

N7AIG

Interesting... reminds me of the process of developing adaptive target lists with partial information for the EKV program. Prospective targets were collected into lists that were successively refined - some discarded, others belief levels were increased. Dempster-Shafer algorithm. Could be an interesting application for such algorithms. (Thinking here about how to find prospective SSB signals)

73 de Dave, N7AIG