Doesn't work for non repeating signals?

[amazingfly]

It works great for remotes that send repeating signals. But I have some remotes that send one signal per button click. I've tried messing with nedges and clicking the remote button as fast as I can etc, but to no avail.

Any thoughts? Or ideas on how the code could be edited? As of yet, I've gone over the RX class and can't quite figure how it could be changed to work with non repeating signals.

And most importantly, thanks for the great code!

[peterhinch]

This could be tricky. As explained in the docs, the receiving device tends to introduce jitter so I average over many frames to determine the pulse lengths. This (obviously) assumes a repeating code. If the code changes on each press presumably it cycle over N codes, returning to the original. I can see two scenarios, one where you know the value of N and the other where you don't.

In the first case you could redo the statistics to sort the returns into N bins, finally averaging and outputting the contents of each bin. The second case would be more difficult. If you have a scope or logic analyser you could figure out the value of N, otherwise you'll have some interesting coding to do. I don't have an instant answer to that one.

Out of interest what type of remotes do this? I've never seen this behaviour.

[amazingfly]

It seems I made a few bad assumptions. It looks like the remote uses 304mhz not the 433mhz I have connected to my ESP32. I really appreciate the reply, and thanks again for the great library!

[amazingfly]

So, seems I actually still have that issue, with the proper modules for the band, according to the FCCID data. It is a remote for a ceiling fan / light. I can't seem to get a successful read on any of the buttons. As they do actually change. For example the light on/off button if held starts dimming or brightening the lights.

I'm considering figuring out how to start and stop the RX so I can capture individual key presses. Though my first experimentation with that seem to completely wreck the quality scores, and not work. I was using my other remote that works just fine as is with you code.

Any thoughts? Do you think my approach may work with further tinkering? I thought about trying to come up with some sort of pattern matching algorithm, to be able to cut down on the random noise. Though that sounds pretty "in depth" and I'm not sure where to start with something like that.

Anyhow thanks agian :)

[peterhinch]

An interesting challenge. Even with repeating signals a statistical approach is needed to deal with jitter. Non-repeating signals will need more advanced stats as described above. It's an interesting project, but not one where I can offer much help as I don't have similar hardware. You could ask in the forum, but I don't recall anyone else tackling this one.