Yeah, I think I was wrong about 7 - I can't remember why I thought that? Maybe there was room for 7 between each howl? In any case, I agree that there are usually 5-6 audible purrs. This image shows the two different types fairly clearly. I've tried to remove as much interference from the howls as I can, but I think that there are still masked purrs.
I wonder if the drifting in frequency of the purrs is consistent with doppler motion? If you are moving past an object, the apparent recession velocity changes (even if your actual relative velocity is fixed). I suppose that the ship was slowly drifting past the artefact as the sound was recorded.
edit: The pulses we see are not compatible with Manchester encoding as there are things like HHLLHL which I think cannot occur in a manchester encoding. Possibly this is Delay Encoding:
https://en.wikipedia.org/wiki/Delay_encoding
Where the HHLLHL would translate to 001, i.e. if there is no transition then it is zero a transition in a pair then it is a 1.
OK, so it is 5 or 6 then. Drat.
It's not doppler in this case, the ships are - most of the time - stationary compared to the UA. Besides, it does do doppler when you're moving - it sounds different.
Err, hm, compatibility in this case depends on entry point - look at the example picture on the wiki page - the GE Thomas version between the second and fifth data bit. it's exactly HHLLHL, but to get a manchester encoded value, you need 2 more blips on either side. (?H HL LH L?)
That said, the delay encoding does worth a look too, as it's description is exactly what we see here. "It's guaranteed to have transition every other bit"
Finished 'And Here the Wheel'...
My thoughts:
- UAs are not kosher.
...
Yeah. Erm. Eww. Uck. I'm pretty sure it's not! Fish sounds or not
(I know whales are not fish)
A Cmdr on our G+ group suggested Gray code?
https://en.wikipedia.org/wiki/Gray_code
I don't remember seeing that one before. It might not be anything, but I'm throwing it out there for consideration.
Doh. Work cut out for and stuff. MOAR coffeee. Now.
I believe the sounds you refer to as "high", "low" are individual symbols/letters. I will try to illustrate what to look for in the audio for others that do not know what we are talking about, once you know what to listen for you will hear it in every video/recording.
This is a video zenith uploaded on youtube 10th july with "sleipnir like" audio (it is actually recorded in sleipnir hehe).
https://www.youtube.com/watch?v=KUNlNDULuuA
Because the noise is so cleared up it is easy to show on a spectrum what the different sounds "look like". Below is just a segment from the video showing the different parts of the audio. The sounds in question is marked with grey boxes.
View attachment 55362
Now, below i have tried to cut out and illustrate all of the sounds in question (we can call them "words") from the video (watch it and try to listen for these yourself

) as well as identify the 9 different reoccurring "sub sounds" (we can call these "letters") plus a "small purr". I have named them after the way i differentiate them myself.
I know all of this just sounds like wobbly noise and that these cant possibly be separated into individual sounds just from a spectrum image but try to believe me when i say they are indeed distinguishable.
View attachment 55358
These 9 "letters" return in different combinations in all recordings. There are reoccurring "words" but i have yet to find any repeating series of words (sentence if i may).
Whooa! Hold on there buddy! I relate to this the same way Drax does to metaphors. I believe you when you say you can hear the difference, but it... goes over MY head.
What if we dial back a notch? I think you could have possibly identified the different types of sound files these are made out of

Here's an experiment for you: can you take a look at the actual Morse code chittering - I'm pretty sure there are at least 3 types of beeps, and 1 or 2 boops (using my less than completely functional ears), still since it's Morse, we ought to "normalize" these to translate it.
Is it possible to categorize the 9 sounds into 2 distinct groups based on pitch? Then look at what you get in "one in-between purr session" - as I described above: high-high-low being one example "session" - to try to form "words".
Or 3 distinct groups and have a go with that?
Besides, if you did make a list / spreadsheet with your findings as is, I would love to take a look! Who knows maybe it is 9 letters (numbers? coordinates?) we just need more data... Who knows? Well, we know who knows. But they remain silent for now.
Love the theory.
The trouble with it is repeat deployments do not seem to produce the same set of purrs.
Aye it's possible it cycles and the "entry point" is different each time but people have looked at this and not seen a repeat.
The trouble is it's quite a long time ago I've looked at this and you become not quite so sure of what your conclusions were!
We've actually done this test before, but I think for slightly different reasons but...
Perhaps a good test for your theory would be to UA deploy in a system with a single star only, ideally with a short morse name (to hear more purrs) record, then repeat and compare both audio tracks?
I do have quite a few videos that I made in Bast (short name) - when it's not close to a planet, because that's long. Could use those for a preliminary run.
It is a bit worrying that the purrs don't seem to repeat. If we could find a pattern there - a cycle as you say, that would count as a breakthrough in my book.
I may be wrong I don't think they have been well analysed as you say, it's just people have seen and looked at them and I think generally they think there's nothing much to them and attention goes elsewhere. All I can tell you about them is that there appears to be 3 or so variations, ie they are not dynamic as such like the morse, more they are one of a small set of types, like each honk/wail is from a set of two types, or the purrs are one of a set of two types.
OK then, let's see what we can make out of them.