Ah interesting - I was trying something similar last night (thinking slow-scan TV or weather satellite images etc) but was struggling with weird software - this looks like a promising programme though.
4 sections sounds promising as we have potentially 4 quadrants on image. Different encoding schemes/op modes might reveal more information ... or at the very least can be disproved and rejected!
I'd posted on this a couple of weeks ago. A number of CMDRs have reported using FLDigi, Ham Radio Deluxe in the last few days. This will allow reception of FSK (RTTY), PSK, and some other modes that have predefined mark / space widths, baud rates, coding schemes etc. If the transmission (we think is) in the UP audio magically adheres to just one of a huge number of predefined 20th/21st century coding schemes, with a likelihood of an amateur radio mode just because of the wide availability of free decoding software, it'll be a disappointment to me. It would be a solution shoe-horned in to allow people to solve it, whilst having no in-game explanation. The UA/UP morse is bad enough, but was more generic to be (just about) believable.
If you look here
http://www.sigidwiki.com/wiki/Database as has been mentioned also recently, there are a huge number of other variants that you won't be able to decode with the software that is currently being played with.
If we think the signals are Frequency Shift Keying (FSK) / RTTY, then we are limited to standard baud rates 45, 50, 75, 100 etc. We could then examine other baud rates by shifting the speed in Audacity or whatever by a few % to cover the combinations. And then trying the different combinations of bits per character, parity etc. Over the air data like this isn't stereo, unlike our in game sounds. I've taken different channels, merged channels etc, but no joy. I did note some more recent UP sounds posted (around Maia I think?) were not as balanced left / right as the sounds posted on the front page, but that didn't make any difference.
After a long search I found I could play with a demo version of this discontinued software that only runs on older OS
http://www.pervisell.com/ham/skysweeper/GenericDetectionFrame.htm
that had the benefit of being able to
detect the baud rate, and receive any generic FSK / PSK settings as you wished, albeit only with lower bitrate sample wave files. Even with cleaned audio after high / low pass filtering, I was unable to recover a consistent baud rate or data. And I was not happy that really obscure old software accessible to all (or other expensive commercial offerings) might be needed to find a solution.
It was about at this point I fell out of love with this as a possible solution. It was too contrived, not fun, and by far not as simple as other in game puzzles, or the morse or spectrogram plot detection (and consider how long it took to work that out.) Ideally, it might take an age to work out originally, but then the earlier puzzles can be worked through quite simply once you know what to do. Going down this new data route, even if we found the answer, it wasn't looking at as if it would be as simple to replicate the answer - hence my feeling that it was not the correct line of analysis.
I'd not tried Fldigi before, so I've given that a few hours work today with various original, cleaned, speeded up, slowed down .wav files with all of the rate combinations available, and still no joy. I might have just missed the magic combination. Perhaps it is meant to be as random as the CMDRs patrolling in game for days on end. My conflicting thoughts in my head are that the puzzle is not meant to be easy, and requires work, but the whole process should be relatively fun and give some sense of achievement if progress is made.
I have both worked and played in related radio and data fields for over 20 years, and if there's really something there we're meant to find, I'll eat my Tribble. But MB said there's still more in the UP message to find.... somewhere...