In this Meteor M 2-2 observation in the APT 137.150 MHz, Observation 956729, there is no signal, yet Satnogs Client has decoded an image and uploaded it. Is this normal or is it a bug ?
I’m not sure if we should call it a bug but the APT decoder will always return data, in this case an image, even if there is no sensible input.
As far as I know neither Meteor MN2 and Meteor M2-2 transmit APT signals on VHF, only LRPT.
Can @surligas chime in on that point? I thought that this decoder would only return an image if it locked successfully on a signal.
Since nobody else replied:
Back when I first ported the APT decoder to satnogs, I noticed that is rather hard to distinguish between signal with a rather signal to noise ratio, which still yield some recognizable shapes in the image and just plain noise.
It has code to detect the sync markers and sync the image to them, however this is again a rather tricky task for high noise signals.
I’ve tried more sophisticated approaches for this, but either they had high false negative rates,
resulting in the first sync at a very late point in the pass, or high false positive rates, leading to syncs from certain types of periodic noise.
Hence I opted for a simpler sync detection, while leaving the decoder in “free running” mode otherwise that will just happily convert any sequence of samples to pixels, if it can not detect the sync markers.
That way stations with a really clean signals can get nice and synced/straight images, while stations with less optimal condition still get some recognizable shapes in their images, even it is a rather noisy pass.