search

Xerbo / LeanHRPT-Demod

An easy to use HRPT demodulator
GNU General Public License v3.0
57 stars 8 forks source link

MetOp Demodulation outputs 0 byte files #18

Open arelli opened 1 year ago

arelli commented 1 year ago

Describe the bug When I try to demodulate a IQ file (.WAV, from a nooelec 2.56MSPS recording) using the MetOp HRPT Downlink option, it outputs a 0 byte .vcdu file, that is of course not decodeable. This happened on two occasions, when NOAA decoding works well with the exact same setup. Also, when I open the files with Notepad++ they seem completely empty, and when I try to use LeanHRPT Decode on them, it shows immediately "Fingerprinting Failed".

Expected behavior Probably I expect the .vcdu files to be in the order of tens of MBs.

Platform I am working on windows 10 Pro 10.0.19045, on a i3-6006U, 8GB ram / 1TB ssd laptop. The recordings are made with SDR#

Version I am using LeanHRPT version 0.1.3

Screenshots Screenshot (97) image

arelli commented 1 year ago

I also tried the v0.1.0 version of LeanHRPT decode, but to no avail. Also tried writing the file in the bin directory of the LeanHRPT executable, same as before.

arelli commented 1 year ago

I hope this is helpful: https://drive.google.com/file/d/1vwmyjjtRR0lDAaTeb6pCxbqJ2Qp-_rUq/view?usp=sharing A MetOp IQ file from a 84 degrees pass over me. I tried it with the above configuration of LeanHRPT Demod and still the same.

arelli commented 1 year ago

It turns out that, If I load the baseband IQ file in SDR# as a baseband file source, and re-record it in 8-bit-pcm format, the decoding works in LeanHRPT Demod. I do not understand how this can be, given that it is a lower resolution version of the original file. Nevertheless, this doesnt work always. In another instance it failed again with the 8-bit file. Is there any chance that this is normal behaviour? That when there is too much noise, no symbols are registered somehow?