Not having any luck determining the modulation & framing for this uHETSat on 401.500.
As you can see below, it’s a STRONG center carrier followed by a quick flash (just under 1/2 second) of data packets around 40k wide. Not matching up with any of the common modulation types, but I’ll keep trying!
Hi Scott,
I am also trying to get some idea of the modulation used, at the moment I trying to find out what radio is onboard and go from there.
Jan PE0SAT
This observation has an interesting interferer: SatNOGS Network - Observation 8640718
It appears to be perfectly doppler corrected to the TLE in use (58467), but is a bit lower in frequency. SpIRIT’s beacons are also visible in the same observation.
Maybe this is one of the other commercial satellites?
EDIT: It appears to be ID 51832 - SatNOGS DB - OBJECT J - 2022-019
Yeah, now it was set to 58472… and it looks good! Thanks!
Here is my best uHETSat packet from GQRX as a float-32 .RAW file w/ sample rate = 93750:
Thanks, could it be MFSK or PSK?
If a check your file in Signal Analyzer :
Here i’m converting this signal from IQ to real
And then if i open the phase detector, i have garbage all along the strong carrier :
and for the flash i get this :
which seems to be clearly data.
There are more or less 8 tones in data segments :
But i’m far from being an expert so i could be wrong
Latest status update, still we have only 11 from 25 TLE sets, so we are keep watching before finalize identifications:
| Temporary NORAD ID | Satellite | NORAD ID to follow | Other NORAD IDs | Identified | Deployed | Last Update | Notes |
|---|---|---|---|---|---|---|---|
| RECEIVED AND DEMODULATED | |||||||
| 98968 | ION SCV Daring Diego | 58469 | - | NO | YES | Received, continue to follow 58469 | |
| 99320 | EIRSAT-1 | 58472 | 58471(unlikely) | NO | YES | Received, changed to follow 58472, other possible 58471(unlikely), removed 58470 | |
| 99011 | SpIRIT | 58468 | 58467(unlikely) | NO | YES | Received, demodulated by 3rd parties, changed to follow 58468, other possible 58467, removed 58466, 58470, 58471, 58472 | |
| 98971 | ENSO | 58470 | 58471(unlikely) | NO | YES | Received, changed to follow 58470, other possible 58471(unlikely), removed 58468, 58472 | |
| 99032 | Hayasat | 58471 | 58470, 58472(unlikely) | NO | YES | Received, changed to follow 58471, other possible 58470, 58472(unlikely), removed 58465, 58467, 58468, | |
| RECEIVED | |||||||
| 98969 | uHETSat | 58466 | 58465(unlikely), 58467(unlikely) | NO | YES | Received, continue to follow 58466, other possible 58465(unlikely), 58467(unlikely), removed 58468 | |
| NOT RECEIVED WITH FREQUENCY | |||||||
| 98970 | KOYOH | 58465 | all | NO | UNKNOWN | Randomly chosen to follow 58465, all the objects are possible candidates |
Hello,
Any chance you had time regarding this ?
yeah, I hope send you a link by tomorrow ! 
Has anyone been successful decoding SpiRit? FM demodulated wav file here.
TKS es 73,
Bob
N6RFM
If possible, please share an IQ file, SpiRIT seems to be using BPSK
It’s using 9600 baud FSK. See here: SpIRIT Nanosatellite (#267) · Issues · librespacefoundation / SatNOGS Operations · GitLab
Note that they haven’t provided any details on CRCs, so frame validity checking will be a challenge.
Thanks Mark.
bob
Does anyone know, or have a way to find out if this might be a valid SpiRIT frame?
------
000: ac 91 54 c8 a4 d1 13 58 ca 58 9a 7c 31 db 03 42 ..T....X.X.|1..B
010: 0d 5f 01 ec 58 38 8f b7 56 75 bc 92 91 04 fa 29 ._..X8..Vu.....)
020: 08 6a a1 e9 a6 c0 8a 25 a4 ec c4 b3 66 77 2a 18 .j.....%....fw*.
030: 78 e8 e9 e5 e8 a1 ba fa de e2 7f 3f 01 6a f3 ba x..........?.j..
040: cd dc 3c e1 db f0 37 23 40 0b db 29 94 fb 07 27 ..<...7#@..)...'
050: e3 fa 5a a9 94 2a ef f0 83 9a 4f e3 66 2f 57 3e ..Z..*....O.f/W>
060: b4 91 f0 5a 78 da fe 32 72 42 57 f2 1a 10 a0 32 ...Zx..2rBW....2
070: 69 04 c2 06 8c 3b b1 c6 81 0d d3 71 eb e5 cf b5 i....;.....q....
080: 10 58 ca 35 37 c3 0c 89 ba 9e 74 0e dc e5 95 02 .X.57.....t.....
090: 55 5f 0b 76 5b 83 ee e3 59 d6 b1 e8 2f 8c 32 06 U_.v[...Y.../.2.
0a0: c9 12 cc 27 99 6d dd 68 b5 b8 f0 3c fa 71 db 06 ...'.m.h...<.q..
0b0: 51 31 6a 54 9a 8d 79 57 80 0b 10 f0 d2 2a d6 4a Q1jT..yW.....*.J
0c0: 08 8a 6e b0 59 2f 15 02 ba 8a 14 b7 55 f2 d1 44 ..n.Y/......U..D
0d0: 41 19 93 d7 9b c5 98 c1 ea 62 da 0b 94 00 f6 0a A........b......
0e0: e3 b4 c9 7f 6d b9 cc 98 d4 35 67 d0 5e eb 69 e9 ....m....5g.^.i.
------
It could be, the size are corresponding :
- Frame Length (excl. preamble and sync): 240 bytes
That said we don’t have decoder for it and it doesn’t seems to be AX25. But that said i think maybe the data needs to be PN9 decoded first.
Thanks for the reply.
While there is every chance that there’s room for improvement, here is a GNU Radio flowgraph that runs through the bitstream from:
-raw I/Q file
-wav AUDIO file
-ogg AUDIO file
-live AUDIO via UDP (such as from GQRX)
… and when the D391D391 syncword is found, outputs 240-byte frames. There was the one good frame (posted previously) from the I/Q shared by N6RFM and when I processed some of the .OGG files from strong-looking SatNogs observations, dozens (if not hundreds) of frames output. I was encouraged since these are not likely ‘trash’ decodes since the process depends on matching the syncword.
Please feel free to try it out and make improvements!
Very grateful to vk5qi for pointing out that the reason my Spirit output was just random HEX was because there was also a scrambler on the decoded output.
I’ve upgraded the GNU Radio flowgraph to include the additional step and as you can see below, the output now includes some readable ASCII
