updated
Very interesting "mixed-mode" transmission cathced and sent me by by friend Michel (F1GOC)
 |
| Fig. 1 |
Segment 1 consists of an FSK modulation keyed at 200 Baud and 1000 Hz shift (Figure 2). The demodulated bitstream has a well defined period of 288 bit consisting of a data section preceded/followed by a sync/probe series (Figure 3).
 |
| Fig. 2 |
 |
| Fig. 3 |
Segment 2 consists of an MFSK-7 modulation keyed at 30 Baud and 400 Hz separation between the tones. The natural demodulation of the 7 tones (0-6;000-110) shows a repeated series of 940 bit dimension (Figures 4,5).
 |
| Fig. 4 |
 |
| Fig. 5 |
Segment 3 is the most interesting one. This section consists of 7 subcarriers which coincide with the tones of the erstwhile MFSK section (see Figure 1), the modulation utilized for each subcarrier - in my opinion - seems to be ASK2/OOK. The "aggregate" velocity of modulation is 200 Baud, i.e. 28.5 Baud per channel (Figure 6). The autocorrelation function of the signal shows a period of 10 seconds, corresponding to 2000 bit frames (Figure 7).
 |
| Fig. 6 |
 |
| Fig. 7 |
As a simple curiosity, the central frequency of the FSK section does not coincide with the value of the center band of the following 2 segments (see Figure 1). hard to state the user and the intent of the transmission, many "experimental" signals are in the air especially during this period, comments are welcome.
------- update --------------------------------------------------------
As my friend cryptomaster commented (and my friend Nicola besides in pvt), the FSK section is the well-known Russian Intel "F06" waveform. Indeed, after the change of the bit-order is possible to clearly see the typical F06 32-bit sync series 7D12B0E6 (Figure 8). The final part of the decoding, thanks to the rivet_b90 tool, is shown below (Figure 9).
And... yes, I'm a bit out of form since the 288-bit period should have suggested the solution 😁
 |
| Fig .8 - F06 sync sequence |
 |
| Fig. 9 - F06 decoding |
