unid 2400Bd/1200 MSK signals

 

Unid transmissions heard on 14669.0 KHz and 14693.0 KHz (USB) around 0800-0900 UTC. At first glance the signals may appear to be a PSK-4 modulation, as indicated by the fourth degree harmonics and 4-state constellation (figure 1-a,b) but actually it's the MSK Minimum-Shift Keying transmission mode (..even if it could be considered as a form of QPSK), indeed:

 

- the frequency spacing (1200 Hz) is numerically equal to half the manipulation speed (2400 Bd)
- there is no zero-crossing transitions in the 4-state phase-plane (figure 1-b)
- there is a long state staying in one frequency: if should it be PSK, the state should come back to carrier and stay there till a new phase change (figure 1-c)

Fig. 1 - a: speed and harmonics, b: constelation, c: phase-detector

The signal can be easily demodulated in SA using the MFSK level 2 demodulator, although it could be also demodulated as a differential mode by sampling data only from the code positions of the phases 90 and -90 degrees: the resulting bitstreams are the same (figure 2).

Fig. 2 - demodulation methods: MFSK-2 and PSK-2s

The signal has a 100 msec ACF that corresponds to a 240-bit length period of the demodulated bitstream (figure 3): each fame consists of the 47-bit length sequence
11100000101010111110000010101011111000001010101
which is probably used for synch, followed by 193-bit length data block. The synch pattern does not seem generated by an LFSR.

Fig. 3 - ACF and frames

 

The short duration of the transfers (<30 sec) did not allow me Direction Finding attempts, however it is possible to hypothesize a Chinese "source" given the remote KiwiSDR receivers which I used [1][2].

https://disk.yandex.com/d/m9vHldGREa0mxQ  (wav)
https://disk.yandex.com/d/jFVhT7cf-086kQ  (bin)

[1] https://khv.swl.su/  Khabarovsk, Russia
[2] https://nsk.swl.su/  Novosibirsk, Russia