Some days ago my friend cryptomaster sent me an interesting recording dated December 4, 2025 about a test of fresh communications equipment conducted in the Russian Federation. The signal was recorded at the frequency of 15014.0 KHz/USB kHz and its external modulation parameters are akin to NATO standards utilized for serial waveforms, namely the symbol rate is 2400 Baud and the transmission mode is PSK8 (Figure 1).
 |
| Fig. 1 - serial waevorms main parameters |
The analysis of the ACF is very interesting (Figure 2). I think that the real structure of the frame is 106.66 ms which - at the velocity of 2400 Bd - corresponds to a frame dimension of 256 tribit symbols (or 768 bits): a dimension rather usual in specified PSK8 waveforms. In my opinion the strong 1173.33 ms spikes (corresponding to 2816 tribit symbols) could be caused by the interleaver (or the scrambler) length: in this case 8448 bits. As you know, an interleaver (and a scrambler) is deterministic and periodic and due to that, it can introduce hidden periodic structures into the signal, even if the first data looks random. erstwhile you compute the ACF, those periodicities may show up as spikes at circumstantial lags. In this case 8448 is not a "casual" number but it's a rather structured 1 corresponding to 1056 bytes and it is precisely a multiple of the dimension of a single frame, i.e. 768 × 11 = 8448 (as you can see in Figure 2).
 |
| Fig. 2 - ACF analysis |
Something akin happens in MS-110A at low data rates (<1200 bps): in those cases, 1 group of 4 frames number 160 symbols (4×40) and it's just "in sync" with the scrambler dimension (160 symbols) causing the strong 66.67 ms ACF spikes alternatively tha the expected 16.67 ms [1].
About the interleaver, in the Russian MIL context the interleaver lengths are usually specified explicitly in bits or symbols (quantitatively), not categorized as “short,” “medium,” or “long” (qualitatively) like NATO documentation sometimes does, so if I'm right it could be a 8448-bit block interleaver, 1.17 s depth.
Now, why I talk about a Russian waveform? The analyzed signal is actually a section of a longer recording which besides includes any op-chats which are clearly in Russian language (Figure 3).
 |
Fig. 3 - spectrum of the full recording |
My friend cryptomaster kindly provided me with a transcript of the chat. I didn't translate it, as the possible usage of military jargon would have distorted the translation. The full transcript is available as a "transcript note" at the bottom of the post. Anyway, the .wav file you can download contains the full recording (op-chat + signal).
The analysis of a 8448-bit part of the PSK8 demodulated bitstream (only a conversion of phase positions), reshaped to a 768-bit period, doesn't aid to detect/identify any kind of structured framing. Although repeating blocks of data are easy identifiable, the "usual" structure of Known/UnKnown data as we are utilized to seeing in akin waveforms is missing (Figure 4). Given the "experimental" nature of the transmission, it's possible that the same structures are being sent... but this is just pure speculation.
A more in-depth survey of the code is surely necessary, as is the request for further catches of this waveform.
Comments are welcome.
 |
| Fig. 4 - 768-bit period |
https://disk.yandex.com/d/oRognr-CGqxbgg
[1] https://i56578-swl.blogspot.com/2021/06/188-110a-acf-lengths-and-interleaved.html
transcript note
- 121, парусу.
- Парус, я 121, вы принимали от меня телекод? Квитанцию отдал, но вы не приняли по всему. После этого свой телекод отправил, но не принял от вас квитанцию.
- Мы отправку делали, получили на неё квитанцию и ваше сообщение получили и тоже дали квитанцию. Приём.
- Принято, отправьте в нашу сторону ещё одно сообщение.
- отправляем.
(передача сообщения)
- 121 парусу.
- у меня нет приёма. Приёмник не работает.
- (непонятно) или нет приёма, всё-таки?
- "Бозон" работает, а "чинара", у меня приёмник не работает на неё. Дайте счёт до двадцати, я послушаю на приёмнике.
- Ну, канал, может быть, немного ушёл, но эти "ноги" снять сегодня не успеем, как поняли? Прём.
- Хорошо, да, ладно.
- Даю счёт (передаёт счёт до 20), приём.
- Парус, я принимал от вас на четыре бала, наверное, но с прерываниями.
- Может, на другой канал перейдём? Повыше? Приём.
- Давайте, перейдём на другой канал.