We seek to understand and document all radio transmissions, legal and otherwise, as part of the radio listening hobby. We do not encourage any radio operations contrary to regulations. Always consult with the appropriate authorities if you have questions concerning what is permissible in your locale.

Author Topic: MFSK Oddity, possible HFT, multiple frequencies and days  (Read 1654 times)

Offline Token

  • Global Moderator
  • DX Legend
  • *****
  • Posts: 2118
    • View Profile
All,

I have been meaning to write up something on the MFSK Oddity that has been being reported by multiple people to this forum over the past year+.  Unfortunately, real life has gotten in the way and I have not had time to do it.

So I am going to pull all (well, all I can find) the threads together under this one thread and include a couple recordings I have made.

(edit)  Well, that did not work.  I will link the other threads to this one, not merge the threads.

For those not familiar, the MFSK Oddity is a signal that has been showing up since at least mid 2020, and I suspect it was around a bit before that.  Further, I think some of the phantom "1 PPS" signals before that time may be related.

Some listeners tie this signal to HFT (High Frequency Trading).  I do not think that has been proven, but geolocation via TDOA and HFDF, searches of related frequency authorizations for HFT license, and similar frequency use / habits of other HFT tied signals make the HFT association very possible, if not probable.  If not quite proven it is a best guess.  10Band LLC and related businesses are likely candidates.

The arguments against HFT are that the data rate is too slow to be useful for that service, and the data also appears static, unchanging, for very long periods of time.  If you are leveraging RF vs fiber to gain microseconds of advantage, why use a data format that sends bits in 10's or 100's of millisecond intervals?

I have measured bandwidths of about 2.6, 4, and 10 kHz, I have seen other bandwidths but they have been too weak for me to get a decent estimate of the occupied bandwidth.

I have seen pulse rates of between 200 Hz (5 msec pulse interval) to 1 Hz (1 second pulse interval).  I have seen repeating "bit patterns" (for lack of better term) of 8, 16, and 32 bits.

The signal appears to be MFSK like.  Since late 2020 it has generally appeared to be 4 channels of data (prior to that it was 3 channels, or less), each channel with 4 FSK like steps.  There may be some other modulation on each pulse, but if so I have not been able to confirm it.  Arbitrarily, for my notes, I have called the lowest frequency channel of data Ch1, and moving up in freq Ch2, Ch3, and Ch4.  Within each channel I have called the tones, again low freq to high, 1, 2, 3, and 4.

These signals move frequencies regularly, probably as needed for propagation.

Some videos below, sorry, they are not nicely annotated or explained, but at least they are recordings.  The frequencies listed are center frequency.

4 kHz bandwidth, 4 Channel, 16 bit, 1000 msec interval, 15910 kHz
https://www.youtube.com/watch?v=wdzHIOUuJ-c


4 kHz bandwidth, 4 Channel, 16 bit, 100 msec interval, 14903 kHz
https://www.youtube.com/watch?v=GTYZrmO6rxE


4 kHz bandwidth, 4 Channel, 16 bit, 66.7 msec interval, 10160 kHz
https://www.youtube.com/watch?v=gKb2Z5LOdlE


4 kHz bandwidth, 4 Channel, 16 bit, 50 msec interval, 10170 kHz
https://www.youtube.com/watch?v=T18rzFeMG04


4 kHz bandwidth, 4 Channel, 16 bit, 33.3 msec interval, 13909 kHz
https://www.youtube.com/watch?v=sz2eRqMXVyk


4 kHz bandwidth, 4 Channel, 16 bit, 20 msec interval, 13928 kHz
https://www.youtube.com/watch?v=QM0o4jCjLHk


4 kHz bandwidth, 4 Channel, 16 bit, 10 msec interval, 10807 kHz
https://www.youtube.com/watch?v=_6Riik2BGJU


4 kHz bandwidth, 4 Channel, 32 bit, 100 msec interval, 15905 kHz
https://www.youtube.com/watch?v=7-Kqwf3PGFY


4 kHz bandwidth, 4 Channel, 32 bit, 5 msec interval, 14500 kHz
https://www.youtube.com/watch?v=HhwsXmk77uc


2.6 kHz bandwidth, 4 Channel, 3 step, 16 bit, 100 msec, 6993 and 6988 kHz
https://www.youtube.com/watch?v=V54EMvVuViI


2.6 kHz bandwidth, 4 Channel, 3 step, 16 bit, 50 msec interval, 14480 kHz
https://www.youtube.com/watch?v=oaoMYPot2ko


T!


Related threads:


UNID 9060 Digi
https://www.hfunderground.com/board/index.php/topic,93053.0.html

UNID 6940.0 kHz (CF) 00:00 UTC 01 FEBRUARY 2022
https://www.hfunderground.com/board/index.php/topic,92398.0.html

UNID 6985 (CF) 1310 UTC 09 JAN 2022
https://www.hfunderground.com/board/index.php/topic,91412.0.html

UNID 6988 USB 2223 UTC 09 DEC 2021
https://www.hfunderground.com/board/index.php/topic,90040.0.html

UNID 6965 (CF) 2200 UTC 09 DEC 2021
https://www.hfunderground.com/board/index.php/topic,90039.0.html

UNID 6970 2019 UTC 19 NOV 2021--see waterfall
https://www.hfunderground.com/board/index.php/topic,89284.0.html


« Last Edit: March 21, 2022, 1819 UTC by Token »
T!
Mojave Desert, California USA

Offline ~SIGINT~

  • DXing Phenomena
  • *******
  • Posts: 4103
  • N.E. Canada (FN08)
  • .ılılı..ılılı..ılılı..ılılı..ılılı..ılılı.
    • View Profile
    • Milspec Communication Canada
Re: MFSK Oddity, possible HFT, multiple frequencies and days
« Reply #1 on: November 11, 2022, 1505 UTC »
T!,

New related link for your collection from The SWLing Post:
High-Frequency Trading Service: FCC posts two lobbying documents by DPA Mac

Offline Nomadiq

  • Full Member
  • ***
  • Posts: 157
  • Boston, MA, USA
    • ICQ Messenger - 69
    • View Profile
    • HF Logs
    • Email
Re: MFSK Oddity, possible HFT, multiple frequencies and days
« Reply #2 on: November 14, 2022, 1448 UTC »
This is a great summary Token! The low traffic on these channels is a huge conundrum. It just about always seems to be idle. Just testing still?
QTH: Boston Northern Suburbs
Radios: Icom R75, Airspy HF+ Discovery, ICOM IC-7410, Yeasu FT-817
Antennas: Buddistick, EndFed Random Wire, Philips head screw driver jammed into the center socket