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 permissable in your locale.

Author Topic: UNID 6970 2019 UTC 19 NOV 2021--see waterfall  (Read 240 times)

Offline IQ_imbalance

  • Sr. Member
  • ****
  • Posts: 282
  • Central MD
    • View Profile
UNID 6970 2019 UTC 19 NOV 2021--see waterfall
« on: November 19, 2021, 2023 UTC »
Any ideas...twice these signals appeared simultaneously....


LOG/NE-SW unterminated BOG/DJ-130/800Mhz Yagi
AFEDRI SDR-Net ICF-SC1 SDS-200 various RTL-SDR
Central MD

Online Token

  • Global Moderator
  • DX Legend
  • *****
  • Posts: 2019
    • View Profile
Re: UNID 6970 2019 UTC 19 NOV 2021--see waterfall
« Reply #1 on: November 21, 2021, 1415 UTC »
The signal you have here is what I call in my log books the "MFSK Oddity", I keep meaning to come up with a catchier name for it, but apparently I suffer a lack of imagination ;)  I am not sure when I first saw this signal, but it was well over a year ago.  I also suspect I saw earlier versions (not containing as many channels or as many pulses) of the same signal before that, probably back at least 2 or 3 years ago.

What it actually is I can't tell you, but I can tell you some of its habits and my observations.

Some people associate this with High Frequency Trading (HFT).  I do not know if that is a valid association or not, but TDOA generally places the source near Chicago IL.  So it is from the same area as known HFT signals, but it might be guilty by association.

The signal does not appear to contain changing data, at least I have not seen it in a mode that has changing data.  Possibly it occasionally does and I have just never caught it in a valid data mode.  Instead, the signal generally appears to be just sending the same pattern for sometimes hours at a time.  The patterns vary, it is not always the same pattern, but once established it generally does not change for extended periods.

Despite the fact I have not seen the signal carry changing data I will describe what I have seen in data terms, shift, channel, tones, etc.

I have seen the signal on up to 4 frequencies at one time.  Your image has what I would call 2 frequencies in my description, one centered on about 6972 kHz and the other centered on about 6989 kHz.  I normally log them by tuned frequency, in USB, with the signal width included in the logging, in this images case it would be 6968 kHz USB with 4 kHz width and 6988 kHz USB with 2 kHz width.  Note that the two frequencies in your image each contain 4 channels of similar rates and shifts, but with very different rates and shifts for each frequency.  The left one is in a ~4 kHz wide width and what I would call fast speed in my log, the right one in a ~2 kHz wide width and what I would call slow speed.  Another common width is ~10 kHz.  Each width can be found in different speeds, but each channel in a given signal has approximately the same speed.

I have seen this signal on many different frequencies, from below 7000 kHz, as you have here, to above 18000 kHz.  Sometimes it sets on a given frequency for hours at a time, maybe all day long, other times it jumps around a bit, with short periods on each frequency.  Generally, if there are multiple frequencies in use, the frequencies are fairly close together, generally less than 1 MHz between them and most often 10's of Hz, sometimes 2, 3, or 4 frequencies right next to each other.

I keep meaning to do a video on this signal for my YT channel, but real life has different plans.

T!
« Last Edit: November 21, 2021, 1626 UTC by Token »
T!
Mojave Desert, California USA

Offline IQ_imbalance

  • Sr. Member
  • ****
  • Posts: 282
  • Central MD
    • View Profile
Re: UNID 6970 2019 UTC 19 NOV 2021--see waterfall
« Reply #2 on: November 21, 2021, 1448 UTC »
Intriguing….were you receiving directly or via a remote SDR?  I’d think they’d want to have as directional a signal as possible, so it raises the question why either of us would pick up if they’re targeting Europe. I guess I might catch some of the NJ signal, and you the CA-ish signal….

I was wondering about data compression but the only thing they need to transmit is a time stamp, the ticker symbol, and the price info…?
LOG/NE-SW unterminated BOG/DJ-130/800Mhz Yagi
AFEDRI SDR-Net ICF-SC1 SDS-200 various RTL-SDR
Central MD

Online Token

  • Global Moderator
  • DX Legend
  • *****
  • Posts: 2019
    • View Profile
Re: UNID 6970 2019 UTC 19 NOV 2021--see waterfall
« Reply #3 on: November 21, 2021, 1622 UTC »
Intriguing….were you receiving directly or via a remote SDR?  I’d think they’d want to have as directional a signal as possible, so it raises the question why either of us would pick up if they’re targeting Europe. I guess I might catch some of the NJ signal, and you the CA-ish signal….

I was wondering about data compression but the only thing they need to transmit is a time stamp, the ticker symbol, and the price info…?

Direct.  I see it pretty much daily, almost any time I go looking for it and propagation conditions are not in the trash.  For example, as I type it I see it on 16000 kHz c/f, 15995 kHz USB, 10 kHz width.  I see it less often at night, but I think that might be conditions causing that, as I will occasionally find it at night but typically less strong.  Often I cannot find it on European remotes at all.

When I say no changing data I mean nothing changes in the data, it is the exact same pattern for hours at a time.  And often the data rate is slow enough that any time advantage that might be had for RF vs cable / fiber is lost in the slow data rate.

T!
« Last Edit: November 21, 2021, 1626 UTC by Token »
T!
Mojave Desert, California USA

Offline Traveling Wave

  • DX Legend
  • ******
  • Posts: 1365
  • Western N.Y.
  • Let Radio Furnish The Entertainment
    • View Profile
    • Email
Re: UNID 6970 2019 UTC 19 NOV 2021--see waterfall
« Reply #4 on: November 23, 2021, 2149 UTC »
2136 UTC - 23 Nov 2021, just noticed the "MFSK Oddity" on 6988 kHz USB, the thing is transmitting on four frequencies, looks to be 4Khz wide in total. At 2140 UTC it stopped ( deep fade?) and started back up at 2144 UTC, still going at 2149 UTC with S5 signal in WNY. It covers frequencies 6988 to 6991 kHz.
2152 UTC - Signal is fading in and out.
2210 UTC - Signal is now S9+5.
2235 UTC - Signal still S9+5.
« Last Edit: November 23, 2021, 2236 UTC by Traveling Wave »
Location: Western New York ( Niagara Frontier)
Radio :TS480 with RTL-SDR pan-adapter, HDSDR, Omni-Rig Control, Zenith T/O R600,T600
Antenna: 40m dipole, 20-17-15 meter fan dipole. One of N. Tesla's own prototypes. A high-powered mutant of some kind never even considered for mass production.
eQSL appreciated

Online Token

  • Global Moderator
  • DX Legend
  • *****
  • Posts: 2019
    • View Profile
Re: UNID 6970 2019 UTC 19 NOV 2021--see waterfall
« Reply #5 on: November 24, 2021, 1736 UTC »
2136 UTC - 23 Nov 2021, just noticed the "MFSK Oddity" on 6988 kHz USB, the thing is transmitting on four frequencies, looks to be 4Khz wide in total. At 2140 UTC it stopped ( deep fade?) and started back up at 2144 UTC, still going at 2149 UTC with S5 signal in WNY. It covers frequencies 6988 to 6991 kHz.
2152 UTC - Signal is fading in and out.
2210 UTC - Signal is now S9+5.
2235 UTC - Signal still S9+5.

In the next couple of days, with the long weekend, I will try to grab some images to add to this thread, to demonstrate what I am talking about below.

In the last couple years I have seen several different versions of this signal.  Without a screen shot or a recording I am guessing at which specific variant of the signal you are seeing in your description above.  So my below is a guess, based on your description, and one possible version of the signal that I see often.

The 4 kHz wide signal was a single transmission.  I have seen this signal on up to 4 different frequencies at one time using bandwidths of 2, 3, 4, and 10 kHz for each frequency.

All of these versions can be found in various speeds, from very slow to very fast.  Sometimes a very slow transmission will be a few kHz away from a very fast one.

In the 4 kHz wide version there appear to be 4 individual, 4FSK, channels.  Arbitrarily I have called the 4 “data” channels inside the signal CH1, CH2, CH3, and CH4, with 1 being the lowest frequency set and 4 being the highest frequency set.  Each channel has 4 shift tones.  Generally the signal cycles in a 16 bit cycle.  I mean each cycle has 16 pulses or combinations of shift tones, and then the cycle repeats, over and over.

One common rate is one set of shift tones every 100 msec.  Since there are 16 bits or sets of shift tones in a cycle that means the signal cycles and repeats every 1.6 seconds.  A faster version of this same signal, identical in every way except speed, sends a set of tones about every 10 seconds and completes a cycle in about 0.16 seconds.

All 4 data channels in a given transmission are in sync, and cycle together (pulses occurring simultaneously in CH1, CH2, CH3, and CH4) but the bit cycle is different for each channel.  The same sequence is not being sent in CH1 as in CH2, or CH3, or CH4.


T!
« Last Edit: November 24, 2021, 1758 UTC by Token »
T!
Mojave Desert, California USA

Offline Traveling Wave

  • DX Legend
  • ******
  • Posts: 1365
  • Western N.Y.
  • Let Radio Furnish The Entertainment
    • View Profile
    • Email
Re: UNID 6970 2019 UTC 19 NOV 2021--see waterfall
« Reply #6 on: November 24, 2021, 2119 UTC »
Token, it definitely was the 4 kHz wide version with the 4 individual, 4FSK, channels as you have described.
« Last Edit: November 24, 2021, 2122 UTC by Traveling Wave »
Location: Western New York ( Niagara Frontier)
Radio :TS480 with RTL-SDR pan-adapter, HDSDR, Omni-Rig Control, Zenith T/O R600,T600
Antenna: 40m dipole, 20-17-15 meter fan dipole. One of N. Tesla's own prototypes. A high-powered mutant of some kind never even considered for mass production.
eQSL appreciated