Messages

91

Spy Numbers / Re: Cuban spy numbers

« on: February 01, 2022, 1434 UTC »

The "noise" is a digital mode called DIGTRX. It's sending files in encrypted text.

Small nit here.

The mode is called RDFT (Redundant Digital File Transfer), the most common software used by hobbyist to demodulate that mode is DIGTRX.  The RDFT mode was developed as a ham radio mode around 2001, and started being used by HM01 in about 2012.  Prior to HM01, the Cubans used the same RDFT (and a couple other modes, also) with HM01's predecessor signal, SK01, starting in about 2007.

There have been a couple of different beta and experimental softwares that could work with RDFT, but DIGTRX is by far the most common, and the only one that I know of that has a chance of running on a reasonably recent OS.

T!

92

Spy Numbers / MOVED: Don't know fershure

« on: February 01, 2022, 1412 UTC »

This topic has been moved to Utility.

https://www.hfunderground.com/board/index.php?topic=92277.0

93

Spy Numbers / Re: 10920U

« on: January 17, 2022, 1430 UTC »

YL reading numbers in SS  1215

Are you sure it was SS?  The reason I ask is that S06s was active on that frequency at that time and date.  This would be the 175 schedule, every Thursday with frequencies changing seasonally, in January, February, November and December, 1200 UTC on 12155 kHz USB and 1210 UTC on 10920 kHz USB.

https://priyom.org/media/122158/s06s.mp3

T!

94

HF Mystery Signals / Re: UNID 6985 (CF) 1310 UTC 09 JAN 2022

« on: January 16, 2022, 1447 UTC »

Update: 10 January 2022

A reliable source has indicated to me that the waveform is 4 x MFSK4.

Yes, its most common mode observed recently has been what looks like 4x4FSK or 4xMFSK4.  That is, 4 channels of MFSK, each channel with 4 shift tones.

However, it has also used 3xMFSK4, 4xMFSK3, and a couple of other formats.

One format had a total bandwidth of 2.4 kHz, and consisted of 1xMFSK3, 1xBFSK, and 1xMFSK6.  It is possible this was actually a 4xMFSK4, but channel 2 had only data hitting 2 MFSK tones and channel 3 and 4 were so close together that it looked like 1 MFSK6 instead of 2 MFSK3's.

The total bandwidths seen have been from under 2 kHz to over 12 kHz.  4 kHz and 10 kHz appear to be most common in recent use.  It is not uncommon to have up to 3 transmissions or possibly channels right next to each other.

T!

T!

95

Shortwave Broadcast / Re: UNID 9265 AM 0050 UTC 07 JAN 2022

« on: January 07, 2022, 1759 UTC »

I moved this post to the Shortwave Broadcast forum because this is a regular SW broadcast.

At various time WINB (and WRMI also) transmits Shortwave Radiogram.  These are transmissions in various digital modes, such as MFSK32 aand MFSK 64.  That is what you have in that image, and WINB was active 0030 to 0100 UTC with that program.  At 0100 they went to Overcomer Ministries and regular voice transmission.

T!

96

HF Mystery Signals / MOVED: UNID 9265 AM 0050 UTC 07 JAN 2022

« on: January 07, 2022, 1752 UTC »

This topic has been moved to Shortwave Broadcast.

https://www.hfunderground.com/board/index.php?topic=91291.0

97

HF Beacons / Re: Increased Beacon Activity Around 4.095 MHz??

« on: January 06, 2022, 1616 UTC »

TCO sends more than just TCO, it occasionally sends something else also.  I have recordings of it over a couple of days, but I have not yet played them back to look in detail.  Just not had time to look in detail at anything this week, real life getting in the way.

One thing it sent was "TE2=", another was "TPZL".  I have no idea what TE2= might be for, but could TPZL be for Puzzle?

T!

And for anyone that did not make the connection I was suggesting in the above post.  There is a Morse code pirate station that shows up periodically on frequencies such as 6933 kHz that is known as KPZL.  It shows up, transmits for a few days, and disappears, and it has done so for years now.

It is possible that this station, periodically sending TPZL, might be connected with KPZL.  The propagation and signal strength were similar to what was displayed by KPZL this year on New Years day.

T!

98

HF Beacons / Re: Increased Beacon Activity Around 4.095 MHz??

« on: January 05, 2022, 0129 UTC »

Yes, I think you a right -- activity around 4095 +/- has increased. 

This past week I have noted 2 new (not in the list) slow dashers on 4093.12 and 4114.5 kHz. In fact logged them both on the KPH SDR today around 0100Z.
Around S1- to S3.

And now just in the past hour, there is a new very strong signal on 4095.99 - sending "TCO" in very slow CW. S9+ on the KFS and KPH SDRs. Just checked Token's SDR (at 0240Z) where its also S7-8. Big signal.

STENDEC

TCO sends more than just TCO, it occasionally sends something else also.  I have recordings of it over a couple of days, but I have not yet played them back to look in detail.  Just not had time to look in detail at anything this week, real life getting in the way.

One thing it sent was "TE2=", another was "TPZL".  I have no idea what TE2= might be for, but could TPZL be for Puzzle?

T!

99

HF Mystery Signals / Re: UNID 6988 USB 2223 UTC 09 DEC 2021

« on: December 31, 2021, 1459 UTC »

So far I have seen individual waveforms of approximately  2, 3, 4, and 10 kHz widths.  Each can be found at various speeds, from slow to fast.  The same data structure, down to the same MFSK step cycles, but scaled differently, can be seen on both the 4 and 10 KHz widths.  Different speeds can have the same MFSK step cycles, just slowed down or sped up.  I have seen every size alone, and every size with one, or more, companion signal(s).  I have seen up to three examples in close proximate, say within 50'ish kHz of each other.  I have also seen multiple sets of signals up, sometimes several MHz apart.  So far I have not observed more than three wide spaced sets up at any one time, for example a set at 10244 kHz, a set at 11448 kHz, and a set at 14892 kHz, all at the same time.

I do not have the skill or software tools to determine if each MFSK bit also has some other modulation on it, like possibly PSK data on each MFSK step.  But the MFSK steps cannot be the entire data content for the signal, as those go for extended periods without change.

T!

100

Equipment / Re: R-390a Users

« on: December 14, 2021, 2004 UTC »

I have been using the R-390a for over 40 years in my shack.  My shack is fairly extensive but I always seem to gravitate to using the R-390a for all types of listening. I have 3 of them including one at my cottage in the ultra-quiet listening environment along the shores of Lake Huron. I'm curious to know what other users think of them.

The R390a / URR is an undeniably good radio.  I have two, a low serial Collins and a late Stewart Warner.  The receiver noise is very good, very low noise.  The LO phase noise is very good.  And the pre-selecting tracking front end is outstanding.

But, I find I don't often use mine, other than just to use them because I specifically want to.  While good they are not better than several other radios I own, indeed not as good as several, and they do not bring to the table some of the features of later radios.  When I use it, I tend to set the R390a on a frequency or station and leave it there while I use other radios to actively do things.  I treat them very much like I treat other old school radios I really like, my Collins 51J-4, Hallicrafters SX-28 or the Hammarlund SP-600, for example.  I use them more for the nostalgia than any other reason.

For basic receiver capability I place several, many, of my other receivers ahead of the R390a.  I feel that my Flex-5000a, Yaesu Fdx-5000, Icom R8600, and Winradio G33DDC all beat the R390a in every imaginable way.  Several other radios I have here may fall behind the R390a in some way, but make up for it with the other features they bring to the game.  For example my RFSpace NetSDR may not equal the R390a in receiver noise floor or front end blocking, but it brings so much else to the fight that I will still take the NetSDR every day over the R390a.

Time marches on, and while old school top shelf radios like the R390a are fun to use, and take me back to the first time I spun the dial on such radios, there are better radios to be had.  Shoot, I still occasionally fire up the very first receiver I ever bought, a Hallicrafters SX-99 I have owned over 50 years.  And that really is not a good receiver, but the memories I get out of turning the knob on that (and the LM-18 frequency meter I use with it) make it an old friend.

So the radio you use, as long as it is adequate for the task, is much less important that how you enjoy using it.  But, if the radio you enjoy is something as good as the R390a, then so much the better.

T!

101

HF Mystery Signals / Re: UNID 6965 (CF) 2200 UTC 09 DEC 2021

« on: December 11, 2021, 1549 UTC »

Kind of puts the kabash on pirates operating on that part of the band. That is 30 kHz of valuable spectrum for what if you ask me, a crazy idea. Does anyone really believe that high-frequency trading (HFT) over an HF network has less latency than a solid fibre network connection? They are talking microseconds. Seriously.

Yes, absolutely HFT over HF will have less latency than a fiber connection.  HFT potentially leverages nanoseconds of difference in time, and the difference between fiber and radio can be much more than microseconds, it can be milliseconds.  If you ever look at the way HFT works you will find that they use automated software and short duration variations in stock value to "buy low and sell high".  Even a stock that over minutes of time is down, will have short periods when it is up, just a little bit.  They may buy 10 million dollars of a stock at say 6.29378, and 50 microseconds later the value is 6.30052, so they sell.  In that 50 microseconds they made $10,709, even if the total value of the stock is down drastically, say 6.27, 2 seconds later.  The guy who can see, and respond, to that short duration rise first is the guy who can make the most money off it.  Microseconds count.

On average, light in fiber travels about 31% slower than radio waves do, or, if you prefer, the Vp of average fiber is about 0.69 c.  Specifically designed high speed fiber will be faster, maybe up to 0.84 c.  But that is still slower than radio.

Not counting overhead latency, a 1500 km path over high speed fiber will take about 6 milliseconds, and over more average fiber it will take about 7 milliseconds.  Over radio this same propagation delay will be about 5 milliseconds.  Radio links will be 15%, or more, faster.  Now push that to coast-to-coast or transoceanic distances.
 When microseconds count, radio can be multiple milliseconds faster.

In the case of this signal, if you look at Part 5 experimental license WI2XNX and WJ2XXI you will find a close match for the frequencies used.  And the FSK emission description might fit also.  The 4-QAM description might fit other possible HFT signals we see on the bands.  https://fcc.report/company/10Band-LLC

Yes, the signal we are seeing in this case probably cannot be used for actual trading purposes.  Although there are several versions of the signal, and the speeds change, the data in it does not appears to change for long periods of time.  But maybe they are using it for experimental purposes, to build information or experience they need for something else.

This is what it looks like right now (1622z) here:


T!

102

HF Mystery Signals / Re: All 1 PPS pip / pulse / tick logs here please

« on: November 30, 2021, 0345 UTC »

Wcode chu mode (inverted) said this;
Year: 2021,   TAI-UTC: 37s,   DUT1: -0.1s,   no Leap Second,   DSTpatCode:  7
  Day: 330,   Hour:  2,   Minute: 31,   Second: 33
  Day: 330,   Hour:  2,   Minute: 31,   Second: 34
  Day: 330,   Hour:  2,   Minute: 31,   Second: 35
  Day: 330,   Hour:  2,   Minute: 31,   Second: 36
  Day: 330,   Hour:  2,   Minute: 31,   Second: 37
  Day: 330,   Hour:  2,   Minute: 31,   Second: 38

That roughly matches the day and time he reported.  Also, if I remember right, CHU transmits its data mode in seconds 31 to 39.  Listening to the MP3 file there are 9 bursts of data, your decode shows 5 of those 9, and match when CHU would have been sending data.

So, someone (I assume pirate base don freq choice) retransmitting CHU audio?  Or something else?

T!

103

HF Beacons / Re: 6698.98 KHz New Slow Dasher?

« on: November 28, 2021, 1733 UTC »

Discovered this while on the KFS Omi SDR. It has a roughly 30 second cycle: 27 seconds of silence and a 3 second dash.

This is a signal I call, in my Log Books, Long Dashes.  I have logs of this signal going back at least to 2013. 

6699 kHz is a common frequency for Long Dashes, however it shifts around.  If you look around you will find it on more than one frequency at a time, it is common for it to be up on 5 frequencies, one near 6500 or 6700 kHz, one near 8800 or 9000 kHz, one near 12000 kHz, one near 12500 kHz, and one near 16200 kHz.  It sequences these frequencies, high to low, hitting each one every 27 seconds.

This morning it started on 6699 kHz at 1420 UTC.  It was also on 8999, 11938, 12492, and 16192 at the same time.  As I write this (1725 UTC) it is still active on 6699, 8999, 11938, 12492, and 16192 kHz.  Prior to 1420 UTC it was on a different set of frequencies. 

In the current version of this signal (in the past it has had different pulse lengths and cycle times) it is a 2.9 second long pulse every 27 seconds.  The total cycle time is 27 seconds, not 30.  If you look closely at the pulses you will see the first ~200 msec of the pulse has a modulation on it.  Sometimes it has other features also.

I do not know the source, based on propagation I suspect China or the Pacific region.  For a while I thought western China, but now I am less sure.

I have done a couple of videos on this signal on my YouTube channel, here is one I did last year:
https://www.youtube.com/watch?v=vA9-F1-EQOg

I suspect it is a propagation tool of some kind, but more than that I have no clue.  It does not appear to run 24 hours a day.  It is wide ranging, I hear it on EU remotes, Asian/Pacific remotes, and here at the house all at the same time.  Naturally not all frequencies may be heard at the same time in all locations, but generally, if the signal is active, one or more can be heard anywhere.

Right now (1735z, EU after sundown) on EU remotes the 16192 signal has faded to gone in most places, 12492 is getting weak, 11938 is usable, 8999 and 6699 are strong.  Here at the house 6699 is just about gone and 8999 is weak, they were both strong here 2 hours ago.  11938 is usable, 12492 is not quite as strong, and 16192 I would not see if I did not know it was there.  This (both EU signal levels and mine) is consistent with Asian / Pacific propagation.

(edit)  Long Dashes left the 6699, 8999, 11938, 12492, 16192 kHz set of freqs at 1847z and moved to a new set at 1849z, the new frequencies are 6470, 8990, 11920, and 12494 kHz.  I assume the 16xxx kHz freq is either covered by another signal or propagation is keeping me from finding it.

T!

104

HF Mystery Signals / Re: All 1 PPS pip / pulse / tick logs here please

« on: November 27, 2021, 1455 UTC »

I'm not aware of any implementation of G-TOR that uses data packet bursts of well under 1 second, at 1 second intervals.

Although it was quite a while ago, I have seen it done (on an established G-TOR frequency) that way for test purposes.

However, G-TOR for this signal was a WAG on my part, totally going by ear and sight.

It looks much more like CHU's Bell 103-compatible 300-baud modem signal. SigIDWiki notes that "CHU is the only time signal station that uses this format", but since this almost certainly isn't a time station, maybe they're using it anyway? I have no idea what Windows-compatible software might be utilized to attempt a decode.

How often do you see this signal?  Since you posted, I have looked for it but not seen it.  Of course, I am a ways away from your location.

T!

105

HF Mystery Signals / Re: All 1 PPS pip / pulse / tick logs here please

« on: November 26, 2021, 1634 UTC »

Occasional HF monitor here. Have not been following threads regarding these type of signals lately, but I was hearing one earlier (02:30z on 26 Nov 2021) that alternated between 1 s pips and short (1 s interval) data bursts. The pattern appeared to match CHU to some degree, except (1) Wrong frequency; (2) No announcements; and (3) went off the air prior to 03:15z.

Any ideas about this - especially the data format? (I'm not yet up to speed on the Universal Radio Hacker software.) Refer to the following:

Waterfall: https://www.lutins.org/temp/sdr/6965/6965_waterfall.jpg
Audio: https://www.lutins.org/temp/sdr/6965/6965_audio.mp3
Spectrogram of pips: https://www.lutins.org/temp/sdr/6965/6965_pip_spectrogram.jpg
Spectrogram of data: https://www.lutins.org/temp/sdr/6965/6965_data_spectrogram.jpg

Thanks for any ideas!

Sounds and looks kind of like G-TOR to me.

T!