Unknown signal on 19 meters

Status
Not open for further replies.

TVC15

Member
Joined
Jun 20, 2014
Messages
7
Location
Seattle, WA
Hello, I'm new to this forum but thought it was a good place to get some feedback on a strange shortwave signal I came across today on 15480kHz. Maybe 'over the horizon radar'? The bandwidth of this signal was pretty wide at 10kHz so it sounded about the same up on 15485kHz. I was tuning in via the Barcelona, Spain tuner over at globaltuners.com

You can find the signal over at my 'soundcloud' page:

https://soundcloud.com/brunoiswrong/oddity-on-15480khz-am


-Cheers, TVC15
 

Token

Member
Joined
Jun 18, 2010
Messages
2,379
Location
Mojave Desert, California, USA
Yes, this is almost certainly an OTHR. It would be very hard to narrow it down much closer than that, for example where it is from or who is using it. It is very possible that it is JORN, out of Australia, but that is a guess at best.

T!
 

TVC15

Member
Joined
Jun 20, 2014
Messages
7
Location
Seattle, WA
Thanks for the reply! I've never come across this type of signal before and did some research after posting this. There's a couple of similar signals on youtube described as OTHR. I didn't know this was still in use.
 

majoco

Stirrer
Joined
Dec 25, 2008
Messages
4,278
Location
New Zealand
I've had "ScanRec" listening on 15480kHz for a couple of days and absolutely nothing! They (whover 'they' are!) are probably using a variety of frequencies depending on propagation conditions so may never hear them again!

Listening to a remote receiver is always a gamble as to where the signal is coming from. It's not very often that I can hear a weak signal here in NZ simultaneously with the Dutch receiver....

http://websdr.ewi.utwente.nl:8901/
 

TVC15

Member
Joined
Jun 20, 2014
Messages
7
Location
Seattle, WA
Yeah, I continued to monitor the frequency about 10 minutes after the signal ended (my recording captures the last burst) but it did not return in that time frame. I feel a little lucky to have chanced across it.
Thanks for the link on the wideband web-sdr receiver in NL. I think this will be my favorite new listening post for a remote receiver. The waterfall makes finding broadcasts much easier and the sensitivity of this is amazing - I'm sure their antenna setup has everything to do with it.

-Cheers, Karl
 

Token

Member
Joined
Jun 18, 2010
Messages
2,379
Location
Mojave Desert, California, USA
Thanks for the reply! I've never come across this type of signal before and did some research after posting this. There's a couple of similar signals on youtube described as OTHR. I didn't know this was still in use.

Actually HF radar is more active today than at any time in the past. The US, Russia, China, Iran, Australia, the UK, Germany, Canada, and many others all have HF radar.

Keep in mind that all HF radar is not OTHR, and all OTHR is not HF radar (OTHR is also found at much higher frequencies). So you may hear radar signals on the HF bands that are not over the horizon type radars.

Also many radars on HF are not looking for man made things like airplanes or chips, some are looking for natural phenomena, such as wave action, ocean currents, or meteors. Correspondingly, not all radars heard on HF are related to military activities.

And it can be hard to draw a line between a sounder and a radar. They can use identical waveforms, and so it would not be wrong to call them both radars (the sounder is an atmospheric radar), but for my description here a sounder is looking at the atmospheric conditions, not for man made objects. And since these atmospheric conditions change slowly the sounder most often revisits (scans) slowly, maybe only twice an hour in some cases.

My YouTube channel has many examples of sounders and radars on it:
https://www.youtube.com/user/FirstToken

T!
 

TVC15

Member
Joined
Jun 20, 2014
Messages
7
Location
Seattle, WA
Most of the signals I've attributed to radar have a mechanical buzz, pulse, or sweep pattern but this one has many different waveforms during the bursts. It's a first for me.
Great collection of captures over on your youtube page. Keep listening! :)

-Cheers, K
 

Token

Member
Joined
Jun 18, 2010
Messages
2,379
Location
Mojave Desert, California, USA
I've had "ScanRec" listening on 15480kHz for a couple of days and absolutely nothing! They (whover 'they' are!) are probably using a variety of frequencies depending on propagation conditions so may never hear them again!


These types of radars typically are not "channelized". Meaning while they may be wide banded they do not have specific frequencies that they use over and over. It is very possible this radar is hitting around the same frequency each day at about the same time (driven by propagation conditions) but if you are tuned one freq you might miss it.

Most of these radars (when talking radar meant to track man made objects, not atmospheric conditions) have a sounder associated with them, or they draw data from a sounder network, to actively find the best frequency for the radar to use for the desired range and target set.


Most of the signals I've attributed to radar have a mechanical buzz, pulse, or sweep pattern but this one has many different waveforms during the bursts. It's a first for me.


This sounds pretty typical of some kinds of LFMCW radars.

It comes up with a pilot tone and then goes into a burst of LFMCW sweeps. The combination of frequency, width, and repetition rate of the sweeps determine what the target set the radar is looking for. It is optimized for targets of a certain size, speed, and maximum range.

The sweep repetition rate determines maximum unambiguous range (the maximum range at which you can be sure of reported target range, further detection ranges may be possible, but the range itself becomes unsure at that point). Of course propagation conditions must also support that range. But, each individual sweep rate has a (or multiple) "blind speed" (see "radar blind speed"), a radial velocity at which the target essentially appears not to move or be detectable, and thus can be processed out by algorithms trying to separate real moving targets from background clutter and noise (see "radar MTI"). By changing the sweep rates periodically you change the blind speeds, so a target not detectable (actually processed out) in one burst can be visible in the next.

That is what you have here, a radar changing its parameters during successive bursts to mitigate the deficiencies of each burst set. Of course, there are other ways to get around many of these shortcomings, such as integration over longer periods of time, this can allow narrower range and Doppler bins. Faster chirp rates, wider bandwidth, etc, all cause changes.

T!
 

ff-medic

Completely Banned for the Greater Good
Banned
Joined
Oct 12, 2007
Messages
728
Location
The Appalachians - Next to the tent and campfire.
I know it is 19 meters.....but from what Ilistened to, it sounds like a digital pager transmission.

If one has ever punched in a radio frequency for a alpha-numeric pager and listened ( for me a VHF freq ), that is exactly what it sounds like to me.

The radio transmittions is almost....if not continous in spacing.

I am not saying it is a paging system.....it may or may not be. I am just saying that with my previous monitoring of VHF paging frequencies ( I was bored, and working the "main gate" - so I experimented ) it sounds EXACTLY like a alpha-numeric pager broadcast.

FF - Medic !!!
 

Token

Member
Joined
Jun 18, 2010
Messages
2,379
Location
Mojave Desert, California, USA
I know it is 19 meters.....but from what Ilistened to, it sounds like a digital pager transmission.

If one has ever punched in a radio frequency for a alpha-numeric pager and listened ( for me a VHF freq ), that is exactly what it sounds like to me.

The radio transmittions is almost....if not continous in spacing.

I am not saying it is a paging system.....it may or may not be. I am just saying that with my previous monitoring of VHF paging frequencies ( I was bored, and working the "main gate" - so I experimented ) it sounds EXACTLY like a alpha-numeric pager broadcast.

FF, you got me thinking here, it does sound rather what I remember hearing from some paging systems. I am not much of a VHF/UHF monitor (not a lot around here) and I do tend to look at every signal from an HF angle first. So I took a fresh look at it (the recording) from the direction of a pager signal, assuming I had been biased in my first assessment.

The short answer:

I still say, with almost no doubt, this is an OTHR, I just can't ID which specific one (not unusual, you generally can't specify the majority of the HF radars seen other than maybe country of origin, although a few are easily identifiable). If the recording had been made in SSB mode I could be even more sure, by confirming a chirp rate, something a pager would not have. Since it appears to have been made in AM mode that is not possible to determine.

The longer answer (skip it if you don't want a minor tome):

Why not a pager?

First is frequency, I don't think I am aware of any pagers working below about 27 MHz on a regular basis. Of course, it is possible that the OP missed a digit and meant the RX was on 154.800 MHz. And it is also possible that the receiver used was experiencing imaging issues, and a VHF signal was being presented at HF frequencies.

The OP identified which remote he was using (Barcelona node on Globaltuners) so it was easy enough to jump on that node and tune up 154.800 MHz to see if any pagers were present. None heard in 45 minutes of monitoring that freq.

Image / overloading was still a possibility. Those kinds of issues tend to be persistent to a location and hardware, meaning if it happened once it is probably a common occurrence. So again on the Barcelona node I tuned to 15480 kHz. I checked it several times for up to 45 minutes per, no sign of a similar signal. Not proof, but possibly an indicator that it was not overload or imaging.

Next, most two tone pagers I am aware of have a small gap between the first tone and the second tone. This signal does not have such a gap.

Most two tone pagers I am aware of use higher tone frequencies. This transmission always starts (in the 14 complete examples presented in the recording) with a 100 Hz tone and that is followed immediately by second tone of around 30 Hz.

Looking closely at each burst it makes a lot of sense as a radar. Keep in mind we only have 14 complete bursts and 1 partial burst in the recording. But a cycle is seen.

The bursts occur on average slightly less than 5 seconds apart, start to start. This burst timing is variable with actual revisit times ranging between 2.75 seconds and 9.19 seconds.

Each burst starts with 20 fast chirps, at a 100 Hz rate. This means one chirp every 10 millisecond, or 200 millisecond of time. After the end of the 20th sweep the waveform transitions to a lower rate sweep, these lower rate chirps are always sent 64 times. This lower rate occurs on one of five fixed values. Arbitrarily assigning the number 1 to the first event heard on the recording, then number 1 cycle is a sweep every 31.2 millisecond, number 2 is a sweep every 29.9 millisecond, number 3 is a sweep every 32.8 millisecond, number 4 is a sweep every 28.6 millisecond, and number 5 is a sweep every 34.7 millisecond.

Throughout the recording the cycle stays the same, #1, #2, #3, #4, #5, repeat. Since each burst is well defined, 20 fast chirps always sent at 100 Hz followed by 64 slower chirps always sent at one of 5 fixed rates, the length of each of these 5 different cycles is always the same (cycles #1 are always the same length, cycles #2 a different length from #1, but all #2's are the same, etc). In the recording each is sent 3 times (the first number 1 is a partial, the recording starts after the burst has started).

This stepped and varying sweep rate is the FMCW radar version of a staggered PRF or staggered PRI in the pulse radar world. It has advantages in both interference reduction and in killing blind speeds.

The two tone transmission (or two sweep rate transmission) also has an advantage in the radar world. Higher repetition rates, like the first 200 msec of each of these bursts, results in higher average power returned from a target, and a higher probability of detection for smaller radar cross section targets. The faster chirp rate of these faster sweeps (assuming the same swept width for fast and slow sweeps) also increases range resolution (the ability to separate two targets close to each other in range). However such faster sweep rates with an unencoded transmission mathematically limits maximum unambiguous range. In this case, 100 Hz, it might make detection more probable, and smaller targets more likely to be detected, but for any target beyond 1500 km the range becomes uncertain. You can detect it, but you do not know, for sure, what the range is. For example a target detected as at 1000 km might indeed be at 1000 km, or it might be at some multiple of 1000 km, like 2000, 3000, or 4000 km. This is called an "nth time around target".

The addition of the approximate 30 Hz portion of the waveform pushes the unambiguous range out to roughly 5000 km. So in the ambiguous range 1000 km example above we could tell, for sure, that the target was instead at an unambiguous 3000 km (for example). The fast portion of the waveform increases detection probability and range resolution, the slower portion of the waveform increases unambiguous range while the varying sweep rates decrease susceptibility to noise and jamming while also eliminating radar blind speeds.

A nicely put together radar waveform.

T!
 
Last edited:

TVC15

Member
Joined
Jun 20, 2014
Messages
7
Location
Seattle, WA
Thanks for all the information The recording was made in AM mode. In retrospect I would have checked the sideband but didn't think of it at the time. Maybe next time. :)
 

Token

Member
Joined
Jun 18, 2010
Messages
2,379
Location
Mojave Desert, California, USA
When making recordings of unknown signals to try and ID it is best to use USB mode first with as wide a filter as you can get away with, I generally try to have about 3.5 kHz of audio passband for unknown signals. This will generally yield the best, most usable, results.

Why USB?

Most utility signals are in USB, but even an AM signal can be heard properly in SSB, so USB works for both AM and USB.

If USB yields no usable audio then try LSB.

Why a wide filter?

For most communications listening you want a filter just wide enough to allow the signal through, while clipping out unwanted traffic. But for an unknown signal you want to make sure to capture the entire width of the signal if possible. A common SSB filter width is between 2.4 and 2.8 kHz.

But if you used such a narrow filter on an unknown signal that might leave a significant identification trait outside the recorded audio bandpass. For example, with a Russian AT3X04D digital signal you would not capture the data stream plus the immediately identifiable 3300 Hz pilot (Doppler) tone.

In the case of this OTHR signal with a USB mode recording (the mode itself is less important than identifying it as definitely either USB or LSB) and about a 3.5 kHz bandpass a person can tell the chirp direction and chirp rate, both are identifying features.

T!
 

Token

Member
Joined
Jun 18, 2010
Messages
2,379
Location
Mojave Desert, California, USA
Token given the frequency of the bursts it sounds like a lot like superDARN...

SuperDARN has staggered PRI's (most often consisting of a repeating 7 pulse pattern) during each burst, this one does not have a stagger. SuperDARN does not use FMCW or IFMCW, while I can't be sure this one does because of the mode of the recording, it does sound to me like it is probably one of those modes.

T!
 
Status
Not open for further replies.
Top