27 March 2022
"TV DX" refers to the act of receiving television broadcasts from very far away transmitters that wouldn't be reachable during normal conditions. Usually, this is done when specific atmospheric conditions arise, allowing the radio waves to travel in a path they previously couldn't. This is also referred to as "anomalous propagation", and does more often happen for lower frequencies. However, as practically all of TV broadcasting transitioned into UHF (and some upper VHF), TV DXing became less common, although not yet impossible since certain propagation mechanisms still work for UHF and above.
Another factor that makes TV DXing much less common in today's era is the fact that the majority of broadcasting switched to digital modulation. Digital broadcasting is more efficient with its spectrum use compared to analog, which came with defined video and audio carriers with a much higher power concentration. While this is good for ordinary viewing, it results in DX conditions requiring a higher signal level received (though when it happens it does provide the signal in mostly full clarity).
In short, with analog "what you receive" is "what you get". Even a weak signal can result in a poor quality picture, while a high quality picture requires stronger signal. With digital, for the most part you decode nothing until your signal reaches a certain level, at which point you get a clearly legible image.
However, despite all this, due to a set of rather strange circumstances, a few days ago I managed to receive an analog UHF TV broadcast originating from a transmitter over 3500km away. So, let's pretend you didn't read the post title and ask the rhetorical question; How is this possible?
To understand the background behind this, we have to go back to the simpler time of 2019. It was then that the Parkes Observatory radio telescope detected its first SETI candidate signal for the Breakthrough Listen project - now known as the (in)famous "BLC-1" - at the frequency of 982 MHz.
As of now it is widely believed that the signal was of a less intriguing (human) source, however at the time many theories about its origin sprung up as they always do. One of the more grounded ones was from the longtime radio amateur Scott Tilley. He noticed that the frequency that BLC-1 was heard at was consistent with the downlink band used by the old Soviet "Molniya" satellite constellation, in the identically named orbit.
These were primarily communication satellites, used by the USSR to relay both military and civilian data for areas high up North where coverage from geostationary satellites was poor or outright non-existent. The Molniya programme was very successful, with 164 satellites being launched in total before the aging constellation was replaced by the new Meridian series.
The Molniya and Meridian satellites both use a type of transponder that listens to one section of the radio spectrum, and re-transmits everything it detects in a different section. VHF, UHF, and even higher bands were observed to be used by these satellites, however, as previously mentioned, of specific interest was the upper UHF transponder downlink known to have been used by Molniya, overlapping the same frequency where BLC-1 was detected.
And indeed, shortly after the BLC-1 detection was announced, Scott found out that this Molniya frequency is now still in use by the Meridian satellites.
One thing that helped with the (re)discovery of this transponder was a strong carrier emission around 992.45 MHz. Upon closer inspection with a higher gain antenna, this turned out to be the video carrier of an analog broadcast TV SECAM signal, even with an accompanying faint FM audio carrier 6.5 MHz higher, revealing that this particular transponder on the Meridian satellites was in fact relaying an entire analog broadcast TV channel.
Some more inspection and observations of the signal by Scott and others also helped with identifying the source as the Türkmen Owazy music TV station, broadcasting from Turkmenistan. As luck would have it, this nation that just happens to use the perfect frequency for its TV channel to be picked up by a legacy ex-Soviet satellite, also happens to still use the analog SECAM standard for broadcasting, a real blast from the past and an unexpected new opportunity for a satellite-aided analog TV DX in the 2020s.
I received my first signals from the Meridian satellites in early March 2022, after a series of particular events brought more attention to these Russian military satellites. Just like many others before, I observed that different signals can be heard on the transponder depending on where the satellite currently is located, relative to the Earth's surface.
In particular, the Turkmen Owazy TV signal would usually be very faint and barely above noise, however it would greatly go up in strength just when the satellite was about to set below the horizon of Turkmenistan, only to quickly and completely disappear soon after. This indeed is expected behavior from a TV tower being relayed by a satellite; most terrestrial broadcast transmitters have their antennas set up in such a way that they focus all of their radio emissions towards the horizon around them - in some cases only in particular directions but almost never above. This obviously is because any signal radiated by a TV or radio tower into the sky would simply be wasted power that could have otherwise been used to improve the coverage of the ground.
In the case of this TV tower, it essentially broadcasts the vast majority of its signal in a circle around it, just like what the beam from a lighthouse would trace. And since the Meridian satellites are not geostationary, a few times a day one of the active ones crosses this beam, allowing it to receive the signal from this tower at full power ("full" relative to what's possible that far in space). This alignment lasts only a couple minutes, either before the satellite sets below the TV tower's horizon or until it gets above the beam, but during that time any receiving station in the satellite's large footprint spanning a big portion of the Northern hemisphere can receive this relatively strong relayed TV signal.
When the relayed signal is received, the total free space path traveled by it on its way up to the satellite and back down to the ground is over 60 thousand kilometers, far longer than what the Earth's circumference is, making for a solid (albeit questionable legitimacy) TV DX contact. One might say that this isn't impressive at all, considering that this is routine intended operation for the countless TV satellites in geostationary orbit at a similar altitude, though in my opinion that lacks the certain "magic" that a retro analog TV broadcast from a totalitarian dictatorship accidentally relayed by a high-end Russian military communications satellite has.
Even when the satellite is aligned with the TV tower, the relayed signal strength is far from great. For example, even during its peak strength moment, I couldn't see even a tiny trace of the SECAM color signal carrier, though the FM audio became perfectly clear at times and the brightness (black and white) signal carrier got loud enough that a basic picture appeared in a real-time decoder I was using.
The image below shows an FFT/waterfall screenshot of the SECAM black and white video signal relayed by Meridian.
As described earlier, the "advantage" of analog over digital, at least in this case, is that even though the majority of the 6.5 MHz wide signal was still way below the noise floor for me, at least the brightness signal was well above it. If this was a digital broadcast, I would have needed the entire signal to be at least a few dB above noise, plus an entire DVB-T(2) channel multiplex would likely be too wide for the transponder anyway.
After sharing my reception of the faint image, @2019Casandro got in touch with me and developed a basic yet perfectly adequate decoder shortly after. Processing the same recording in his decoder showed very promising results, as his program was much more suited for very faint signal like this, so I tuned into the TV broadcast again on a different day to get an even better recording. After running that through the decoder, I finally ended up with the video some of you have probably seen, in which it is possible to make out clear features of the content. On that occasion I also recorded the FM audio signal alongside the video, so I could combine the two into a "proper" viewing experience (which you can access by clicking the link).
Note that the decoder currently completely omits any video sync, instead simply duplicating the view in both width and height. Not only does this reduce its complexity, but attempting to sync such a weak signal really only ended up detracting from it (as seen with the previous real-time decoding). Perhaps a reliable way to sync will be developed in the future, but for these short demonstration purposes I find the current method satisfactory. It's not like you're going to be watching a full day's worth of Turkmen TV programming through Meridian anyway (though if you are I fully encourage it).
The 990 MHz Meridian transponder isn't particularly strong, requiring a high-gain antenna for it to be copied successfully. I used a 2.5 meter offset dish with a LHCP helical feed tuned to 995 MHz (Meridian is RHCP, dish inverts polarization). Right after the feed was a 990 - 1000 MHz LNA, custom-made for me by teroz.cz. I already tried receiving this transponder about a year ago, but without this filtered LNA my receiver was overwhelmed by nearby cellular network signals. Speaking of which, I used a HackRF SDR as the receiver, though just an RTL-SDR V3 or an equivalent would suffice. In fact I used the Nooelec NESDR SMArt V4 alongside the HackRF during my latest reception attempt, with the HackRF recording video and the NESDR recording audio (easier for me than having HackRF record the full bandwidth on its own).
The feed arm of my dish is very high above the ground, forcing me to use several meters of coax between the LNA output and the SDR input, so to make sure all of it gets though I also use a cheap off the shelf satellite TV line amplifier that I modified to accept and pass through 5V bias-tee power. This will most likely not be necessary if you use short coax and/or a higher gain LNA (I ordered mine with a lower gain option on purpose since I knew I would pair it up with the line amp).
A smaller antenna would most likely also work to receive signals from this transponder. In fact, during his original discovery, Scott Tilley simply used a small omnidirectional RHCP helix aimed at the sky. While this won't be enough to realistically decode anything, it will be satisfactory for those who wish to simply monitor the overall activity of the transponder. Going a step higher and using a larger directional helix aimed at the satellite might be able to provide you with good results, although I personally haven't seen anyone copy any individual signals on the 990 MHz transponder with an antenna smaller than a 2m dish. If you do, please share your achievement and tag me!
As mentioned earlier, TV broadcasting from Turkmenistan is only one of the many things you can hear on this transponder. Scott Tilley has been detecting signals consistent with UHF radar emissions from the currently besieged country of Ukraine, and due to the uplink band used it is common to see other broadcast and cellular signals appearing every so often. The transponder also has its actual intended use by the Russian military which I will not speculate about here, though once again Scott Tilley has published an extensive analysis of just that if you are interested.
The 990 MHz transponder also isn't the only one used by the Meridian satellites. Depending on the individual spacecraft, it can use a combination of transponders with downlinks including 278 MHz, 484 MHz, 990 MHz, and higher frequencies in the C-band and the X-band. Particularly the 278 and 484 MHz ones might be of interest to you, as those can be received with a much more conveniently-sized antenna and very often carry some interesting traffic, including amateur radio repeaters or air traffic control. Just like the 990 MHz downlinks though, the 278 and 484 MHz ones can be easily overpowered by your local broadcast signals. While I managed to receive the 278 MHz one, I am out of luck when it comes to 484 MHz as a DVB-T2 channel multiplex decided to settle right on that frequency.
[2022-03-28] As of today, only Meridian 8 and Meridian 9 have an active 990 MHz transponder downlink. Meridian 10 launched just a few days ago and might soon also activate it.
The image below shows a screenshot of the VHF Meridian transponder downlink, with many short narrow-band signals as well as what appears to be a wider channel and an FM radio station.
If you do manage to consistently receive one of the transponders, no matter which one, it will enable you to basically listen to a part of the radio spectrum anywhere in the Northern hemisphere, as long as you time your listening sessions properly, keeping in mind the required satellite alignment I described using the Turkmen TV tower as an example. As far as terrestrial transmitters go, whichever ones are near the satellite's footprint will have a higher likelihood of being picked up by it and relayed. Of course there will be many exceptions, but it is a good general rule of thumb. Of course a ground-based transmitter has to be strong enough in order to be picked up by the satellite in the first place.
I hope I have managed to describe the process behind receiving signals from these rather unique satellites. I didn't just want to have a Tweet saying "I received x and y", which is why I wrote this rather extensive article about the admittedly not so impressive feat, in the hopes that I can at least use the opportunity to share more detailed information with you similarly to my previous content about receiving the Starlink beacons with an SDR, which had very good reception (pun intended).
I would also like to emphasize the fact that I wouldn't be able to do any of this without the hard work of people like Scott Tilley and Casandro and all the support from my followers. As always if you have any comments, questions, or suggestions, feel free to reach out to me via Twitter as I still haven't figured out how to run a proper website with a comment section.
Thanks for reading!