December 2018 - January 2019

-- How did Priyom.org come to be? What does it do?

What would become the priyom.org community first came together around the events of summer 2010 regarding the station known as the Buzzer, or by the erroneous callsign UVB-76. The lore goes that it is a shortwave station transmitting short buzzing tones, continuously, 24/7 on the same frequency at 4625 kHz, and that has been doing so for decades without interruption - except for very rare coded messages in a Russian voice. And just why anyone would go through the effort of broadcasting a buzzer on shortwave for whole decades, is the question - and you can imagine its appeal to mystery and creepy story enthusiasts.

In the summer of 2010, some activity was noticed that appeared to break with that smooth decades-old routine that the station was believed to follow. There were signal interruptions, frequent voice messages and other surprising transmissions; and thus interest in the mystery spiked and it became a topic of discussion on online forums. Now it is better established that the Buzzer is a Russian military channel marker, that got tangled in the reorganization of the Russian military at that time in 2010, and whose command network now carries frequent and regular messages (or maybe it is just monitored more thoroughly now).

The Buzzer can be heard quite easily at night in Europe - but that is for people who live there and own a shortwave radio, which would be quite a small fraction of the internet users reading about it on forums. Something defining for our community happened when someone set up the website uvb-76.net, taking the audio from 4625 kHz from their local radio receiver in Estonia, and re-streaming live it onto the internet - giving a way for the online masses to directly partake in the listening of this mystery for themselves. This website came with an embedded chat box, and a small community, including myself, found their way and gathered there.

In the following months, our interests expanded from the Buzzer to other mysterious shortwave stations and numbers stations in general. While looking at the existing available material about them, we realized that there was space for more modern ways for presenting and sharing that information, and we decided to found our own website: priyom.org. A bit later, due to technical issues with the original chat room that we gathered in, we founded and moved to out own chat channel named #priyom. And thus our community was born.

Among what we do is documenting the numerous different numbers stations - strictly speaking that usually means transmissions for intelligence purposes only, but we also track similar shortwave oddities that are part of military networks (like the Buzzer) or diplomatic ones.

One very important tool is our transmission schedule: numbers station transmissions take place on fixed times and frequencies that are set up and scheduled in advance with each recipient. Discovering them, keeping track of changes, and knowing where and when to tune is a must for proper monitoring. We put it all together in a single calendar view, that always shows what time the next numbers station transmission will happen. This way, we know when to be there to tune in, and it also makes it very easy for curious people who want to catch a numbers station just to see for themselves what it is like, without having to monitor static for hours to get lucky.

We transcribe and log transmissions. We study the data and try to make sense of how the different stations and transmissions relate to each other and how they should be sorted, organized and best presented to easily bring out relevant patterns and help further discoveries. We favor a synoptic approach trying to figure out the operations of agencies behind related stations as a whole, the characteristics of their transmission networks, and evolving trends, rather than just individual stations and transmissions.

We try to better understand messaging protocols, and in particular to reverse-engineer the proprietary digital modems used by some stations, and build decoders for them. By nature, contrary to intelligence transmissions, military and diplomatic ones involve numerous stations networked together: we research those networks, try to identify and locate each station, the frequencies and the schedules that they use when applicable, try to understand the architecture and different branches of the network, and so on.

Whichever kind of station, we try to determine the origin of the signal: country, city, down to the exact shortwave transmission site if possible. The location gives us clues about the organization operating the transmission, and vice versa. We usually do this using receivers in many different locations, working collaboratively, taking advantage of equipment or capabilities for techniques like triangulation or multilateration. But sometimes - rarely, a member gets to investigate this in person, by going up to the presumed transmission site to confirm.

Most of this stuff requires some tinkering with software tools or elements at some point, so this is something we do too. We rely a lot on online software-defined radio receivers, software decoders for digital modes, and so on. We contribute to these software projects, and sometimes take the lead on particular pieces that we need. Ultimately, our aim with providing technical solutions and fostering a vibrant, complete ecosystem, is to enable any newcomer from the web to tune in and listen to numbers stations in just one easy click in their web browser.


-- Why do you think it is important to track and record numbers stations?

I am not sure that any of us thinks it is absolutely important to track numbers stations. It is more like a hobby. I think that the importance we put on it is rather comparable to that of a hobbyist collector, who enjoys expanding their collection or making sure it is exhaustive. Some of us put importance on different stations, or even individual scheduled links, based on personal preference and interest.

As explained in our mission statement on our website, there is also the all relative importance that putting this information in the open is in the general public interest. I'm pretty sure that the monitoring we do is already done by state intelligence services of at least major powers, however they treat the information they gather from these broadcasts as state secrets. Alternatively, this information is gathered by unscrupulous for-profit players, and sometimes taken from open hobbyist sources with no regard for attribution or copyright, only to be resold at overpriced rates by companies posing with the - sometimes unwarranted - clout of professional intelligence circles, to whomever is willing to spend budget on it. This is somewhat similar to the military-industrial complex, in terms of how the money getting funneled through it can get disconnected from actual value.

Even without getting so ominous, the mere lack of open and reliable information leaves the door open for all kinds of disinformation, evidence-free eccentric or sensationalist theories, and what can be called "doom porn" from random enthusiasts hyping themselves up over possibilities like imminent nuclear WWIII. Sadly, these are also too often rehashed and peddled by shallow journalism, even from reputable outlets. In the end, the relative importance of our work is to put out there an open and reliable alternative to such things as disinformation, manipulation and corruption. But mostly, we just take our work as an enjoyable hobby :)

To come back to a more practical side, we do have some sense of urgency in our work in particular circumstances when we get a rare chance to observe, make a record of, and maybe figure out, some unusual transmission or pattern: for example, some stations that are only sporadically active, some rare transmission contents, or some end-of-year resets or vacations that help us better understand the underlying operations - and also, more often than you'd think, some revealing operation errors.

In the long run, the technical purpose of properly tracking and logging numbers stations is that gathering sufficient amounts of data is the first step for any further meaningful study. With enough data, patterns start to emerge, and you can sort and classify relevant logs together, start understanding the formats and meaning of each field, reverse-engineer the message encoding (especially for digital stations) or even in rare cases take a stab at the encryption.

As for recordings, their importance is two-fold. First, radio and numbers stations are primely an audio experience, so you pass by a big part of it if you focus only on text transcripts. In the case of numbers stations, the emotional appeal and creepy atmosphere largely go through audio listening - and recording. Even in priyom.org's relatively short lifetime in shortwave monitoring, there have been numbers stations that went extinct, whose voices can't be ever heard on the air anymore; and recordings are primordial from a conservationist point of view.

Second, recordings contain the whole, raw information if you ever need to go back and re-analyze transmissions. This is especially important for digital stations when you manage a breakthrough in reverse-engineering the modem, now you can go back and decode or re-decode all your previous recordings. And even in voice transmissions, a text transcript typically doesn't carry certain subtle clues like alterations in pitch or spacing, or transmitter artifacts, that can be important to go back on and check to confirm a theory coming up later.


-- What has the outside response to the organization been like? (either on the site or on social media, etc.)

I would say that the response is generally favorable, although sometimes mixed. It seems that the people visiting our website find that it is a great and very informative resource - they tell us so. I have seen our work getting spontaneously linked as a reference by users we are not involved with. We have some solid Twitter following, and our blog posts occasionally get reposted by others as newsworthy. But of course this is all difficult to objectively appraise.

The way I would rather approach this question, is how we have managed to be part of a disruption in the field of shortwave monitoring and among other shortwave monitoring groups. One long-time monitor who witnessed the arrival of the priyom.org community felt pleased to see the new enthusiasm, and different feel and thought process that we brought to the hobby, to quote him. However disruption usually doesn't happen all smoothly.

There have been real feuds with other groups, but those were largely based on personal antagonisms between members. An example of friction would be about the numbers station calendar widget on our website, that we originally put together based on classic schedule charts from the ENIGMA 2000 group. At some point they noticed it and got a bit upset about this reuse of their material, but this was quickly worked out and resolved with prominent attribution notices putting credit where it was due.

Some other tensions came from bringing up new elements, new stations or discoveries about them, with an effect of challenging the status quo and consensus about how to view and refer to those stations. It seems to me that rather than anything like argued disagreement, the issues have been more rooted in vagueness and inconsistency in some areas of the hobby, themselves coming from lacking or fleeting interest.

The reference station designator system associates a letter with a number, to give a clear identification to a rather abstract numbers station that otherwise wouldn't have much going for it to give it a name: for example, E11, S06, M12, V15... It was devised by the ENIGMA group created in 1993. While some stations like the Buzzer initially received designators (S28 in this case) due to the way research historically unfolded, there was some uncertainty as to whether others should, considering that military stations like the Buzzer are not strictly within the scope of intelligence numbers stations. When priyom.org researched other and new Russian military channels, there was some disagreement and frustration on both parts about how we took the initiative to refer to these stations outside of the official designator system (managed by the ENIGMA 2000 successor group by then), superseding this system and straying from the unhelpful umbrella designator under which it would have been supposed to fall.

Another example is when we and others gave more attention to digital stations not adequately provided for by the classic designator scheme. I think that research and reports on this topic were met with a certain lack of interest and apathy, from the people who were de facto responsible for keeping the designator system up to date, again generating frustration.

I suppose that part of this matter is simply the benefit of hindsight, and looking at a system with a history behind it using new, fresh pairs of eyes: the quirks and rationally unsuited parts can be more evident or even glaring, and it is easier to put forward change. This is a good thing.

Now, I think that our place, work and input are acknowledged and welcomed by other communities. I believe that our numbers station schedule has become the reference, and we have taken the lead on digital stations with a new class of designators using the letter F, for FSK: F03, F06, F11... And some people are happy to work with us; but that still doesn't mean that everyone is ready to.

As I mentioned above, technical tinkering sometimes requires more direct involvement and cooperation with people managing other pieces of the ecosystem. Some people thrive on giving away their help in hopes that something good comes out of it, but some people have a more closed, protective and selective approach to sharing when you ask them. One example is when I wanted to make a shortwave receiver map with a wider focus than just select, narrow amateur or broadcast bands - as numbers stations do not constrain themselves to official allocations... When I set up rx.linkfanel.net, I had several alternative data sources to turn to; one of them warily refused - going as far in his reaction as adding technical measures to deter reuse of his data; and one of them cheerfully accepted.

Whether with our audience or other players in the community, I try to have a friendly approach and mind what they are looking for.


-- How would you described the community?

The priyom.org community is an online one. Conversations and exchanges are based on an IRC chat room, accessible from a web widget on the home page of our website to any visitor, so it makes us pretty open. There is no formal membership - just hang around and participate if you want. We welcome and have people from many parts of the world, although on average we are more centered around Europe.

I would say that we are diverse in our interests: we have people who come here for the intrigue and mysteries, we have electronics hobbyists, military buffs, licensed radio amateurs - some of which with tall antennas and very expensive gear, nostalgics of DX shortwave broadcasts, software techies, transmission protocol hackers, radio transmission professionals... We also have a lot of friendly "off-topic" conversations outside all of this.

I think that a deeper and important factor here is how the community originally came together around a popular culture phenomenon - the mystery around the Buzzer. Similarly to how CB radio was an element of popular culture in the 70s, this tends to drive whole different demographics from the more classic radio backgrounds centered mostly around technical interest or professional experience in communications and electronics.

And considering how radio is a medium whose peak of popularity lies in the past, I think that you would find that a striking difference is that priyom.org is a community of younger people. We have members of all ages, but a lot are under 30 years old, or even teenagers. You could say that the average age is maybe one whole generation younger than that of other typical radio groups.

As I mentioned above, the community originally gathered around a web chat for an online repeater stream of a radio signal. That would already be a strong filter resulting in people who are more comfortable with the internet than actual radios. Many of us are digital natives, and many particularities come from this.

Perhaps it is because we are so experienced with it from online exchanges, that we have internalized realities such as Sturgeon's law - 90% of all content is of poor quality, or the Pareto principle - 90% of content is produced by only 10% of participants - which also means that 90% of users are mostly passive consumers barely ever contributing a measly 10% of content.

Many radio communities have exclusive approaches and rules about membership, contributions and access to content, to try to ensure high quality standards for content and members, and fight the statistics above that they view as undesirable and parasitic profiteering; but to us, their struggle seems at best utterly vain (at worst conceited). We come from a world that sees efforts to prevent information or content sharing as not only pointless, but often misplaced and rather evil. We simply embrace the fact that many people will lurk our chat room without ever speaking, and many users will browse the information on our website without ever participating to the research or monitoring, and that's okay; so it's only natural for our community to be so open.

Another consequence of our digital approach to radio listening would relate to, and I quote our mission statement on our website, one of our goals "to drag the radio listening community online and into the 21st century (kicking and screaming if needed)". Behind this rather humorous statement is the idea that we make a big use of remote web receivers, and in particular SDR ones. The WebSDR of the University of Twente and the KiwiSDR network are great tools for us. If the classic approach is that, by necessity, each monitor finds interest foremost in the stations that they are able to receive locally, then with this, a single member can monitor stations in Europe, America and East Asia all alike, and can tackle the task of studying a worldwide embassy network all by themselves.

Digital receivers often feature a waterfall display, and in particular with a wideband SDR waterfall, it can make it easier to search for, spot, then properly tune to and understand signals of interest. Beyond this obvious matter, there is something more subtle coming out. When reading reports about signals, sometimes by the way they talk about it, I can really get a feel that the reporter caught this signal while dialing across the spectrum on an analog radio, rather than tuning up to it with a waterfall display. So I think there is something fundamental going on there with how technology conditions the very mindset with which you approach radio listening, and even more monitoring. And we promote awareness of these modern possibilities and wish for it to increase the levels of understanding and monitoring.

We have a vision of how shortwave monitoring could be made more efficient and automated. When a given station gets a short spike of popularity, it pains me to see receivers overcrowded with hundreds of people making a poor use of expensive and limited hardware demodulating resources by all redundantly tuning to the same frequencies at the same time. Instead, we offer web audio stream relays of popular stations like the Buzzer, that can easily handle all these connected users much more efficiently, without tuner waste.

We extract and also save the data from these channels, and glancing back over spectrum views reveals in just a fraction of time if and when there were transmissions, that can then be recovered and replayed. An adequate signal processing algorithm could even detect, cut out and upload message transmissions automatically. This is for fixed-frequency channels with round-the-clock activity, but it only takes plugging our numbers station schedule into just one more tool recording the right bits of spectrum, to have the same with them. Then theoretically, with good enough speech recognition software (or Morse or digital decoders), we could have all automated message transcription and logging. Eventually pouring these into a big data server cluster - the same kind that powers recommendation algorithms and self-driving cars - would open endless data analysis and prediction possibilities.

Of course this is far off and we have not reliably achieved half of that. But we can be proud of one of our members who put up an already impressive work at maintaining a custom WebSDR receiving setup, then implemented the channel extraction mentioned above on top of it, and even went as far as creating a radio "time machine", to extract and replay signals from any frequency from the past. I think that this is the same thing as agencies like the NSA do, archiving bulk spectrum recordings to come back later to check with hindsight for any transmission that they think might have happened or found a reason to become interested into; except that the way I imagine it, they record many bands of radio spectrum, not just shortwave, and use a whole network of receiving stations with worldwide coverage, and never ever delete the data; whereas due to disk space limitations, we can only recover past transmissions at most a couple days back from the present. Regarless, I find it pretty amazing that he built something replicating the same concept.

This too changes the way that we can view monitoring. This can speed up discovery and confirmation of new numbers station schedules immensely, and greatly improve opportunities as you can "rewind" a transmission caught in progress, recover duplex or more complex sets of transmissions after spotting just one piece, or investigate after vague signal reports by someone else. And this is the kind of things that our community strives to build on and beyond.


-- Why are you interested in numbers stations? How did you learn about them?

Speaking for the priyom.org community in general and historically, we were brought together around the Buzzer, and we expanded our interest to numbers stations in general because the Buzzer is often presented as a numbers stations, or among them, or at least as part of that world - and for good reason. News articles romanticizing the mystery tend not to make strong distinctions, and the existing literature and resources dealing with the Buzzer often do so as just one station among many which they cover, or at least they link to numbers stations as a related topic. So it was only natural for us to find about them when looking for information.

I think that many other people learned about numbers stations when coming across news articles about these mysterious shortwave signals, or YouTube videos such as "Top 10 most mysterious radio signals". These would quote priyom.org as a reference, leading viewers to the community. Occasionally, a popular YouTube channel will also make an episode on the topic, and this can drive a lot of curious people in. Some people will browse forum threads attempting to monitor on shortwave the beginning of WWIII and the end of the world, and priyom.org gets mentioned as a useful resource in those.

One instrumental source of material is The Conet Project, a collection of numbers station recordings professionally published as a CD set. This has been a vector for numbers stations to be sampled in the mainstream culture by artists looking for their exotic sound. Several music bands have sampled extracts in songs. The science fiction thriller movie Vanilla Sky makes use of them. This is a way that the curious fans reading trivia about these works can be led to the world of numbers stations.

Other popular TV series like Lost or Fringe have used the concept of numbers stations as a plot device without necessarily incorporating real recordings, but considering that the fabric of these shows runs on strings of fantastical weird phenomena, I would be unsure how many viewers would understand that numbers stations are actually a real thing, to be discovered outside the show.

Personally I guess numbers stations found me more than I found them; although my grandfather was a Morse radio operator for Air France, and hobbyist shortwave listener, so maybe I was predisposed. I was unable to closely follow the priyom.org community between 2010 and 2014, from before it even got that name, so coming back to it is what introduced me to numbers stations as they were monitoring by then.

And I think that this is also a great vector. People come to priyom.org through various ways, because of the Buzzer, another station, WWIII or something else; but hanging around the chat room and the community exposes and introduces you to everything else. In particular, the numbers station schedule will advertise to you most of the currently active stations. And I believe that even among audiences who are already shortwave monitors, putting a given station in our schedule will bring and spread awareness of it, it will make people discover a station that they might not have paid attention to otherwise, and increase reporting and logs about it.

When looking at the ways through which people learn about numbers stations, one of the things that transpires is the appeal of the mystery around them. People are interested in numbers stations because they are mysterious, they are creepy and eerie oddities. People are interested in numbers stations because they offer a window into the spook world. There is something primal about being somewhat part of the secrecy, feeling in the know, being able to observe so tangibly actions that you would expect governments to keep hidden from the public view. And when you walk that line between knowing some about what remains a governmental mystery, that is also a field of interest for conspiracy theorists.

Numbers station monitors also walk that line between the known and the unknown. Some of them claim that if it was known for sure what numbers stations are, if their mystery was solved, they would stop being so interesting. I think that this is very revealing. As I mentioned earlier, there exists many eccentric theories about the purpose of the Buzzer and numbers stations, that I will not propagate here. These theories are often baseless, and not only lack any supporting evidence, but thrive blissfully ignoring contradictory evidence that proves them technically or factually impossible. Not only do they resist contradictory evidence against themselves, they proliferate on the grounds of the commonly accepted belief that we don't know what numbers stations really are - a belief so entrenched that it is embraced and relayed even within the monitoring community - that resists all the numerous evidence piled up from declassified state secret documents, witness accounts from former spies and legal cases brought against arrested agents, showing that we do know that numbers stations are indeed intelligence transmissions.

If the idea that numbers stations are a mystery is so powerful, perhaps it is because it allows them to be what you make of them, to be whatever theory you put into them. They are a tangible recipient for people's fantasies. To such a literal extent, that I have seen fan fiction personifying individual stations as its characters.

Just like the saying goes, "ignorance is bliss", I think that it is a valid approach if some people would rather enjoy numbers stations as a mystery to them; but there are also people who will choose to pursue the truth, no matter how unsatisfying it turns out. The KiwiSDR network introduced a TDoA multilateration feature, and it strikes me how many people seem to toy with it just to check that they find the correct transmitter location for known reference time signals. Technology is not meant to just stay on the safe side of what is already known! Radio amateur signals are already more interesting, since you never know whom you will come across, but they are also very eager to give you their callsigns and QTH already. I find that the most interesting and satisfying transmissions to research are those that you know little about, such as numbers stations.

Working out numbers stations can be a great and interesting technical exercise. But even just observing the technology and design that they rely on is fascinating. One-way voice links combined with encryption using one-time pads are extremely simple and require very little skill or equipment, yet they are very robust and provide unmatched secrecy and security for field agents. Shortwave propagation enabling long-range transmission provides for a powerful vessel to reach foreign agents. It is a wonder. Personally, I have an interest in languages and network protocols, so I am also drawn to analyzing the strategies and metadata that they use to address and deliver messages under the constraints of one-way communications and unreliable shortwave propagation.

In the end, monitoring numbers stations is also just a fun hobby. Each of them is unique in some way or flavor, and you just gotta catch 'em all!


-- Which stations would you identify to be particularly active and/or popular with listeners?

This is a good question, because that largely varies on your criteria for "activity", and maybe behind that, what you are looking for in a numbers station. For example, if you look at air time, the most active station would be HM01, from Cuba: it transmits basically all day every day. However, transmissions repeat in a loop the same set of six messages all day long, and messages are not necessarily renewed often either, so the real amount of traffic isn't that big. But its near constant presence on the air makes it an easy station to catch.

Most stations use a more classic sparse scheduling, as in they turn up for a few minutes, sending a message, then disappear. Among these, the Polish station E11 would be the one with the most activity: about 40 different twice-weekly schedules known. Presumably, each of them is unrelated and targeted at a separate recipient.

Another very active station is S06s, transmitting from Ukraine, with again several different scheduled links on each weekday. S06s's traffic really stands out, in that it always sends new but very short messages. Moreover, the way that these are coded or encrypted is unusual and reuses lots of contents. So what could be another metric for activity - the amount of data carried over all the messages - baffles us with this one.

Activity does not necessarily correlate with popularity. I would say that the most popular in the community are voice stations based in Europe. Not all numbers stations carry messages as spoken audio, some use Morse code and others use digital modes. As I said before, it is very much an audio experience, and it seems that hearing an ethereal and nonsensical voice coming out of the static plays a big part in the appeal and enjoyment drawing people in.

And eerie voices are not the only element possible with the analog audio medium that creates and gives a popular character to a station. Many past, now extinct stations, but also still today the South Korean V24 for example, have used varied musical pieces and songs as introduction before messages. The Egyptian E25 does too, and also features goofy strongly-accented voice samples, frequent audio glitches and sometimes even messages spoken by live operators - and indeed E25 is a favorite to some for this.

Morse code stations would perhaps be felt as the most characterless - at least to most listeners who don't practice Morse themselves. However the auditive approach strongly carries over to digital modes in my opinion. The most popular digital numbers stations would be the slow MFSK ones like XPA2: they sound good and somewhat melodic, but more than that, I think that people enjoy the feeling that they can hear, to an easy extent, the numbers mapping behind each tone. In contrast, fast (100 Bd or more) BFSK stations sound screeching and grating, and cannot seem to get a hold of listeners' interest - even though these too can be somewhat decoded by ear once you get to know them.

The situation with PSK modes is worse as they more or less sound like static, or a different and louder kind of static. You can easily spot on a waterfall display that there is something there, yet... it's not that listeners just don't care, nor only that they disdain it as noise, but sadly too often people don't even see it as one of a multitude of identifiable modes. To quote some official description from the ENIGMA 2000 group of the STANAG 4285 mode, that they call "the Jet", illustrating this mind frame, it "is not considered to be deliberate jamming, although a nuisance." [http://www.signalshed.com/docs/ENIGMA%202000%20Active%20Stations%20List%20V1.3.pdf#page=43]

In the end this kind of factors can drive a real disconnect between activity and popularity. For example, look at this breakdown of the number of active schedules among the different stations of a single Russian agency: http://priyom.org/media/184376/russian-6-schedule-stats-2017.png  F01 and F06 are digital "grating" BFSK stations. As I mentioned above, despite being the clear minority, voice stations based in Europe are nevertheless the most popular.

Another example of disconnect would be North Korea. They operate a numbers station, V15, that they reactivated in 2016 after 15 years of absence. It runs as thinly-veiled programs on their PBS Pyongyang Pansong national radio station. As such it is usually preceded and followed by colorful North Korean folk music, and pre-recordings are spoken with all the pompous style of North Korean news anchors. As we saw, that would already be a factor of popularity.

Popularity is also greatly affected by the aura carried by a station, as represented or misrepresented in the traditional or social media. So when the crisis about the North Korean nuclear program intensified between them and the United States during 2017, there were many - unwarranted - reports from enthusiasts associating V15 activity with missile tests, imminent nuclear WWIII and whatnot. The hype spiked.

Bear in mind that V15 follows known weekly schedules and sends a message once a week usually. Meanwhile, we at priyom.org were researching the North Korean diplomatic network, that uses the DPRK-ARQ digital modem, based on BFSK bursts and which can actually be commonly heard across the shortwave spectrum. We were surprised to find that this network carries a lot of traffic, with dozens of encrypted or encoded telegrams distributed across the embassy network every day. We published an exclusive full research dossier; but even then, it was never received with any interest comparable to that of V15. Despite the huge disconnect in terms of traffic, activity, and even relevance in trying to figure out the geopolitical crisis.

It is up to what place countries choose to give to shortwave communication channels (main, backup or none at all), compared to satellite for example, but military and diplomatic stations can be by their networked and official nature much more active than intelligence numbers stations. The Buzzer sees a lot of traffic - it is most active on Thursdays according to this chart: https://imgur.com/uxxbjoy  It is also still popular to newcomers drawn by its historical aura, and due to the geopolitical place that Russia has grown back. Conversely, the HFGCS network of the US Air Force which sends frequent EAM and Skyking voice messages relative to strategic (nuclear) commands, is both quite active and popular for similar reasons.

In the end, if I had to take away something from what makes a numbers station popular, it would be a reflection on our information society. There are theories that some (or even all) numbers stations are just dummy operations ran as psyops against competing states. When thinking of Russian intelligence in the current years, you would also think of disinformation and manipulation of public consciousness to undermine Western democracies. Yet, in an ironic way, the real impact that numbers stations make on us might not come from abroad: the real factor at work might be the way that our own media, social media, and our own folklore that we, amateurs create by ourselves around numbers stations, skew and warp what they really are.


-- You said you're involved with some technical research for the organization; could you provide a brief explanation of what that means and what you do?

One thing that we do is to reverse-engineer digital modems used by numbers stations and others. There are many layers to this, usually that starts with figuring out some block or packet structure around which the transmission is organized, then going through some integrity features like interleaving, checksums or error correction.

After isolating and recovering the payload, we can look at how the message itself is digitally encoded. Usually it comes down back to an equivalent using the same kind of familiar analog numbers as in sister voice stations ran by the same agency. For example, when studying the Russian intelligence station F06, we made a breakthrough when we found out that each digital piece of 10 bits (1024 possibilities) mapped directly to, and enclosed, a string of 3 classic decimal numbers (ranging from 000 to 999).

Numbers stations are simpler because transmission protocols are one-way and involve only one station on air and no feedback; but for two-way modems there are also flow control mechanisms to figure out. We've seen several diplomatic network modems where each side can switch to idle mode, data transmission, or even sometimes an interactive chat mode.

Something else that we do is studying and trying to determine, or at least guess, where the origin and recipients of signals are located. The whole point of numbers stations is that you cannot locate the recipient, but a rough geographical area helps to pick the best receivers to listen to it. It can also help to understand the role of the station. For one, if a station uses frequencies such that all its recipients are within its country of origin, then surely it must serve some domestic purpose, and must not be foreign intelligence.

The origin of the transmission is always easier to determine, using techniques, such as triangulation, older than radio itself. An origin country points to the government behind it. If the signals come in turn from many different capital cities across the world, then it could confirm that this a diplomatic network.

A more ambitious task is to understand which kind of metadata, which number fields in the headers, identifies a given schedule, link or recipient, while pinpointing down to a city the location of each of them; and exhaustively listing and matching them up in a neat table. This kind of data has been monitored for a long time through decades before priyom.org existed, but we are proud to have been able to bring substantial contributions.

Behind this, there is also documentary research work. Sometimes the data that you have now is not enough, or its history matters in some way, and you need to look at the past. Maybe to correlate it with some hypothesis that you have, or maybe because stations disappeared or do not operate in the same way anymore. Then you have to search through old logs, past archives of shortwave monitoring newsletters, and other resources, to try to find logs that are probably partial because the signals were less understood at the time, and any other information that can help you as evidence.

Something we do very little of is cryptographic analysis, because we don't have any member with that kind of expertise. We have had to research content encoding, because certain messages are actually transmitted in cleartext, just encoded in an unusual way. But in one case we have been able to break part of the encryption of certain Russian telegrams, only to make the puzzling discovery that they start with a timestamp of the telegram, repeated three times in succession. Numbers station monitoring never ceases to mystify us.


-- Lastly, because this is an Amateur Radio publication, I was wondering if you are involved in other Amateur Radio activity?

Some of us who are radio amateurs do enjoy QSOs and participate in contests or DX-peditions; but amateur radio is not within the scope of priyom.org, and as a whole we aren't involved with it so much as we sometimes overlap. Some of the most frequently asked questions that we get are about signals that really are radio amateurs, for example: "What is the spooky music on 7076 kHz?" Which is the popular 40m band frequency for the digital JT65 amateur mode. Or again: "I heard voices speaking under the Pip," which is a Russian military channel marker sitting on 3756 kHz at night, right in the 80m band.

A shortcoming of KiwiSDR receivers is that they only provide a small, limited number of simultaneous user listening slots, usually 4 or 8 depending on the configuration; so sometimes it can get frustrating that there aren't enough slots for everyone to tune in, and you get curious what other people are using their slot for. We do see that amateur radio is somewhat popular among these online listeners, with some user slots tuned and listening to amateur bands for hours on - and to commercial broadcast bands as well. Also, a popular practice by KiwiSDR operators is to run continuous WSPR sessions, sometimes even on several of the user slots; although to be correct, perhaps that one has more to do with assessing the performance of their receiving setup, than participating into the amateur network.

I, too, have happened to occasionally tune in and listen to some ragchew; after all, some of them could be considered radio oddities in themselves :) However, I think that what is fundamentally at play here, is that much of what amateur radio can manage - the ability to make contact and chat with random peers on the other side of the world, might have been an impressive feat decades ago, but is lost on digital natives, because this is already a staple of their online world that their generation takes for granted. They just don't see the point of going through convoluted, expensive setups and technical licensing just to talk with strangers, when they already do that every day on their handheld consumer devices. Of course there is always some arbitrary part in a hobby, and there is much more to operating a radio yourself and being part of a community, but even with that in mind the appeal might just not be there. I feel hardly appealed to having my own shortwave receiver myself!