Ultraslow radio for decentralized global digital communication (2013) | Hacker News Reader

Ultraslow radio for decentralized global digital communication (2013)

mail-archive.com

112 pointsTomte43 comments

> Getting a balloon to 90km altitude is straightforward

No, it is not. The altitude record for a high-altitude balloon is 53 km, set in 2002. Most high-altitude balloons are limited to 37 km or so. 90km is nearly at the Karman Line.

He was probably thinking 100K ft

FT-8 is the current "state of the art" in long range QRP[1].

Sadly it's pretty much only for exchanging a signal report + callsign for contesting. You get 75 bits per 15 seconds.

The other tricky part is solar conditions change so different bands don't work during the day/night and local RF pollution can make it really difficult to receive anything in a dense metro area.

[1] https://en.wikipedia.org/wiki/WSJT_(amateur_radio_software)#...

I had the great fortune to take a class from Dr. Steve Franke. It was a class on radio communication circuits, and a big chunk of the class was devoted to talking about the different modes. I remember he got really animated talking about phase-shift keying. Dr. Franke is the F in FT-8.

That's very cool, PSK or QAM is pretty darn incredible. I highly recommend breaking open say Numpy or Matlab and working backward from quadrature(sin+cos) into basic 2-PSK, 4-PSK and N-QAM RF mixer modulation/demodulation.

The math behind it is mostly basic trig and really impressive considering it is the underpinning of all modern RF modulation(wifi is 2/4/16/64/256-QAM modulation per OFDM carrier for instance).

If you're a software person, reading how liquid DSP does these modes is very insightful https://github.com/jgaeddert/liquid-dsp/tree/master/src/mode...

That looks really cool. Most of work these days is upstream from TNCs/etc but def keep that in mind if I end up playing around at the modulation level.

GNU Radio too.

Go Fighting Illini!

Oh wow he went to UIUC. Another reason to be proud myself. :)

That's very interesting. I've gotten a bit into ham radio recently (just have my technician's & two little radios, nothing fancy) and by far the most interesting thing is long-distance communications.

Do you know anything about the reliability (e.x. 1000km transmissions between two rural areas are possible 5% of the time/50% of the time/90% of the time), as well as the kind of antenna you'd need for that? It'd be very cool if a portable setup was possible.

1000km may be stretching things but 200-300mi over say 40 meter band(that stays active day/night) is possible with something called an NVIS antenna[1] the downside is you probably still need ~100W to make it effective which makes portable operation pretty tricky.

If the portable HF thing is something that interests you there's something called Summits on the Air which involves hauling a radio up to nearby mountain tops and try to make as many contacts as possible. Even with tech license VHF and HF CW are still something you can do. Elecraft's KX line are really common for these(I own a KX3[3]) and really incredible pieces of kit. With a radio, ~30ft wire and a tree/pole you can definitely make 800mi+ contacts.

[1] https://en.wikipedia.org/wiki/Near_Vertical_Incidence_Skywav...

[2] http://www.sota.org.uk/

[3] http://www.elecraft.com/KX3/kx3.htm

1000km is achieved routinely with modest HF transceivers; 100W mobile units, for example. Band conditions change, however, so given the limited power an amateur license allows you operate on the bands that are working best at the moment; even 1500W won't work when a band is 'closed.' Fortunately amateurs have access to a lot of bands so some frequency will usually serve.

You can do a lot of optimization if you're working fixed sites. Directional gain and optimal take off angles for a given path can be targeted with excellent results. HF antennas are large. A simple 40 meter horizontal dipole (good night time antenna) is 20 meters long, and it needs to be raised to some altitude that achieves the 'bounce' you want, usually 20 meters or less. Smaller, much less efficient antennas exist.

I've had QSOs from my home in California with hams in Hawaii, Japan, Australia, England, and Poland, among others. It depends on a number of things, like the band you're on, the time of day, antenna, power, mode, etc.. I don't have a big antenna farm or 2KW linear amps, just 100 W radio over wire antennas. In fact, I've made long distance contacts in QRP mode, just 10W or less. It's a lot like fishing. You keep throwing your lure into the water until something takes it.

We’re on a solar cycle minima at the moment which slightly ruins ionospheric propagation on HF which is the least “line of sight” set of bands. However it’s on the up again so in a year or so expect 1000km regularly.

I regularly hear East European stations here in London, UK with a random length bit of wire strung off a home made 40m receiver and into a tree though. Noise here is terrible so I have had to use very heavy narrow bandwidth audio and IF filtering.

If you want long distance, be prepared to use very narrow bandwidth i.e. learn CW though (its not that hard although I have trouble copying some of the faster stations).

FT-8 is a game changer. I've never seen a mode take off like it as long as I've been a ham.

Yeah, while FT-8 is really well put together nothing about it that is really that esoteric.

I really wish things like Forward Error Correction and more complex modulation made it down to the VHF bands. The fact that the standard protocol there is 1200bps without FEC and basically stuck in the 80s is really a shame. (Yes I know DSTAR/DMR/C4FM exist but they're largely locked into specific mfgrs).

WWVB, operating from Fort Collins transmits its time code signal at a rate of 1 bit/second at 60 kHz using carrier power modulation. This translates to a wave length of 5 km yet your cheap "atomic clock" is able to detect and sync to the signal with a short folded quarter wave antenna that fits inside the clock or inside of your wrist watch. Coverage is over most of the continental US

The iPhone app "Clock Wave" is able to generate and transmit this signal from the speaker (!) in your iPhone, so you can accurately set radio-controlled clocks and watches if you cannot receive the real thing. It can simulate the five major time transmitters around the world.

I'm always amazed that it works at all.

This page has a good explanation of a similar application: http://www.jrcomputing.com.au/Set_Watch/Set_Watch_Auto.html

The app has the phone's audio output circuit send a modulated 20KHz tone to the headphones or speaker. This incidentally results in faint EM emission at the third harmonic (60KHz) which can be detected by the watch at short range.

You can call this the "Part 15 loophole" [1].

[1]: https://www.gpo.gov/fdsys/pkg/CFR-2016-title47-vol1/xml/CFR-...

Is it actually receiving WWVB, et al., or jusy synchronizing with, for example, an NTP server and then using that to generate an audio signal you can "manually" sync with?

I don't think the iPhone nor any Android phone have the necessary circuitry to receive the WWVB signal directly. THey do, however, keep time sync with NTP or some NTP-like service over the wireless network or wifi. That should keep it's knowledge of time accurate to a second or two.

It simply broadcasts the time on your iPhone (the source of which is usually time.apple.com) as a WWVB-compatible radio signal.

> Shortwave radio (1.5MHz-30MHz) reflects off the ionosphere and can be received intercontinentally. This includes a couple of ISM bands in which unlicensed operation is permitted internationally: 13553–13567 kHz (22 meters) and 26957–27283 kHz (11 meters) [...] QRPers consider anything below 1 watt as "extremely low power", or QRPp, and they consider a thousand miles per watt as a difficult benchmark to meet.

Unfortunately, the maximum legally allowed transmitter power is on the order of milliwatts for the 22m ISM band, and tens of microwatts for the 11m band. (At least in the US; see 47 CFR § 15.225, 15.227)

Given the range of these signals, does the US government have a say if you were to transmit at 1 watt from, say, South America?

That's usually the role of the ITU[1].

[1] https://en.wikipedia.org/wiki/International_Telecommunicatio...

Earth-mode is interesting. https://sites.google.com/site/sub9khz/earthmode with frequency being so low and the power needed it might not be practical. Those people seems to have tried to use utilities as antennas. The French around WWI apparently had some ground telegraph developed and produced about 10k units of it.

From the blog link:

> Foremost in sub-9kHz through the ground communications was John, G0AKN, who sadly is a silent key.

That's an endearing way of putting it. I like it.

I’d be curious to know in what kind of context such an all-encompassing line of reasoning occurred: is it in the context of prepping or somesuch? (Not that I look down on it because of that, much the contrary.)

Thinking of this kind of scenario is part of the reason I am so fascinating by SDRs capable of both reception and transmission: it seems to me that having one or two such devices linked to a ham-radio-grade amplifier and antenna every ten kilometers or so “makes sense just in case”.

The title of the mailing list is "Kragen thinking out loud --- half-baked ideas." It does seem to just be some guy's random thoughts on a variety of subjects over a period of ~12 years. Check out the archives: https://www.mail-archive.com/kragen-tol@canonical.org/mailli...

This whole article is very well written and explores a number of different ways to get decentralized global communication. It's a topic I've thought about myself for years, and I keep coming back to the conclusion that ham radio has to be the answer.

Also, it'd be interesting to read more writing on the topic. Does anyone here know of any other articles or anything?

It would be nice if there were a visualization of this. Perhaps watts on the x-axis, distance on the y-axis, and baud rate of specific technologies or wavelengths as dots connected by contour lines? Probably on a log/ log scale.

(Sorry I don't care enough to make it, but I promise to look if someone else does ;)

Is there an "internet" out there based on these sorts of communications?

To me it makes sense to tie these things together into a global network independent of the fiber/copper internet. But it never seems like anybody's actually done it.

Not really for a couple reasons:

1. No encryption on any of the ham bands. They don't allow commercial traffic so to enforce that all transmissions must be decodable.

2. Bandwidth is scarce. 40 meters, which is one of the few bands open day and night is only 300kHz wide(7Mhz-7.3Mhz). For a frame of reference 802.11 has channels that are 22Mhz wide, almost 2 orders of magnitude larger.

3. The internet is really good at being the internet. When you move to RF you impose all sorts of constraints(see mobile/satellite vs fiber) so unless you have a really strong use case(one:many, mobile or no infra) wires are going to beat out airwaves every time.

That said there's things like FaradayRF[1] that are working on building hardware/protocols that can solve the interesting use cases like delay tolerant networks and last mile delivery.

[1] https://faradayrf.com/

I've thought about this often enough you could almost call it an obsession.

So have I! I'm glad that there's a handful of people also obsessed about this. It just seems so important and surprising how few people talk about it.

What have you found? What do you think the most promising options are? If you (or anyone else reading this) would like to send me an email about this, I'd love to talk more about it, my email's in my profile.

Can it be used for decentralized global scale consensus ?

Can you share more about your interest in this? Eg just cryptocurrency or for something larger?

Just? Imagine a globally scalable cryptocurrency where consensus is not affected by network topology and forks, in fact, just can't occur.

Somebody would put a layer allowing forks on top of that anyway, for one reason or other. Physical layer has only so much impact on abstractions built on top of it.

(Also, please don't. It's enough that crypto starts to show up in global energy use statistics; we don't need for it to also crowd out the RF spectrum.)

With this kind of radio you don't need to rely on P2P peers only. It becomes impossible to hide information (the best chain in case of cryptocurrency). Also, crypto shouldn't be necessarily energy based, it can use proof of stake.

> With this kind of radio you don't need to rely on P2P peers only.

> It becomes impossible to hide information (the best chain in case of cryptocurrency).

That's only because of... the law (something crypto people usually want nothing to do with). The reason amateur transmissions are not encrypted isn't technical, it's that the laws regulating amateur radio service forbid it explicitly.

> Also, crypto shouldn't be necessarily energy based, it can use proof of stake.

I'll believe it when I see it. To the best of my knowledge, there isn't any working, at least minimally tested, and actually deployed implementation of PoS anywhere. Moreover, from what I read around the block, there's a good reason to suspect that you have to waste energy somewhere in the process to ensure trust in, and integrity of, the blockchain. Some people actually think it's a feature...

> The reason amateur transmissions are not encrypted isn't technical, it's that the laws regulating amateur radio service forbid it explicitly.

It shouldn't be encrypted. The only (and the most important) thing which needs to be transmitted globally is the hashes of the several last blocks. It is not possible to "decrypt", so from my understanding, it's not "encrypted" information.

> I'll believe it when I see it.

Bitshares, NXT, Reddcoin, to name a few.

good idea; nikki tesla says, high five!

Reply on news.ycombinator.com