The United States Frequency Allocation Chart
beautifulpublicdata.com
beautifulpublicdata.com
One main reason is that allocations are accompanied by footnotes that provide further clarification for how each frequency band can or should be used. For instance, footnote US334 states that the in the bands 17.8 GHz to 20.2 GHz, fixed-satellite services transmitting from space to earth via geostationary and non-geostationary satellites may be authorized on a primary basis. For people/companies looking to provide services, this is the difference between being able to operate in a band and not.
Additionally, it’s simply much more practical to look through and filter a table of allocations. As an example, this very simple site allows for easy comparison’s between the allocation tables of 25+ countries: https://spectrumdatabase.io/. It’s not as pretty, but it’s more practical.
Either way, it’s great to see people interested in frequency allocations!
[0]: https://www.flickr.com/photos/llnl/9403051123
[1]: https://www.flickr.com/photos/acme-laboratories/31849902131
Not only it clips some colors for no reason (look at the blue background around "broadcasting radio" house -- the gradient transition is totally gone. I assume the editor just used lasso tool to select "white-ish" parts and replace them entirely with a single color, instead of using color level?), it didn't even correct the background to pure white (but some reddish color)? Ugh!
That said, much does still hold true. Longwave is still used by giant shore installations to communicate with ships. (Mostly by coincidence. Long wavelengths can reach submerged submarines.) The aeronautic band is still used by aeronautic non-directional beacons. The AM broadcast band is still AM broadcast. Some of the amateur bands are still in the same place today.
The physics and electronics depicted seem accurate. Its creators were clearly informed of the cutting edge in the 1940s; it depicts a magnetron, which had been invented about a decade before, and not widely used until high-power models were developed and then applied to radar c. 1940. A section about the health effects of ionizing radiation is conspicuously absent to my modern eye.
I scanned through it looking for any obvious errors or things we've changed our understanding of a lot since then.
There's one picture of using Radium radiation in hospital treatments that seems very old-fashioned.
"Variable speed of light" - https://en.wikipedia.org/wiki/Variable_speed_of_light
Here is a picture of the chart: http://canadianspectrumpolicyresearch.org/wp-content/uploads...
And here is the original document from Industry Canada, in PDF but with a much higher resolution: https://www.ic.gc.ca/eic/site/smt-gst.nsf/vwapj/2014_Canadia...
(This is the 2014 version so it might not be exactly accurate anymore, but I'm sure you can find more recent versions from Industry Canada)
Edit: here seems to be the latest (2018) version; a little bit less pretty at the top but still very cool: https://www.ic.gc.ca/eic/site/smt-gst.nsf/vwapj/2018_Canadia... (via https://ised-isde.canada.ca/site/spectrum-management-telecom...)
Also, does anyone know why 275 GHz is the cutoff?
Terahertz, here we come![1]
Incidentally, radio regulation, by international agreement, ends at 3 THz. Above that, it's called "light".
[1] https://docs.fcc.gov/public/attachments/DOC-356588A1.pdf
That's true, but it's also tautological.
If you gave an intelligent but blind alien the task of describing our international radio regulation they'd probably put the visible signal lights on airplanes and ships under the same umbrella as WiFi or FM radio.
In reality visible light is also highly regulated, and in much the same way as the radio spectrum.
The regulations I am aware of relate to safety (laser precautions or FAA requiring markers on tall structures, e.g.), public nuisance, or environmental protection (e.g. for protection of sea turtles or light pollution reduction).
Overall it's probably not as formal as the radio spectrum in being regulated by international convention, but US and EU regulation is a de-facto standard when it comes to things like any external lights on cars, ships, airplanes and other vehicles.
Just counting headlights, tail lights, break lights and indicators the average car has at least 10 different sources of light.
If you own a car that number of "visual radio" transmitters[3] on your car alone probably outnumbers that other transmission devices you might have, and certainly if we account for "signal strength" (your bluetooth headphones probably don't have a range of many miles, but a car will easily "transmit" light over that distance).
1. https://www.ysmarines.com/marine-lighting/cxd7-cxd7-b-type-m...
2. https://www.imo.org/en/OurWork/Safety/Pages/Preventing-Colli...
3. Radio just being another form of light, and (in this example) light being another form of radio.
None of the researchers were cognizant of the fact it was spewing noise outside its frequency allocation until some relevant government agencies quickly visited and had it shut down. We were impressed with the response time :)
Not sure if that's still done these days but I'm sure with GPS receivers and a couple mobile units, a position could be pinpointed very quickly.
There's a list of monitoring stations here: https://www.mdarc.org/regulations/fcc-monitoring-stations
(And yes, GPS-linked direction finding units are definitely a thing that can be purchased commercially. They're big in the amateur radio world for "fox hunts" - competitions to try and find hidden radio transmitters.)
https://www.sciencedirect.com/science/article/pii/S187705091...
"We only ask once"
Does the range change as frequencies increase (i.e. does the margin of error get smaller as you go up the spectrum)?
Some allocations are harmonized globally. Others are divided up into one of three global regions - roughly the Americas, Europe/Africa, and Asia/Oceania. Then these are further divided into one of 90 zones that roughly follow country-specific boundaries.
All of this is decided very bureaucratically. In some cases decisions are constrained by the laws of physics (HF signals are going to bounce across the globe), in other cases by practical necessity (aviation/maritime allocations are harmonized to facilitate global movement of vessels, satellites are going to orbit across the globe). However, in some cases allocations will differ between ITU regions or zones (especially local broadcast and land-mobile communications users).
Further, priority is usually given on a first-come/first-served basis, but harmonization is encouraged. Individual countries can modify their local allocations, provided this don't interfere with other global users.
You can view the global allocations by looking up the ITU Radio Regulations, Article 5. (PDF here: https://www.itu.int/pub/R-REG-RR-2020)
If you're curious about the process behind all this, look up the World Radiocommunication Conference, which is held every ~3 years: https://en.wikipedia.org/wiki/World_Radiocommunication_Confe...
Also — and this is sort of a hybrid case of "grandfathered in" and "practical necessity" — I presume that, despite the ITU making no ruling on such, the band centered around 2.45GHz has ended up being "commercial use" in pretty much every country? My thinking being that it is the lowest-cost-to-generate dielectric frequency for cooking food, so microwave ovens almost exclusively use that frequency; and microwave ovens are common in pretty much every country (or is there a country that outlaws microwave ovens?); so the local regulatory body of each country is inevitably stuck with that radio interference source on that band, mucking it up for any "dedicated, clear-channel" licensed use; and so they may as well give that band away to industry, to come up with noise-tolerant use-cases for it.
The ITU-R was founded in 1927 (it was known as the CCIR back then). At that point, we were mostly worried about telegraph operators using spark gap transmitters, which have been effectively illegal since 1934. There's not much to grandfather in.
> I presume that, despite the ITU making no ruling on such, the band centered around 2.45GHz has ended up being "commercial use" in pretty much every country?
In the case of 2.45 GHz (the ISM bands), that's done under Part 15 rules (in the US, similar rules elsewhere) that allows for low-power, unlicensed transmitters. Your WiFi signal is intentionally run at extremely low power so that it won't travel more than a few hundred feet before it's attenuated by the environment, and in that frequency range there's no risk of it bouncing off the ionosphere. Meanwhile, your microwave oven is shielded to keep any signal leakage to a minimum.
This falls under the "non-interference" provision I mentioned earlier - the signal is unlikely to leave your neighborhood, let alone leave your country, so the ITU doesn't care. They're more concerned with signals that travel hundreds or thousands of miles.
If you check the ITU allocation, 2.45 GHz is allocated to radioloation (radar), licensed mobile radios, and fixed broadcast users. Plus a slice for amateur radio just below it. ISM use isn't even mentioned at this level.
WiFi and microwave ovens are effectively secondary users here. They get to share any spectrum left over provided they don't interfere with the primary users - and can't complain if they receive interference from the primary users. (In other words: If the government complains that your microwave oven is interfering with their radar station, your microwave oven will lose. Guarenteed.)
I guess one question is: are there amateur/unlicensed uses that are starved of bandwidth (Wi-Fi does strike me as something that fits but is there anything else)?
WiFi 6e / WiFi 7 are getting a huge swath of bandwidth between 5.9 and 7.1Ghz in the US and Canada (but somewhat less in Europe/Japan/Australia/etc)
Even with current 5 Ghz WiFi, a lot of routers seem to not let you select the DFS/TPC channels above 48 even though they're perfectly legal to use in most places. As a result, those channels are far less congested.
The equipment itself also has false positives, which lead to much of it being unusable.
This has never noticeably happened in my experience (UK). I've used DFS bands on my routers for years. It's far worse to use non-DFS bands because if you live in a built-up area they are almost always congested, causing poor performance (latency spikes, packet loss, etc).
It's a huge performance penalty if some other nearby network is sharing your channel, so DFS is one small downside for a very big upside, IMO!
There's certainly some devices (cheap WiFi light bulbs, or anything really old) that don't support 5 Ghz at all, but any 5 Ghz device I've ever used found my DFS AP just fine.
5 Ghz WiFi bandwidth is really severely limited without DFS. If you're using an 80 Mhz channel then you're using almost the entire available non-DFS bandwidth, overlapping/conflicting with nearly all other nearby non-DFS networks. But with DFS there's room for 7 non-overlapping 80 Mhz channels!
I have laptop from 2013 that is 2.4 Ghz only for example.
This market-based analysis, which is typical, omits the factor of nearly complete regulatory capture, i.e. our reality.
> Coase anticipated that full, fair, well-informed evaluations would find that decentralized resource owners generally outperformed state diktat.
That omission is pervasive, as if "state diktat" is some force independent of these same resource owners.
Regulatory capture would prevent entrants from even having the option to bid for spectrum.
Your reversing cause and effect.
And it's not like every auction winner is guaranteed to make money, there are cases where they overbid and ended up losing money, to the benefit of the regulator, which is indirectly benefiting the taxpayer.
Which cause and effect am I reversing? Who deserves what is based on identity, i.e. citizen vs corporation, not money alone. So the auction determination of who deserves what, based only on money, is what is in some cases reversed.
>it's not like every auction winner is guaranteed to make money
Whether corporations made a sound business decision in the process of elbowing citizens out of their rightful resources while attempting to extract rents from them is beside the point. The damage to the People is done regardless.
This is an astonishing claim.
If you don't believe the law, judiciary, etc., decide, then your views are likely shared by almost no one.
Current adoption and technology has been shaped by precisely the corporate dominance of the spectrum I'm referring to. In light of that, there's a circular implication to this point you're making.
Let's imagine a future where mobile data is fueled by citizen-run mesh/P2P networks. Who truly knows what's possible (not just in tech terms, but adoption/behavior) until it happens, but we can be sure what won't be possible if we (the People) don't even have the legal access to experiment.
Intersecting with that issue -- it's not just about "starved for bandwidth". Different bands have distinct propagation characteristics that make them suitable for different purposes, so the variation is important too.
And legal access to experiment is already free and available to everybody. It’s called a Special Temporary Authorization. Corporations use them all of the time for various purposes, including experimental applications. Citizens are allowed to apply for them too, but nobody ever does.
I don’t really care about hypothetical future scenarios that will never exist. Unlicensed and amateur spectrum has its place, but it’s not exactly deserving of any more bandwidth than it already has.
Many of the very real, tangible rights and benefits we now enjoy started out as hypothetical future scenarios.
> Special Temporary Authorization > Citizens are allowed to apply for them too, but nobody ever does
It's easy, you just need to deal with the government and fill out paperwork for a permit and get it approved.
Maybe because again, the resources of these megacorporations and an individual in the general case are no comparison. Big companies have staff attorneys and compliance departments for whom the marginal cost for discovering these processes and doing paperwork like this is absolutely nothing relative to an individual. There are all kinds of things you can do with special permits and licenses and they are by that process guaranteed not to be popular with the general public, that goes for anything.
> mesh networks are actually dying off. Nobody cares.
It's niche right now because it's not easy for the average person to participate in and industry is deadset against it. The telecoms don't even want municipalities to have ISPs let alone individuals.
> not exactly deserving of any more bandwidth than it already has
What did these companies do to deserve almost all of our spectrum, capture our democratic processes with political bribes? You think Verizon is in the position they are because they "deserve" it?
They paid for that spectrum, and they fulfilled licensing commitments to built a network that hundreds of millions of people can access and use.
The FCC has special licensing terms and incentive programs for small businesses, non-profit organizations, and individuals. It actually is dead simple, and it’s cheap or free too. Their only ask is that is that the licensee serve the public interest and cover the population with services that they can actually use. This is the exact reason why mega corporations look like they’re preferred over the small players: the small players can’t fulfill that commitment.
The EBS spectrum, which is now used so profitably by Sprint/T-Mobile, was given away for free to academic institutions. They initially intended for it to be used for academic television, but most of those channels ended up not being used, so they gave them free reign over what to do with it. Unburdened from restrictions on how to use the spectrum, and what did they do with that freedom? They leased it to Sprint for the sweet sweet cash payout. As far as I can tell, there isn’t a single academic institution using that precious midband spectrum for a mesh network.
So again, the premise here is simple: all you have to do is build something that people use. If you can do that better than a mega corporation, and at low costs to the end users, you will almost certainly have the favor of the FCC. They’ll fucking give you spectrum for free. So just do it already.
It’s just a tradeoff where handing them more spectrum reduces infrastructure costs not really performance for most users.
Arguably given the above it would be useful to have a larger range of frequencies available to amateur/unregulated users, but it's also probably true that the general public gets more utility from more 4G LTE spectrum, if we're honest.
Hierarchy is something like:
1. US military/intelligence
2. Transnational corporations
3. Other large orgs/governments
(distant) 4. People
I was thinking of combining this with a 4G repeater so in theory you’d have phone data you can flip on and off with a switch.
Now that I look at this chart for a bit I realize its the radio equiv of the Vignelli map of NY subway. It looks nice, it feels like good design, but its actually unusable. The Y axis within each spectrum isn't used for any meaning other than just dividing into halves, thirds, quarters whatever. Sometimes there's two or more boxes of the same color and label in the same spectrum with no explanation about what distinguishes them. It color codes every minor distinction in application with its own color, which actually makes it less clear.
Here's my first pass at a redesign. All the categories are reallocated into a 2-feature category system, (Sea, Land, Air, Space) x (Science, Navigation, Communication, Broadcast). The first half-category goes on the y axis of each spectrum bar. So all the space stuff is along the top of the sub-spectrum chart, all the sea stuff along the bottom. The second half category goes into a color, of which there's only 4. The existing color stripe for federal/non-federal is good as is, but could perhaps also include operator type more generally (military/civilian/government/public, reserved etc).
I don't know if there's something like that for the United States.
Though ideally an interactive one, with zoom and some capacity for search and for personal annotation.
As a garage door opener owner/operator isn't licensed, it woudn't qualify as FIXED (though the transmitter wouldn't be fixed in all cases) and it certainly wouldn't qualify as "Federal-Exclusive."
(They're not in the ISM spectrum [0].)
[0]: https://en.wikipedia.org/wiki/ISM_radio_band#Frequency_alloc...
The idea is that by keeping the radiated field sufficiently weak, you avoid long-distance propagation and won't interfere with anything not in your immediate vicinity.
This is why your garage door opener only works when you're right next to your house. (Compare with a licensed mobile radio, which will easily transmit 100-300 miles given a clear line of sight.)
There's a third "level" (you can look it up on the FCC's website if you like) that includes things that have low priority (any of the items listed in the chart above supersede them) and are normally low enough power that they will not cause any interference to the primary or secondary frequency holders in the chart. This is why you don't see things like wifi or garage door openers in the chart, but they are part of the band plan, usually under "RF Devices" which falls under Part 15 of the FCC's Code of Regulations.
https://docdb.cept.org/document/593
[PDF] - https://docdb.cept.org/download/4162
[Excel] - https://docdb.cept.org/download/4089
[1] https://www.fcc.gov/engineering-technology/policy-and-rules-...
[2] https://www.ecfr.gov/current/title-47/chapter-I/subchapter-A...
Some of the amateur bands also overlap with allocations for things like radar. For example, a subset of the "5Ghz" WiFi channels (the "DFS" channels) can only be used if the access point will auto switch to a different channel upon detecting a radar pulse.
This is actually a chart that shows who is legally protected from interference. A license holder of a service listed in a colored bar has the right to complain if there is another emission on their spectrum. (There are also "primary" and "secondary" users for some ranges, the latter expected to yield to the former). According to that chart, a radiolocation service is the primary user. Amateurs can then work around them (1.5KW earth-moon-earth wifi anyone?) and finally wifi is allowed to exist with all the other random gadgets.
Yes, WiFi is given nearly the lowest priority for a small slice of the spectrum that is full of garbage from other users. No wonder it's so good at beamforming, spread spectrum use, channel hopping, and so on.
https://www.ntia.doc.gov/files/ntia/publications/fifth_annua...
What a time to be alive the early 00's were - where the possibilities of 3G was endless along with the profits. £22 billion for 3G on a tiny island was real money in 2000, and a lot more than the £2.3 billion or so they managed in 2014 for 4G and £1.3 billion in 2018 for 5G
then, small nails.