3

(a) No amateur station shall transmit: <...> Music using a phone emission except as specifically provided elsewhere in this section; communications intended to facilitate a criminal act; messages encoded for the purpose of obscuring their meaning, except as otherwise provided herein; obscene or indecent words or language; or false or deceptive messages, signals or identification

Can *encoded* information be transmitted? For example: If you have a defensible reason for encoding (ie compression), does that count as message encoded "for the purpose of obscuring meaning" and is it allowed?

Does this change if you use a custom protocol to encode (not cryptographically secure but technically obfuscated)?

Scott Earle
  • 4,599
  • 1
  • 17
  • 30
belkarx
  • 133
  • 5

3 Answers3

6

97.113(a)(4) doesn't get involved here at all. Every digital message is "encoded", but that doesn't mean it's "encoded for the purpose of obscuring its meaning".

Instead, much hinges on 97.309(a)(4) and 97.309(b).

97.309(a)(4) says

An amateur station transmitting a RTTY or data emission using a digital code specified in this paragraph may use any technique whose technical characteristics have been documented publicly, such as CLOVER, G-TOR, or PacTOR, for the purpose of facilitating communications.

Which basically says that so long as what you're doing is decodable by anyone in the public, without secret algorithms, or secret keys, it's fine (that "a code specified in this paragraph" is a sticking point that everyone seems to ignore; it only includes Baudot, AMTOR, and ASCII, and therefore would seem to exclude nearly every modern HF digital mode that uses an 8-bit code or convolutional codes like FT8 on a silly technicality... but either my interpretation is wrong here, or, like I said, everyone including the FCC is just happy to ignore that detail). In any case base64 or any widely-known compression algorithm is certainly a "technique whose technical characteristics have been documented publicly".

97.309(b) says

Where authorized by §§97.305(c) and 97.307(f), a station may transmit a RTTY or data emission using an unspecified digital code, except to a station in a country with which the United States does not have an agreement permitting the code to be used. RTTY and data emissions using unspecified digital codes must not be transmitted for the purpose of obscuring the meaning of any communication. When deemed necessary by a Regional Director to assure compliance with the FCC Rules, a station must:
(1) Cease the transmission using the unspecified digital code;
(2) Restrict transmissions of any digital code to the extent instructed;
(3) Maintain a record, convertible to the original information, of all digital communications transmitted.

Which is saying that on some band segments (at a glance, it appears to be the data segments on the 6-meter and shorter bands), you can even transmit data using completely undocumented encoding schemes, unless you're doing it internationally to a country that doesn't approve of that, and unless you're doing actual encryption. However, if you do that and the FCC tells you to stop, you have to stop.

hobbs - KC2G
  • 12,322
  • 16
  • 31
  • To clarify, do you need explicit permission from the FCC to use undocumented encodings? – belkarx Mar 21 '22 at 04:09
  • 1
    @belkarx no, you only need to stop if requested, and do it in a segment that has the "a RTTY or data emission using an unspecified digital code may be transmitted" annotation in 97.307. And operate within your license class, of course. – hobbs - KC2G Mar 21 '22 at 04:12
  • 1
    (I'm guessing that they could also blanket ban everyone from doing it for a period of time or indefinitely without going to the trouble of amending part 97... nothing there says they have to write letters to individual hams, just that you have to stop when a director "deems it necessary".) – hobbs - KC2G Mar 21 '22 at 04:15
  • Does undocumented mean a preamble/signature is unnecessary? Encoding looks a lot like encryption if the algo isn't specified, so how is the conformation (or lack thereof) to FCC's guidelines even verified? – belkarx Mar 21 '22 at 04:28
  • 1
    @belkarx the old-fashioned way. Which is to say, mostly honor system, but if you try to skirt the rules and get caught, everything up to and including prison. – hobbs - KC2G Mar 21 '22 at 04:30
  • Ooh fun, makes sense, thanks. Are there precedents of people getting caught doing similar things? I know people have been jailed for harassment and illegal interference but I haven't seen anything beyond that – belkarx Mar 21 '22 at 04:31
1

Well, Morse is encoding. All the digital modes, too.

Most regulations on the legality of encoding Amateur radio messages are usually clear that the encoding cannot be secret.

The US regulations are interesting because they mention legalities hinging on the purpose of the encoding, which has led to no little confusion by US hams over the years. Taken at face value, I think the meaning is clear: you can't encode data in a secret manner just to obscure the intelligence carried with that data. (Thus, cryptographically strong encryption is right out, but convolution for the purpose of compression is ok. Similarly, by a strict reading of the regulations a ROT$x$ Caesar cipher would be contrary to the regulations if the protocol did not specify $x$ -- even though such a system would be trivial to break.)

So, for example, all those digital modes are published protocols such that anyone can monitor and decode them with the right setup. That is, there isn't a secret algorithm or a secret key necessary to monitor the communications.

Base64 would fall into this same category, as it is both a well-known published and open standard, and also the reason for its use is not to obscure, but rather "armour" text that is sent over some communications channel using some protocol.

Obviously, there isn't a single criterion that is used by regulators to determine the legality of a new mode or encoding, especially over time. I don't know what the exact process is (or even if there is one), but historically the encoding was often only used experimentally and then perhaps reported in club proceedings. There didn't have to be some official RFC (and I bet many encodings predate RFCs) or anything; though for the example of Base64 posed by the question at hand, this encoding does have an RFC anyone can point at. The regulations are clearly in place to provide enough interpretation that the intention for the use of the encoding is just as important as the technical aspects.

  • Conclusion being, a custom protocol is not permitted? – belkarx Mar 21 '22 at 02:22
  • 2
    @belkarx I'd say that you can do a custom protocol, but you need to publish the details of how it works and how to use it in some easily-found place online or in print. – nerdenator Mar 21 '22 at 02:36
  • 1
    @belkarx it's a good question and may already be answered here. I don't actually know how something like FT8 goes from a bright idea by a Nobel prize winning physicist who wants to solve a specific radio comms problem with a new encoding, and that new encoding being considered acceptable by any regulations. Certainly step #1 is publishing the protocol and its encodings in some publicly accessible manner. –  Mar 21 '22 at 02:50
  • 3
    In some ways, hams were into "open source" before it was cool. –  Mar 21 '22 at 03:10
  • @jdv how would you refine the protocol if it was not acceptable by regulations? Probably you would keep some kind of experiment log... – user253751 Mar 21 '22 at 13:57
  • @user253751 as hinted at in the accepted Answer, all of this is on the honour system. So I imagine in the old days things were published in journals and then CQ magazine so people could knock ideas around and experiment. Refinement is really two things: getting to version 1, and then versioning afterwards. –  Mar 21 '22 at 14:02
  • Great. I'm looking forward to reading the details of the VARA protocols, and the AMBE encoding used in D-STAR, DMR, Fusion, etc. I wish the open publication requirement was true - it would be a blow for freedom. – Deepstop Mar 24 '22 at 13:12
  • Yes the D-STAR protocol was published by JARL, but it is not completely open as the standard refers to a vocoding method only available to public in integrated circuits provided by a single company, although a software implementation is sometimes licenced to equipment manufacturers. The details of AMBE are not disclosed. Similarly, VARA is available for a fee, but the protocol and encoding methods are not published anywhere. So publication does not appear to matter to the authorities, I suppose because the decoding software and/or hardware are widely available. – Deepstop Mar 26 '22 at 13:04
  • @jdv It might be unclear to you, but it isn't to me. While you claim D-STAR is not open, it is not completely so. – Deepstop Mar 27 '22 at 13:37
  • @jdv. Despite your italics and ranting, you contradict your previous comments. Perhaps you are trying to troll me. I don't know. I have made no assertions about the FCC or regulations other than they don't take publication into account, and you have no convinced me otherwise. 73 my friend. I hope you can keep your blood pressure under control. – Deepstop Mar 28 '22 at 14:37
  • @Deepstop the question wasn't about those modes you mention that would strictly appear to run afoul of the regs. There is other Q&A about that you are free to contribute to. To me it is obvious that the tech used in radio has grown organically, and this process is bound to encourage inconsistencies. To be clear: I did not say or try to imply that "official" publications were a single factor for "acceptance". I suggested, for the protocol in question, that there is an RFC for Base64 and that its use isn't primarily to "obscure". Taken together this would probably satisfy the spirit of the law. –  Mar 28 '22 at 18:01
  • @Deepstop, I will point out that some regulations actually do specifically call out a mode being made "public" as one of the preconditions for its acceptance. If the first legal-ish test is that a mode cannot be "secret" then making the details public passes that first test. Since the interpretation of "public" is one of those things you know when you see it, an example might be an "RFC". Or published in CQ magazine. Again, it is an interesting question how commercialized modes can be used, and this is discussed in another Q&A, but not (as far as I know) fully answered. –  Mar 28 '22 at 18:09
1

WSPR and FT8 (and other similar digital modes) use a similar (but more complicated) arithmetic bit packing as Base64. But FT8 is a widely published encoding (online pdf from a major university, and in ARRL QEX back issues), so any message contents are not obscured to anyone who has access to these public documents, or to commonly available software tools (wsjt-x, et.al.)

However, both WSPR and FT8 have recognizable signatures that aid in identifying which encoding is being used.

hotpaw2
  • 13,316
  • 6
  • 45
  • 76