This question asks if proprietary encodings are "in keeping with the nature of" of ham radio. The accepted answer notes, at least in regard to FCC regulations, that if anyone can buy encoders and decoders (as hardware or software), then it's not considered encryption.

Is that a documented ruling by the FCC? (I don't doubt that it's true — I just wonder if they wrote it down.)

There is a difference between proprietary/patented/licensed and secret encodings. You can have something that is proprietary, patented and requires paid licenses, but the format is openly documented. E.g. video formats like H.264 and HEVC.

The FCC and international amateur radio regulations say that transmissions "shall not be encoded for the purpose of obscuring their meaning" (with an exception for satellite control).

My understanding is that the AMBE voice codec used by D-STAR, System Fusion, and DMR is secret.

If such an encoding is secret, then the purpose is very much to obscure the meaning. Specifically, the purpose of the obscuration is to prevent an independent implementation, requiring you to obtain a decoder from them or their licensees. Only they hold the key to unobscuring the transmission.

I can see how it would be legal to have to buy a license or a licensed device, but how is it legal to have a secret encoding that obscures the content of the transmission?

  • 7
    $\begingroup$ This is my first question here. Please be gentle. :-) $\endgroup$
    – Mark Adler
    Commented Jun 19 at 2:49
  • 1
    $\begingroup$ I think it will be hard to get a clear answer to your questions, which have been discussed many times. With respect to the US, FCC Part 97 prohibits "messages encoded for the purpose of obscuring their meaning". It may take courts to decide how this applies to codecs for which decoders are readily available. I think it would be hard to argue that the encoding is being done for the purpose of obscuring meaning. $\endgroup$
    – gschro
    Commented Jun 19 at 4:00
  • $\begingroup$ @gschro that's why I'm happy it's asked in "is it a documented ruling"; that's verifiable, objectively answerable :) $\endgroup$ Commented Jun 19 at 7:58
  • 2
    $\begingroup$ @MarkAdler by the way, welcome here! It's very nice to see you come here and ask questions on source coding :) $\endgroup$ Commented Jun 19 at 7:59
  • $\begingroup$ Somewhat related: ham.stackexchange.com/q/20669/26657 $\endgroup$
    – clvrmnky
    Commented Jun 19 at 12:59

3 Answers 3


My understanding is that the AMBE voice codec used by D-STAR, System Fusion, and DMR is secret.

That is mostly incorrect. The IMBE/AMBE+2 codec is standardized and documented as part of P25. The same codec is also used in DMR, System Fusion and NXDN.

It is an official ANSI standard, ANSI/TIA-102.BABA. Thus, it is similar to your examples of H.264. (Search for BABA).

The original full-rate IMBE codec was published in 1992. As it is 32 years old, the codec used in conventional P25, System Fusion Voice Wide, and NXDN 12.5 kHz should be patent-free.

The half-rate AMBE+2 codec, used in DMR, P25 Phase II, System Fusion Digital Narrow, and NXDN 6.25 kHz was published in 2009 and is believed to be patent encumbered until around 2028.

The outlier is D-STAR's 3600/2400 bps codec mode, which is incompatible with either P25 mode. I do not know if that one is officially documented anywhere, but the D-STAR spec was published in 2001, it should be off-patent now or soon.

  • $\begingroup$ Thanks! Being patent encumbered is not an issue in my question, and is orthogonal to its secrecy, which you have addressed. $\endgroup$
    – Mark Adler
    Commented Jun 19 at 20:36
  • 1
    $\begingroup$ Found this: "Also, I'm not convinced that the so-called "AMBE+2" is the same as the halfrate vocoder that is documented in the p25 specs (the former is apparently used in dmr, ysf, and nxdn, and p25p2)... There are "features" in AMBE+2 that are not publicly documented...", but that is hearsay, so I am taking your answer as definitive. $\endgroup$
    – Mark Adler
    Commented Jun 19 at 20:45
  • 2
    $\begingroup$ @MarkAdler It's the same as H.264, encoding is not unique, implementations can have better or worse quality while still conforming to a standard bitstream. Indeed, there were multiple IMBE and AMBE+2 full rate codec versions to solve issues with background noise and sirens in P25 public safety applications, however they were all backwards-compatible encoder side changes. Experimentally, MMDVM cross-mode can take the bitstream from one mode and stuff it in another and it works fine. $\endgroup$
    – user71659
    Commented Jun 19 at 20:50
  • 1
    $\begingroup$ Understood. If the publicly-documented decoder always works, then it's public in my book. $\endgroup$
    – Mark Adler
    Commented Jun 19 at 21:13
  • 1
    $\begingroup$ @MarcusMüller Correct, the spec only describes the IMBE/AMBE variants used in LMR, so the spec cannot be used for the codec modes in e.g. satphones. That's actually the whole quote that Mark Adler copied. It's similar to H.264 profiles: you give the DVSI chips a bitrate and feature vector. However, with the exception of D-Star, LMR uses the two fully documented P25 codec modes. Disclosure of what satphones and satellite radio uses is outside FCC ham regulations. $\endgroup$
    – user71659
    Commented Jun 20 at 22:56

First, thank you for asking this question, it's a fun one to explore!

Is that a documented ruling by the FCC?

I haven't found any documented ruling by the FCC on your question... but I believe your question includes some subtle mistakes in the facts available regarding 1) purpose vs effect, and 2) secret, both of which combined might explain why the FCC hasn't been moved to rule.

Think RTTY

Before software RTTY decoders were widely available it had the exact same effect though obscuring comms were clearly not the purpose or intent. You had to buy equipment or you were not able to decode. This is after WWII and before FCC explicitly contemplated RTTY (which was make/break on HF because FSK wasn't yet an approved mode)

USPTO has it, and so do you

I think you're mistaken regarding "is secret" : The algorithm is patented and filed with the USPTO, making it a matter of public record. It may not be commercially reproduced, but since private individuals can recreate it for their own use this seems to satisfy the spirit of the regs. There's even open source software available for decoding.

But what if?

Because I love hypotheticals in the law, take a quick trip with me.

Even if the encoding was still secret, an argument to be made might be as follows:

The relevant reg is §97.309(b) (RTTY and data emission codes). In part:

... a station may transmit a RTTY or data emission using an unspecified digital code ... data emissions using unspecified digital codes must not be transmitted for the purpose of obscuring the meaning of any communication... [emphasis added]

In your question you note that:

If such an encoding is secret, then the purpose is very much to obscure the meaning

I think you're conflating purpose with effect. The purpose of Icom et al in using AMBE vocoders is not to obscure, but rather to achieve high quality digital comms (whether they hit the mark or not is outside of the scope of the question and this answer). A side effect of this is that people without their equipment cannot decode.

Also, a proper legal analysis would probably note that it's not Icom's intent/purpose that matters in that FCC reg... it's the Amateur Operator's intent! I would argue that an operator, by using widely available commercial technology that may be obtained by anyone with or without a license, cannot be considered "intending to obscure." I would expect that a judge would be more likely to agree than not, especially in light of the apparent decline in the Chevron Deference principle in recent SCOTUS rulings.

Now, it's certainly possible that if the codec were secret that there might be some other FCC rule or broader USC reg that would control, but I define that as out of scope for my hypothetical :-)


I see where you went wrong. DMR, P25, P25-2 D-Star, Fusion, and NXDN FT8 are not secrets; they are licensed technology standards anyone with cash can use. You can buy a radio or applications with one or more digital protocols, or you can stay conventional and not be allowed to play digital radio. You have to pay to play.

Concealing a conversation is easy and legal if you follow the rules. Use one of the digital text modes, like AMTOR, DominoEX, Olivia, or MT-63, which no one uses. Few hams have P25 radios, so your conversations are private by default. OTOH folks like me, native Indians, can talk in our native language to prevent rubbernecking. CW conceals your discussion from the public and the vast majority of hams.

  • $\begingroup$ I don't think you read my question. It's not at all about having to pay or about licenses. It's about secret encodings. I did not think that having to pay to use the technology or get a license was in any way in conflict with the FCC regulations. However I thought, and to some extent still think, that secret technologies, i.e. formats that are not openly published, are in conflict. Your examples of non-English languages and CW are not secrets. They are openly published encodings. $\endgroup$
    – Mark Adler
    Commented Jun 20 at 19:46
  • $\begingroup$ By the way, the accepted answer notes that I was incorrect about DMR and System Fusion, as their audio codec formats are in fact openly published. However D-Star is not, so your answer is not entirely correct, at least as far as the participants in this discussion are aware. If you can link to better information, please do. $\endgroup$
    – Mark Adler
    Commented Jun 20 at 19:47
  • $\begingroup$ I said "to some extent above", since it is valid to question whether the regulation applies only to operators and not to manufacturers, since an operator of a D-Star radio is showing no intent of obscuring the message. Only the manufacturer is demonstrating that intent. $\endgroup$
    – Mark Adler
    Commented Jun 20 at 19:49
  • $\begingroup$ i'd love to see an example regulation that might apply to manufacturers in the amateur radio context @MarkAdler! $\endgroup$
    – webmarc
    Commented Jun 23 at 0:55
  • $\begingroup$ Like this? § 97.315 Certification of external RF power amplifiers. (a) Any external RF power amplifier (see § 2.815 of the FCC Rules) manufactured or imported for use at an amateur radio station must be certificated for use in the amateur service in accordance with subpart J of part 2 of the FCC Rules. No amplifier capable of operation below 144 MHz may be constructed or modified by a non-amateur service licensee without a grant of certification from the FCC. $\endgroup$
    – Mark Adler
    Commented Jun 23 at 1:53

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .