While M17 might sound like a new kind of automatic rifle (as actually, it is), we were referring to an open source project to create a ham radio transceiver. Instead of paraphrasing the project’s goals, we’ll simply quote them:
The goal here should be to kick the proprietary protocols off the airwaves, replace DMR, Fusion, D-Star, etc. To do that, it’s not just good enough to be open, it has to be legitimately competitive.
Like some other commercial protocols, M17 uses 4FSK along with error correction. The protocol allows for encryption, streaming, and the encoding of callsigns in messages. There are also provisions for framing IP packets to carry data. The protocol can handle voice and data in a point-to-point or broadcast topology.
On the hardware side, the TR-9 is a UHF handheld that can do FM voice or M17 with up to 3 watts out. The RF portion uses an ADF7021 chip which is specifically made to do 4FSK. There’s also an Arm CPU to handle the digital work.
We were struck by the similarity of the TR-9 to a cell phone since it has an LCD display, an SD card slot, and a 9DOF sensor. Maybe some open hardware cell phones and open hardware ham radios could find common ground.
This is quite ambitious, but generally, small ham rigs are having a resurgence. Having high-quality RF components available as chips makes a lot of difference.
How come no one has lost their amateur license for using proprietary protocols.
I’m not a ham, but I always thought part of the license requirement was no encryption and only public protocols ?
Going to guess it because the hardware is commercially available. If it was an unpublished, home brew setup, the story might have been different?
So what you’re saying is that maybe corporations can do things on the amateur bands that amateurs can’t.
I’ve wondered this, myself. I think it may come down to this: the protocols are not secret, but they are patented, so that other manufacturers can’t build them into their radios without a license. This would not preclude individuals from building them for themselves, though. Just a guess – I don’t know if the details of the modulation schemes are published or not.
Remember that for a while, way back when, superheterodyne radios were covered under patents. I don’t think this legally prevented amateur operators from building superhets for themselves, though.
Not really saying that corporations can do things on the amateur bands that amateurs can’t. My point was that if anybody can get a radio that can access a codec format/modulation format, it would be hard to apply the “encryption” label to that format. Maybe I am reading too much into it, but I think the intent of the prohibition on encryption is to prevent obscured communications.
I don’t think that the encryption prohibition was intended to enable other manufacturers to produce a product. If that were the case, using public key encryption on a digitized voice stream would satisfy the ability for other manufacturers to produce equipment and still allow obscured communications.
I think that all of the various digital modes are legal to use because their algorithms are publicly available.
It sounds like we are both pointing at the same conclusion from different directions.
Corporations can not operate on HAM bands accept in the case of a “shared” band. We share some bands with commercial and government entities. That has nothing to do with your original question though. In those cases one entity will have priority over the other. Hams are on the bottom and the Government is on top of the priority list. I don’t think there is any band all three of us share.
How about the County DAS system? I’m a Fire Alarm installer. I just got my Technician class license(KN4YOJ) in 2019. I was trying to educate myself of radio frequency because my company is now installing the DAS for the Fire Department to be able to use the 800Mhz frequency in buildings.
My company does not have a Ham radio license. Yet they can through and as a commercial buisness suddenly be able to install thus equipment and test it. I think this shouldn’t be.
ERRC or public safety das is not ham radio. The frequencies are not ham. The license the county has is a private one, not ham. Your designer and installers have to have a GROL commercial license, since it is a commercial system.
And again expensive toy I cannot afford. UHF toys and upper bands stuff like digital protocols will be expensive but then the hardware that commercialise it too, thats what I do not like about the hobby, commercial people taking over the amateur status of the hobby I do not like, It will kill the creativity in the hobby, and its still a hobby for the real hams, creating stuff on a low budget aint it, thats what I like about the Amateur status, its open, clean, and sharing, not over the top out of reach that is hardware driven.
In the US a DMR hand held is only about $45 on Amazon so a much cheaper option.
because of a pathetic exchange rate even that is elevated about 17 times for me. Not everyone else’s problem but not all HAMS originate from country’s that is “rich”, but still, it is a hobby and the “Amateur” status says it all
Have a look at DROIDSTAR. Its an app that you can use on digital voice modes including M17. *It is in Beta* and documentation is scant but shows a lot of promise. I used it last night to check into a C4FM net. You will need a DMR registration # to use it – one way to confirm users are actually licensed.
There are different regulations around the world when it comes to amateur radio use of encryption and protocols. Encryption is ok in a few countries.
Here in the US, whether or not encryption is legal doesn’t stop me from going encrypted when my fellow hams are being lids. 2M is filled with plenty of licensed and unlicensed clowns on in my area. P25 AES256 is nice…
Just because it is capable of encryption does not mean it uses encryption. Proprietary protocols just mean someone owns the patents and if you use it for your radio, you must pay royalties making those radios more expensive. Usually encryption requires additional licensing and, many times, hardware.
You can use proprietary protocols in digital as long as they are not used to “obscure the message.” In other words they must be available to the general public, even if for a fee, and must not encrypt the data with a cipher.
You are correct inasmuch as the vast majority of countries do not allow encryption (Poland being the only exception I am aware of).
The M17 Project is fully open source, AFAIK it is the only completely open source digimode and hardware meant for a non-HF band. That qualifies it as a known ENCODING protocol that is capable of being decoded by anyone, including that country’s governing body for radio.
DMR, P25 and System Fusion are all proprietary digimodes. They are allowed on amateur radio bands because anyone can buy the requisite hardware/software to decode the signal, the hardware/software are not sale restricted. The companies can choose to only allow certain manufacturers a license, but because John Q Public (or Uncle Sam) can walk into a shop and buy a digital radio with that mode to listen with, it’s allowed on the public airwaves.
DMR and P25R standards. The DMR and p25 standards define the minimum requirements of an air interface to be called p25 or DMR that is anybody on manufacturer can download specifications the public specification or the standard and build a radio that will work with other DMR radios in the minimum configuration It does not prevent a manufacturer from adding bells and whistles or making it better okay so when you say that Motorola has proprietary on DMR you are wrong Motorola has priority proprietary bil
Bells and whistles that they’ve added to DMR okay DMR is still a standard and by the way standards are never state of the art but because they are standards everybody knows how to do it if you don’t have a standard that’s when you get 10 different proprietary systems and by the way Motorola dropped its proprietary smart net protocols and adopted p25 which was a open standard and opened itself up to competition from other manufacturers like Harris and and others. but a standard is a minimum requirement It is not a maximum requirement and every manufacturer that finds a better way to do something I want to add a little whistle for people to buy their product that’s fine and that can be patented but not the minimum standard nothing in the middle standard is can be patented. So
Actually, if you want to use proprietary encoding protocol in device you build, you have to pay for patents. It is not allowed to use protocol without paying for patent even if it is available to you (and it is not, protocol is sold as chip you can use without knowing how protocol actually works). And that it an issue for most hams to use such protocols.
F17 resolves all that.
OMG…Punctuation, dude…
None of the amateur digital voice codecs are encrypted. It used to be that some of them were not fully disclosed. It turns out that the AMBE chip algorithm was disclosed as part of adopting it for a system commonly used by fire and police called APCO 25.
However that’s different from their being Open Source. And the Open Source codec happens to be technically superior in terms of bandwidth use and voice quality.
Because there’s nothing illegal about it. There is a really big difference between proprietary protocols and encryption. Yes, it would be nice if amateur radio manufacturers stopped using these antique things.
There is a big difference between a protocol and encryption. The most important thing to understand is encryption is meant to hide the meaning of communication. A protocol is used to make communication orderly and reliable. A protocol may also include things like compression that look like encryption to the untrained eye but the purpose is to reduce bandwidth or otherwise improve communication.
Nowhere does anything say a HAM needs to use an open protocol. There is even one exception to the encryption rule where we can use encryption for command and control of satellites. By the way us Hams have satellites.
Source – I am an Amateur Extra class which is the highest class currently available in the US.
Because we are using a published standard. DMR, DStar, NXDN, fusion, P25 are all published standards. We do not use encryption, unless the key is published. This makes it legal. FreeDV is totally open source. Not sure what they are reinventing the wheel here.
They are reinventing the wheel because all of the digital standards you just listed, while legal for ham use because the standards are published, are also illegal to build and sell equipment for because they are encumbered by patents. While that is, imo, objectionable but not a deal breaker, I’ve heard enough about heavy-handed tactics used by the patent holder against potentially competing truly open standards that I will not spend a single cent on equipment whose sale results in royalties to them.
The only slight exception to this is P25, because the governmental bodies that created the P25 standard had the clout to strong arm the patent holder right back, so they agreed to some sort of blanket license for P25 equipment manufacturers.
I am totally with you on this. Clearly, their nature is to bully.
Because the FCC requires the manufacturers to make their code publicly available. Intent is to digitize for system function, not to hide meaning.
That’s true. FCC getting complacent.
How do I become part of this project?
Join our chat first
https://m17project.org/chat/
97.309 only states that the protocol has to be publicly documented. It doesn’t require it to be freely available. This is why AMBE can exist on the ham bands. It is publicly documented but it is patent encumbered, so DVSI still has a monopoly on the vocoder chips.
I’m a month late, but the fig leaf apparently is that the protocols are “open”, in that you can find and read the protocol spec. You can’t actually implement them without proprietary components, but apparently that’s enough to get the FCC off the vendors’ backs.
I think it’s BS, personally, but it is what it is.
I don’t think there’s actually a rule against closed protocols, or protocols protected by patents. In my opinion, it’s certainly against the spirit of amateur radio, but enough licensed operators have put their money down to invalidate that argument.
Right, the rules only prevent encrypted signals from being transmitted, but DMR/D-Star/SystemFusion aren’t “encrypted” because there’s no “secret” in a cryptographic sense.
And on one level, I get the appeal, it’s certainly more reliable than HF for long-distance communications… But so’s Zoom, you know?
I have a VHF/UHF handy-talkie myself; I bought it because it was the quick-and-easy way to get on they air. But I don’t think of this as a true “ham” radio, since it’s commercial off-the-shelf, and depends on a network of (commercial off-the-shelf) repeaters. So any range and reliability I’m able to get has nothing to do with my own skills. I certainly don’t look down on Baofeng users, because I know they have their own reasons, and I only bought a “better” brand because I could afford to pay more for a higher confidence in the product. Doesn’t make me a better ham.
a company can create a proprietary system there is no law against it. Yaesu has their fusion, encryption is illegal on ham bands. Not sure how this company gets around that. Maybe they wont once the fcc finds out about it. By posting the m17 program on the internet that is now public. That doesnt mean the code they wrote has to be open source or public. They can keep the secret sauce secret as long as the program is available to the public.
This might warrant a flurry of complaints to the FCC. The whole 2m and 70cm ecosystem is full of this crap. Yes, of course a company can invent a proprietary modulation/encoding. But that doesn’t make it legal to use on amateur bands. It’s one thing when licensed amateurs develop ways to extend the state of the art, because they are usually happy to share the details. But when companies do this to exclude other companies’ products, it seems like a line has been crossed.
DMR and P25 does not imply encrypted. While DMR and P25 can have encryption enabled it can not be enabled legally on ham radio bands. DSTAR does not have encryption as an option nor does C4FM. While these modes have licenses and require a chip that carries a fee that is paid to the patent holder it does not prevent anyone who has paid that patent holder rights to receive the signal. I would love to see a radio that can operate in all digital modes with an upload or a license key installed to receive and transmit. 73,
Lin
Since when in encryption allowed in ham radio?
It isn’t, whatsoever(Aside from IIRC unencrypted but authenticated communication), but if you turn it off or use a default password, you can make hardware and protocols that are useful in both ham and commercial radio.
The one execution I recall in the FCC Part 97 rules governing the Amateur Radio Service is that a control operator of an Amateur Radio satellite may send encrypted telecommands to control the satellite.
Amateur Radio and Amateur Radio Satellite Service are separate services, if memory serves.. :)
In the US zones it is flat out prohibited for Ham bands.
In some other areas encrypted messages are often accepted with some caveats:
1. the protocol specification must be open and publicly published (M17 did this already)
2. you don’t violate your local community band plans
3. same data content restrictions apply as other Ham channels
4. your connection minimizes broadcast power and spectrum use
5. call sign identity beacon must be active unless directly controlling a drone
73
VE7NTP
Encryption is not universally prohibited among Amateur Radio services.
(And even in the US, whether it’s prohibited is debatable: http://www.n5dux.com/ham/files/pdf/Data%20Encryption%20is%20Legal.pdf
There is a difference between Encryption and Encoding. The M17 Protocol ENCODES the voice/data in a digital method using a known standard with no pretense of hiding or obfuscating transmissions.
ENCRYPTION on the other hand operated specifically with the pretense of obfuscating the voice/data contained therein.
This distinction is how the DMR Encoding standard allows for encryption as an option (for applications where it is legal). Same with M17, the encoding protocol has an encryption scheme for counties allowed to do so.
Exactly. ASCII is a code. It’s right there in the name. Morse Code is a code, ditto. Both of these, and a myriad of other standards, are used to encode information without obscuring it.
Not in the US but there are quite a few countries where amateurs are allowed to use encryption. Poland for example. However that’s not really the reason why M17 has encryption capability but the fact that without the capability there’s no chance it could compete with the existing digital protocols
Its not illegal in Sweden and i don’t think its illegal in Denmark either but you need to identify yourself in some way,
but its not recommended to use encryption.
Same thing here in Poland.
It is not allowed for international traffic at all. Domestic traffic it is up to the regulator.
On a side note there is one specific instance where encryption is specifically allowed, and that is for telecommand command of space stations. And we can all agree that’s for very good reason.
M17 is either a pistol or a gas mask; it’s not a rifle. On the ham radio side: encryption is forbidden. But it must have a well defined protocol allowing quick and non-proprietary translation. The OP needs to post more details to explain this discrepancy.
I don’t know why this was even mentioned in the article. a) any modulation mode can send digital data “supports” encryption. It’s just not legal for people to take advantage of that fact while operating under amateur licenses.
I’m going to go out on a limb here, but after looking at some of the developers of M17 and donating a small sum to them, they are not from the USA……….ergo Amateur radio is a World Wide endeavor and not all laws are the same as ours. ;)
When the OP begins with mis-identifying a pistol as a rifle and reinforces it with “It actually is” it’s rather obvious they might not be trustworthy for anything that follows. They need to provide thorough information as well as clean up an article that calls their editing of a press release into question. ;)
73
edit: “as actually, it is”
Bushmaster manufactured an M17 rifle and Zastava Arms still does.
Sorry, I meant to say “while operating under U.S. amateur licenses. Unintentional Americentric comment.
The leader of the project is based in Poland, where encryption is allowed on ham bands.
I like it a lot
The US FCC law regarding bandwidths and modulation are in Part 97.307(f). The problem is the 300 baud rate limit. There are ham radio signals that are allowed in the EU that aren’t legal in the US because of this. The FCC occasionally waives the regulation when disasters happen so that Pactor 4 modems can be used for a higher data rate.
Part 97.113 (4) covers encryption which limits it to a few specific cases.
That’s only for frequencies below 30 MHz. This is a VHF walkie talkie.
It’s a 300 baud limit in HF, but a 19.2k or 56k baud limit for VHF/UHF. This walkie talkie can exceed that.
It can also be used as a hammer, what’s your point? Just because it’s capable of exceeding rules on baud rates, doesn’t mean that it will be used in such a fashion.
I think arbitrary baud rate limits are a poor idea as opposed to spectrum emission based rules.
The FCC symbol rate limits date back to 1980. It’s time that they catch up with technology.
Normally I’m very supportive of projects, DIY efforts and open source; however, I tuned out upon reading the phrase, “replace DMR, Fusion, D-Star, etc”. What’s next to get rid of? Morse code? I’ve been a ham since 1977 and see no reason to _”replace”_ existing protocols. Such things grow and die by whatever support they manage to rally. Go ahead, come up with something and see if the community adopts it. If Dr Taylor (K1JT0 did it, so can you.
Obligatory links:
https://physics.princeton.edu/pulsar/K1JT/wsjtx.html
https://physics.princeton.edu/pulsar/K1JT/refs.html
If Morse code included a requirement to pay Mr. Morse a royalty to build a keyer, I would humbly suggest it too be replaced with something open in the true spirit of ham radio.
Dr. Taylor’s excellent digital modes aren’t replacing Morse code, but I think the comparison is actually quite apt: M17 or another open protocol should “replace” the proprietary offerings in the same way, by being better and thus widely adopted and eventually driving the older, less capable (and, imo, un-hamlike because patent-encumbered) stuff to the fringes where it will still be used by enthusiasts and diehards but the new hotness will get the lion’s share of practical usage. I wish this project and others like it the best!
That was a comment made by another Ham that was interested in our project early on in the beginning. We’ve made it our ethos, in a way. Nobody’s trying to take your protocols away, as a matter of fact we’re developing one that’s more open to the Amateur Radio community by replacing proprietary offerings available to you currently. Its your choice to use it or not. Just as its your choice to expand the hobby or be stuck with your key, pounding out to your heart’s content.
I Hope They Don’t Go That Route With DRM To Replace HAM Radio Protocols, I’m Studying For My HAM Ticket,What They Trying To Do To AM Radio Broadcast Band,They Are Trying To Do To HAM Radio,!(This Tenickracy!!!!),We Got To Stop Them!!!!
So what does this new protocol bring to Amateur Radio that CODEC2 (http://www.rowetel.com/wordpress/?page_id=452 ) doesn’t?
M17 uses CODEC2 for encoding voice. The M17 spec defines all the other layers around it.
And in case you have missed that:
http://www.rowetel.com/?p=7324
It’s the most viewed entry on Rowe’s page.
Hello from KG5ZEF Ray, Ponca City Oklahoma. I can remember an open source project around 25 years ago called “IP over ham” on Linux and I haven’t revisited the outcome, but this hardware looks like it could do it as is provided the correct software can be used. When storm chasing there is often data dropouts on cellular networks and this causes issues with delays of radar updates. It would be cool if this could be used as a backup run off a unit at home, but the rules and regulations would prohibit it from being fully usable.
Thoughts? 73.
There have been many, many “IP over Ham” projects lately. New Packet Radio comes to mind as an example. However, file and image transfer is a major component to the protocol as it’s being developed. Your thoughts are on track with M17!
@RAY said: “Hello from KG5ZEF Ray, Ponca City Oklahoma. I can remember an open source project around 25 years ago called “IP over ham” on Linux and I haven’t revisited the outcome…”
“IP over ham” as you say has been around as long as packet radio has been around, and it’s not a “project” it’s simply the stacking of existing open protocols. The radio and modem are the Physical Layer (OSI Layer-1), AX.25 packet switching is the Data Link Layer (OSI Layer-2), IP is the Network Layer (OSI Layer-3), TCP or UDP is usually used as the Transport Layer (OSI Layer-4), etcetera. You don’t have to use AX.25 to move IP packets around, anything that can bridge OSI Layers-1 & 3 will work. It’s just that AX.25 is the “Amateur” radio optimized version of the generic X.25 packet-switching protocol.[1][2] Actually there’s nothing “Amateur” about the AX.25 protocol at all, it works perfectly with commercial radio-modem equipment.
References:
1. AX.25 – Wikipedia
https://en.wikipedia.org/wiki/AX.25
2. Configuring an AX.25 interface for TCP/IP
https://tldp.org/HOWTO/AX25-HOWTO/x1194.html
DMR is not proprietary. I implemented part of it and you can just download the standard. Some voice codecs may be, but this is not related to DMR itself.
While you’re right that the DMR specification is publicly published, parts of it are still patent–encumbered, and licensed by Motorola.
Oh well, the same could be side of D-Star.
D-Star is all open, except for the main codec – that nasty voice codec.
There are software devoders thatcpuld be used with an SDR, but even on a powerful PC/Laptop you’ll need an USB dongle with that proprietary voice chip to make it legal. 🙄
Don’t get me wrong, but that violates ham spirit.
The heart of amateur radio was and is experimentation and right of being allowed the build your own radio. In the analogue world, you could always construct radios with different schematics, yet you were still able to communicate with all radio amateurs.
That voice codec is akin to binary blobs on some Linux platforms (Raspberry firmware). It’s as if a company patented a whole modulation type (like SSB) that only their radio line is allowed to use.
I’ve always felt this argument was oversold. Building your own radio always meant buying components — while technically you could mine your own copper, smelt it, draw your own wire, make your own grids for tubes, and blow your own glass, that presupposes you have your own copper mine and high-grade silica source. And lots of time and skill, of course. As long as their are multiple sources for components this all goes out the window. The larger issue with things like DSTAR are vendor lock-in throughout the stack, and that already seems pretty moot.
In 1979 there was NBVM, Narrow Band Voice Modulation. Commercial scheme for improving SSB, it folded voice over into a smaller segment, answered a compander. It was presented as a product, but fanfare, then a bit later a schematic. It was complicated, all wholly so plenty of active filters. I can’t remember if it showed modules off it was so complicated that it in effect required buying the modules. It was in the 1979 ARRL Handbook, and never again.a
There was debate of the actual.improvement, the commander seemed purposeful than frequency compression. It was costly, so few took it up, which in turn meant little incentive for others to follow.o
There was no “organic” flow, just a finished product that added maybe a bit of improvement. No third party projects,no improvement as schematics were passed along.
If I can’t buy a blank chip, compile open source code and install on that chip and then be compatible with whatever encoding/modulation is being transmitted, then that encoding/modulation/whatever has no place on HAM Radio frequencies. Replace the hardware intensive method with software for an SDR if you like. Same thing.
While you have every right to decide what belongs in your shack, you do NOT have the right to decide what belongs on the Amateur Radio Service bands. CFR Title 47, Part 97 defines that, in the U.S., anyway.
Many Hams make the choice to avoid proprietary protocols that are for all practical purposes restricted to owners of a particular brand of equipment. I’m one of them. So yeah, it annoys me when others choose to operate “walled gardens” created by Motorola or Yaesu. But that’s their choice, even if I consider it a stupid one, and it’s perfectly legal.
I DID say “for all practical purposes”, though. There are less practical alternatives. For protocols that some commenters have labeled “encumbered by patents”, I say,
a) I am not a lawyer.
b) My lack of legal knowledge is in no way limited to U.S. law.
c) Patent protection prevents the commercial exploitation of intellectual property described in a patent without the permission of the holder of that patent From what I have been able to determine, it is not unlawful to build equipment that make use of the information in patent applications for personal use, and since patent applications are public documents, it is also not unlawful to help others to also exploit this information for THEIR personal use.
d) Notwithstanding the provisions of the Digital Millennium Copyright Act, radio emissions are subject to analysis, and such analysis can in many cases reveal details of the encoding of those emissions which are NOT revealed in patent applications or other freely available documents. Exploitation of THIS fact is sometimes termed “reverse engineering”, and this has been responsible for free and open source descriptions, specifications, and implementations of a great many “proprietary” protocols,
So it sounds to me like you’re saying that since many manufacturers of radio equipment do not make it EASY for you to freely make use of their ideas, they should be banned, ostracized, or otherwise excluded from your hobby. To which I say, “waaah”. Buy it, hack it, or just shut up.
@BrightBlueJim – You’ve stumbled on a common misconception. There is no allowance, at least in the US patent system for personal use. You are NOT technically allowed to build things for your own personal use using information from a non-expired patent.
Of course, it is very unlikely that anyone will catch you. And even if they did, most big corporations have better things to do with their expensive lawyers than go after a single hobbyist working in his/her garage on a personal project. (Although they might if you share your build instructions online).
But technically you are breaking the law. Is that what ham radio is supposed to be about? Building technical skills that you must keep secret and cannot share all while breaking the law?
That’s not how it was sold to me.
Do you have any source for this other than the one I used, i.e., “that’s what i heard”?
“High grade silica”??? You mean….. the beach? How did the Romans do stuff without time travel?
@Kenton Hoover – Where would you draw the line? If it’s only about having multiple sources of a component then why not disallow the building of radios altogether? There are several radio manufacturers to choose from.
I think one big difference between the analog components you mention and a proprietary codec is that we all learn how those analog components work at least as part of beginning electronics. And by the way, people do make most of those things from scratch. Inductors are probably built more than bought. Homemade capacitors aren’t that uncommon, often fixed or even screw-driver-adjustable ones are made from PCB material. There are many postings online showing variable capacitors made from copper flashing. There are a few home makers that are niche-famous for making their own vacuum tubes and even a few who have made their own transistors. (see Jerri Ellsworth).
And what multiple sources are there for patented voice codecs? You might find the chip in radios from different manufacturers but they always contain a chip from whoever owns the patent!
I know a lot of people who have a lot of analog electronics experience look at todays microprocessor enabled black boxes and think understanding them in the same way they understand an analog circuit is impossible. This is not true! In a one-semester college computer-organization course we went from logic gates to the design of a simple CPU. These things can be understood. Granted, a real-life product is much more complicated than our simple classroom CPUs but the base concept is the same. DSS is pretty easy to understand, people were making low-frequency DSS’s with EEPROMs, a clock and a resistor ladder years before these nice SI chips were easily available to every hobbyist.
And codecs? Well, if those were out of reach to the hobbyist then where did Codec2 come from?
“It’s as if a company patented a whole modulation type (like SSB) that only their radio line is allowed to use.”
You mean, like they DID patent frequency modulation.
It is not abut patents itself but what limitations on use patent holder enforces. lots of stuff is patented but free to use.
The problem with digital ram modes is not the mode itself (IE the modulation, codecs, and the communication between two transceivers). The problem is the higher level constructs not letting hams leverage the mode to its full potential . Packet Radio (AX.25) was better in some ways in that nodes could construct a mesh network that was ad-hoc and very flexible. DMR, built on a commercial infrastructure, is extremely rigid, and some hacks were put in place to try and make it something more than just a digital version of any other classic mode. So hams can key up a talkgroup on a repeater, which “subscribes” the repeater to it, allowing the ham to communicate with another set of hams worldwide on that group. That’s pretty much it. Although DMR supports messaging protocols and raw data, none of the back-end infrastructure makes use of it, and most repeater owners have even simple SMS between stations on the same repeater blocked. I won’t even get into the mess that is configuring the zones and channels on a DMR radio, and the rigidity involved.
What HAM needs is a digital infrastructure that is a bit more like a trunked or cellular system, but using ham bands and digital modes. So similar to APRS, there would be national / global frequencies on each band for network traffic to handle repeater beaconing, negotiate QSO requests, etc. So as a person is mobile their radio would be aware of what repeaters are in the vicinity, and they can then switch to those frequencies to ping back to the repeater to receive any QSO requests, receive SMSs and other push type notifications.
SMS is THE killer app for hams. A hundred hams could carry on dozens of QSOs at the same time on the same simplex 12.5 kHz channel using SMS. This would require radios to either have good enough text input for it, or the ability for an app on a cell phone to use the radio (via bluetooth or wifi) to transmit the messages. However, SMS is totally ignored in HAM, largely because it’s incredible difficult to find a radio for which text entry isn’t a convoluted mess. Basically the best we see is T9 type input, which went out of vogue on cell phones over a decade ago.
Anyway, my point is that a proper ham digital mode and transceiver hardware needs to be more than just APRS and the ability to do voice communication. So anything trying to reinvent this wheel needs to truly embrace the digital future. Kids of today would fully expect to be able to do SMS / Snapchat / Messenger type text messaging on HAM, if we want to see HAM radio keep being fully relevant in the future.
PS – Hackaday’s Facebook login feature is not working.
PPS – Hackaday’s Google login feature is not working (Error: Google failed to return an expected code.)
M17 is going to support messaging and file transfer. Image file transfer can be used for SSTV (e.g. for the ARISS).
DMR is bor as a commercial protocol, so it’s expected that some modification and quirks on operation have to be done, it’s like a radio I have that was built to be used on 460 MHz and has a 16 channel rotary adapter, and if I have to change repeater frequuencies on CTSS tone I have to open it and use a DOS program to reprogram it.
Now Mororola is makig DMR radio and hams are reprogramming to use on ham bands, but unfortunately instead of simple fm there’s the more complex DMR stack.
Does SMS require some sort of underlying telephony platform? If that’s the case then the question should be whether to or why implement GSM (or something like it) on ham bands, way before considering why HTs don’t have QWERTY pads.
Not telephony per se, but SMS does require a service. That is, it is not a peer to peer protocol. The second ‘S’ of SMS is “Service”. Consider how it behaves: SMS usually works without requiring a live connection between the sender and the recipient. If your phone is either off, or not connected to a network, the message is not necessarily lost. Usually, once you reestablish connection to the network, the message can then be delivered. This means that the message must get stored somewhere, if you wish this to be the case. This is not essential; SMS can be either store-and-forward, or forward-and-forget. So, depending on how reliable you want it to be, it can operate similarly to email, where your device periodically checks with a message server, and can even send return receipts, or it can be sent to the nearest repeater to where you were last known to be, and completely hit-or-miss. In the forward-and-forget model, this requires that the network have some means of tracking where you are, since this allows it to send messages addressed to you through a route most likely to reach you, rather than broadcasting every message to the whole world. Each time you go on the air, your device searches for a server, and the route to you is updated on all of the servers. In the case of the store-and-forward model, it only requires a server whose nearest repeaters can be looked up, so that your device has a fighting chance of connecting to the server. The bizarre thing about the cell network based SMS, is that each service can use store-and-forward, forward-and-forget, or something in-between, where each node retries a certain number of times, and then gives up. This is why a decade ago, texting to someone on a different provider was iffy at best, and even today, messages are often lost.
The current state of 2-meter and 70 cm usage involves individual repeaters and repeaters that are linked to other repeaters so that they repeat everything on those other repeaters. Neither of these is good for an SMS-like service – an isolated repeater can only deliver messages within its own coverage area, while a repeat-everything link would quickly become saturated since the more repeaters are linked together, and the larger the area covered, the more messages are broadcast to the whole world (for limited definitions of “world”). An alternative is a nybrid, where repeaters send each message to an Internet-based server where the “storage” part of store-and-forward is done, but this obviously breaks down under any conditions where the Internet becomes unavailable, so anybody wanting to have a disaster-proof version of SMS would reject this option.
For models that work under sketchy conditions, you can look to what worked before there were reliable networks. In the ditital world, this was UUCP, or Unix-to-Unix Communications Protocol, in which messages were forwarded in the general direction toward the recipient (usually from one university or corporate time-shared computer to another) until it found the recipient, which could take days, since some of these connections were once-per-night at best. In the human world, we had radio relay, where a person wanting to send a radiogram to someone contacted a ham operator, giving the name, city, and I guess phone number of the recipient, and the ham would batch up messages by city, then make contacts in the general direction of each city, forwarding them to someone closer to each recipient, until it ultimately got to a place where the recipient’s phone was a local call for the ham holding the message. Which is pretty much the same thing, including the “could take days” part, but with human links instead of computers. Both of these were obviously store-and-forward systems, but neither used a central server – if the recipient was not at the location the sender predicted, the message would not get delivered. Looking to other services, we have a fix for that, though: In ancient wired telephone networks, if you changed locations, the service provider at your old location could automatically send back a message giving the new location. Similarly, most postal systems have a forwarding mechanism.
Is there a packet std for this somewhere?
Keep a watch on the project website, it will be updated soon. I don’t want to post the links here since they may change.
APRS could be the service you’re looking for. No IP required.
So, bottom line – Is this good, or do I not renew my Extra Class in 2027?
Most digital radio communication is not simplex, so unless they also intend to provide an alternative network system which supports their protocol, and they also hosts and administer the central servers for this, then IMO their new protocols are never going to be widely adopted.
The other thing against them is the momentum of the existing digital modes.
At the last count there were 152,000 registered amateur radio operators on DMR.
See https://database.radioid.net/database/dumps#!
There are also tens of thousands of registered users on the more proprietary modes like DStar and YSF
And people don’t buy a radio unless they can talk to other people using that radio.
If their radio also supported other protocols, e.g. DMR, they may have more of a chance of people buying their hardware, and then perhaps using their protocol.
But even then, I think the lack of people to talk to will probably mean people would just use their radio for DMR.
There is also some miss-information about DMR.
DMR is an ETSI (European Standards Organization) standard, but the standard only defines the lower levels of the protocol stack. It does not specify the audio codec or the method to send SMS messages etc.
All the manufacturers agreed to use AMBE as the audio codec, but technically any audio codec can be used on DMR as long as the data is 27 bytes per 60mS of audio.
e.g. Codec2 could be used on DMR with some modification.
Different manufacturers have their own additions to the higher levels of the protocol stack, some of which may be patented, but the layers of the stack required to have an audio data conversation are in the ETSI standard.
“Momentum”? If there’s no alternative, we can’t talk about momentum.
TR-9 is both DMR and NXDN capable.
BTW: good luck with implementing Codec2 into existing radios. Codec diversity within one system is the worst thing you can have.
I’d need to ask David if Codec2 could generate 27 bytes from 480 bytes of audio samples relating to 60mS of audio.
If he can do that, then I don’t see a huge problem with replacing the codec block in many Chinese DMR radios.
The firmware in the Chinese radios got hacked years ago, and the entry point and buffers etc of the codec decode and encode is well known.
Yeah, there’s a good reason why nobody did that yet. You just seem to be reluctant to accept it.
Advantege of M17 is that it is open for any codec use, even proprietary. Protocol documentation mentions use of existing proprietary protocols.
That is the most logical approach. Make open system, and let two radios negotiate how they would connect to each other. That makes all radios compatible even on digital level. Right now we have that level of total compatibitiy only in analog layer.
Protocol also defines all infrastructure for networking. And as it is open, people should be more motivated to invest time and expertise to develop tools and software using that protocol.
By proprietary I’m guessing you’re talking about PACTOR? Since PACTOR is primarily used for Marine SSB and only marginally in ham radio, I suspect that’s where the legal loophole is. Still annoys me though.
Published does not equal open.
It’s not open if creating your own implementation can get you sued.
Whoops, missed.
Funny thing is that all the digital ‘proprietary’ protocols mentioned ARE actually open AND published.
Published does not equal open.
It’s not open if creating your own implementation gets you sued.
This is something we really need. But it’s not good enough just for it to exist. I see so many hams that are so excited that their local club is putting up a new repeater using the latest proprietary protocol-x. We need that same drive and excitement, or more for an open system.
It still amazes me when I see this conversation repeated over and over again:
non-ham-> Isn’t ham radio just a bunch of old, obsolete technology.
ham enthusiast-> No, we’re actually on the cutting edge, often leading the way with new technologies
non-ham-> Such as?
ham enthusiast-> Tells about the latest turn-key repeater the local club put up using some proprietary digital technology that was originally designed, debugged and marketed for commercial radio before it ever hit the ham bands.
We need to stop repeating this conversation.
This is so true. I see that kind of stuff every day on amateur radio related Facebook pages (and Reddit). Same thing, over and over again. I’m tired of DMR and AMBE. I’m even amazed that hams still use APRS.
M17 is receiving a lot of hype and we are going to use it to spread the word. I’m really doing my best while working on M17.
“… it’s not just good enough to be open, it has to be legitimately competitive.” That’s the goal. M17 aims to be competitive in ham and non-ham markets, offering an alternative to licensing codecs that are patent-encumbered. I agree that instead of saying “hey, look at this new repeater with X protocol and Y codec that has been used commercially for ages”, Hams should be saying “hey, look at this new repeater with a freely available source and codec that has superior audio quality to anything available commercially, developed by a worldwide team of other Hams!”
This has already been done with “Codec 2” and FreeDV. Obviously, this mode didn’t take off well. I predict the same for M17.
this project is literally an implementation of codec2
I read all, or tried to read and understand some pieces and parts of it….however it is quite evident to me that: due to my old age and diminishing abilities and having been lost in the new methods of modulation and protocol…..I am irrelevant to the whole discussion. Having a different career background (than digital protocalls) about all i can do is wonder what any of this stuff is about. I have managed to figure out some of the digital software and use HRD and SSTV and a few other modes that I still dont quite understand. I would assume the newer and younger hams are initiated in the field of digital everything and I wish them well. Being 80 yrs old its “greek” to me. However I still like to use Olivia, PSK and SSB sp that i can have a qso with a living human. SO, ya’ll have fun with your digital toys and old folks like me will have qso’s with people. 73.
It’s not really deemed encrypted if the codec is public. As long as the government has access to listen to and un-encrypt the transmissions they are not secret. The reason for the FCC rule was to prevent spies and saboteurs from endangering the nation in the time of war. There was a black out of most amateur radio during WW1 and WW2 to prevent the nations secrets from being given to the enemy. We are light years beyond that in technology now. Any of the modes of transmission we use today would have been determined to be secretive in 1942. This is not the case today. Whichever method that is used it should be open source to be legal.
Yes, this only applies to the US, but specific to the comments made about the US laws people do not seem to be reading the FCC regulations in this discussion.
§97.113 Prohibited transmissions.
(a) No amateur station shall transmit:
(4) . . . 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.
The only “except as otherwise provided herein” within Part 97 is §97.207(f) “Space telemetry transmissions may consist of specially coded messages intended to facilitate communications or related to the function of the spacecraft.”
There might just be a logical reason that this contained in the question banks for licensure at the General (G1B07) and Extra (E1D02 and E1F09) levels.
There is nothing that says the encoding must be open source, which is why we have proprietary codecs and they are legal. They are not intended to “obscure the meaning” of the message, but rather to compress it so it can fit within the allowed spectrum.
Proprietary codecs are absolutely against the spirit of what amateur radio is, which is why we should develop a commercially-viable open-source standard that could be used not only in amateur radio but also for commercial radio transmissions. This means that it must have the potential to encrypt the transmissions (as this is allowable for commercial radio traffic). Otherwise, any standard will wither and die.
And question T1D03 for technician.
*ahem *
https://xkcd.com/927/