Continuation of a series begun in Zero Retries 0179 - Explaining the Use Case for Data Over Repeater - Part 1.

Begin With The End In Mind

One of the primary tenets in the great book 7 Habits of Highly Effective People that has really stuck with me in the decades since I first read it is:

🔗 Begin with the End In Mind

I’ve found that simple advice often helps me when I’m stuck in “process” — how do you want this situation to end up? Once you can imagine what you want to see at the end of the “process”, next steps and an overall plan sometimes become clear (but not always), but at least you have an idea of what to tackle next in the process of getting to “The End”.

Thus, what follows is my current thinking of “The End In Mind” with regard to a potential long term project of advocating that Amateur Repeaters be reimagined for data communications, either as dual use with voice, or repurposed to be primarily for data use.

The Current Generation is “Data First”

One of the “barriers to entry” for getting Amateur Radio operators of the current generation on the air and active within Amateur Radio is that they prefer data communications … text¹ … to voice communications for their casual communications.

Just as modern society has migrated to data / text communications for much of its typical communications — text messages, sharing photos, email, social media (both text and short video) — Amateur Radio has begun to migrate its communications (and experimentation) to data / text. Witness the widespread use of data modes such as WSPR, PSK31, FT8, VARA, FSQ, and APRS, and even EME using JT65. We continue to invent new data modes, most recently LongChat (link is to YouTube demo).

But using data communications on VHF / UHF isn’t very convenient and somewhat expensive because to operate simplex on VHF / UHF requires a station to use powerful radios, external antennas, a reasonably high / clear location, etc. Or, data operations on VHF / UHF are limited to “data light” APRS operation via APRS digipeaters.

Digipeaters Work… Kind Of…

This topic is worthy of a longer discussion, and perhaps a re-examination in this era.

What follows is my perceptions of digipeaters, my memory, my technical knowledge. What I will state in this section is my best understanding of the state of digipeaters, without resorting a thoroughly researched, very long “deep dive”.

Digipeaters are a reasonable technology for creating ad-hoc networks for short data bursts such as APRS transmissions of weather data, text messages, position data, etc. However, digipeaters don’t necessarily work well for wide area use, or larger numbers of users, or longer data transfers. In a word, digipeaters were developed as a hack in the early days of packet radio by using a bit of memory in a TNC for receive-buffer-retransmit operations. In short, digipeaters enabled some networking in early packet radio. The utility of, and the shortcomings of digipeaters were somewhat overcome by a number of improved types of digipeaters, including Net/ROM / TheNet, TexNet, ROSE, and others I’ve now forgotten². APRS added some optimizations to digipeating such as “digipeat only when it’s a good idea to do so” such as home stations, rarely, and mobile stations, only when moving.

But digipeater operation falls down in a number of ways:

• Digipeaters are generally very simple devices, with limited buffer memory, and thus can accommodate only small amounts of data and short packets. Larger amounts of data or longer packets cause “fragmentation” through the repeater, and reduce throughput.

• If the number of users of a digipeater rises above a certain threshold where too may transmissions exceed the digipeater’s (channel) capacity,

• The Hidden Transmitter (or Node) Problem where some digipeater users cannot receive each other directly, but the digipeater can hear all users equally well, and thus there are “collisions” when attempting to use the digipeater simultaneously. This cascades into the digipeater and users “backing off”, severely reducing the overall throughput of the channel.

• And, generally, digipeaters are “fossilized” in using 1200 bps Audio Frequency Shift Keying (AFSK).

In Contrast to Digipeaters, Repeaters Work Well

Repeaters changed the paradigm of VHF / UHF usage in Amateur Radio, making it easy and effective (and relatively inexpensive) to operate on VHF / UHF and communicate between groups of Amateur Radio Operators in a local area or region.

Repeaters are located in high locations such as skyscrapers, towers, or mountaintops, allowing modest user stations to reliably communicate with each other via the repeater.

An additional advantage of repeater’s high locations is that they allow reliable communications over a wide area.

Repeaters operate in a simultaneous receive / transmit mode³ full duplex by using separate receive and transmit frequencies. When two stations are in communication with each other via the repeater, every user of that repeater knows that the repeater is in use, and thus “collisions” (such as result from Hidden Transmitter / Node Problem) are minimized.

A subtle “feature” of repeaters is that they can act as a “water hole” — a central gathering spot… kind of like a continuous club meeting for a particular group of users. (From other articles in this issue, the water hole effect is especially prominent with Amateur Radio video repeaters.)

Lastly, using a repeater de facto enforces minimal / compatible technical standards of all users, such as requiring a reasonable signal for using the repeater (such as high transmit power or better antennas for users that are farther from the repeater), or use of a particular digital voice technology, or any number of other technical standards.

But… Amateur Radio VHF / UHF repeaters are built for, and used for, almost exclusively voice operations. Yes, there are some data capabilities incorporated into some digital voice systems used in Amateur Radio — see Zero Retries 0179 - “Data capability” on D-Star, System Fusion, DMR, and P25 as explanation of why those systems aren’t very relevant in this era.

The Era of Quiet Repeaters

Simultaneously with the “rise of data communications”, a trend is emerging that Amateur Radio VHF / UHF repeaters are becoming “quiet”. Repeater activity is declining, which becomes a vicious cycle. If a repeater is quiet, there’s less incentive and interest to monitor that repeater for interesting conversations, and the repeater grows even quieter.

Competitive Pressures for Amateur Radio Spectrum

There is also the trend of increasing “competitive pressure” for the VHF / UHF spectrum that Amateur Radio has been allocated, and has been allowed to operate on a shared basis. In this era of almost entirely wireless communications — mobile device networks, Wi-Fi, satellite communications such as Starlink, the competitive pressure is increasing. This is not fear mongering. Current examples:

• The Amateur Radio allocation in 3.3 – 3.5 GHz has been eliminated. This was significant because several wide area Amateur Radio microwave networks were built to use this band and thus able operate without interference from unlicensed systems in the 5 GHz band.

• While there are still some segments of the 5 GHz band where Amateur Radio can operate exclusively, the Amateur Radio allocation has been reduced in favor of allocating almost all of the 5 GHz band(s) to unlicensed operation.

• If FCC Docket 24-240 is adopted as proposed, Amateur Radio may lose the effective use of 902-928 MHz because of competitive pressure to reconfigure that band to move Amateur Radio operations, along with all unlicensed operations, into 907-918 MHz.

• Amateur Radio use of the 1240 – 1300 MHz band will inevitably be reduced in the next few years. The primary use for this band, worldwide, is for Global Navigation Satellite System (GNSS) other than the US Global Positioning System (GPS). These new GNSS systems such as Galileo (Europe), GLONASS (Russia), BeiDou (China) are now coming online and using this band and do not want interference from Amateur Radio operations.

Thus there is an increasing imperative to use our Amateur Radio VHF / UHF bands, and demonstrate actual usage, rather than allocating them within Amateur Radio as “allocated” such as repeaters coordinated and built… but provably not widely used when surveys are conducted.

Putting It All Together — The End In Mind

Thus, I posit, that all of the above trends combine to create a multiple “win” situation that quiet repeaters can be repurposed for shared data / voice operation, or in some cases data mostly operation. Doing so makes it easier for new Amateur Radio Operators to become active on Amateur Radio data modes in the same way that repeaters make it easier to use VHF / UHF for voice operations, and the repeaters become more widely used, and our Amateur Radio VHF / UHF bands are provably actually being widely used.

A Personal Perspective of Data Over Repeaters

I intend this section not as a “stroll down memory lane”, but instructive background from real world experience.

When I moved to the Seattle area in 1987, I discovered a very active Packet Radio group called the Northwest Packet Radio Association (NAPRA). NAPRA was very active in packet radio — constructing digipeaters and later simplex nodes with links using Net/ROM networking, and user education about getting active in Amateur Radio Packet Radio. Bulletin Boards were commonly used. Years later, some of us in NAPRA became tired of the “petty fiefdoms” of the node owners and BBS sysops. There were endless months of debate about the optimum parameters in the Net/ROM network and the BBS sysops often killed (censored) email from folks they didn’t like, and killed bulletins they didn’t agree with.

Sometime in the early 1990s, when it had been ported to MS-DOS, some of our most technical members discovered KA9Q NOS and began talking it up amongst ourselves. We first experimented with it over simplex links… and found that TCP/IP over Amateur Radio was fun and very interesting and more capable than “plain” AX.25 packet radio (and nodes, and BBS’s). For one thing, we could communicate directly between ourselves using email, not messages on a BBS….. with a BBS sysop in the middle of our communications. We gradually grew the TCP/IP network with some folks running multiple ports (radios on different bands) and getting more and more capable until eventually we developed the network described in an article I wrote for the 1995 ARRL and TAPR Digital Communications Conference — The Puget Sound Amateur Radio TCP/IP Network (PSARTN). I use the PSARTN terminology in explaining it to a wide audience, but the actual name of the network and group, chosen by the users was WETNET — Washington Experimenters TCP/IP NETwork.

All of the basics of what I’ll explain in this section are in that article, and I’m glad I wrote it because it’s one of the few surviving bits of documentation of that network — that and one of the 440 MHz repeaters gathering dust in my shop. We were so busy doing and experimenting that we just didn’t write things down except in email lists of the era… and the server of that list is long defunct and few folks’ email archives survived the deaths of the limited life storage mediums of the DOS / very early Linux era. So this description will be partially out of memory, and much out of emotion and remembrance of the excitement of that era.

The PSARTN was groundbreaking in its day for combining five significant advances in packet radio in its day:

• At its peak, PSARTN consisted of three 440 MHz repeaters, one 144 MHz repeater, and one 222 MHz repeater, and some simplex channels. There were a few other repeaters that were attempted.

• The 440 and 144 MHz repeaters operated at 9600 bps using a TAPR big regenerative option on the TAPR 9600 bps modems (on a modified TAPR TNC-2 clone).

• The PSARTN used TCP/IP over AX.25. Our subnet in 44Net was 44.24.x.x.

• The repeaters and simplex channels were all networked via routers; initially KA9Q NOS, then JNOS, then Linux on PCs located at the repeaters.

• There was an Internet gateway and we selectively gated messages from usenet and email into and out of PSARTN.

Most of us… the more dedicated, technical folks, operated on the 9600 repeaters using JNOS, and there were a few bleeding edge folks that ran very early versions of Linux. TCP/IP worked well due to the adaptations for slower speed links that KA9Q put into his NOS code.

Admittedly, some of the excitement of the PSARTN was that it was “our own little Internet”. We were all able to learn about Internet technology and TCP/IP on our home stations — learning the basics of addressing, routing, and the many, many ways you can misconfigure a router, including creating many, many packet storms (why is my radio transmitting continuously?). Mostly we used fixed IP addresses and static routing, but we experimented with dynamic routing and Dynamic Host Configuration Protocol (DHCP).

Generally, PSARTN worked well. The main issues we had were getting new users up to speed on using and configuring their systems for TCP/IP, tweaking their 9600 systems for correct deviation, and the vagaries of the portion of AX.25 that was the Network Layer and Transport Layer. We also had to contend with slow computers (PCs operating at the original PC’s clock speed of 4.77 MHz were commonly in use), setting up KISS in TNCs, bugs in TNCs, and the serial link between the PC and the TNC (had to be faster than 9600 to keep up — upgrading to RS-232 serial cards with 16550 UARTs fixed that issue.

When we were using simplex links with more than two users, it was only partially successful because we had varying modems, radios, transmit delays, hidden transmitters, differing deviation settings, etc. Thus one of the biggest successes of the repeaters operating at 9600 bps bit regeneration is that you could simply start a ping session with another user on the repeater, and just keep adjusting your parameters of your radio or your system. For example, deviation was easy — just adjust for best ping success. If every ping was coming back, your radio was de facto set well enough. And because everyone was working through the same system (the repeater), if you could work one person, you could work everyone who was a user of that repeater.

The three 440 MHz repeaters were regional — one in North Seattle, one on the far East side of the Seattle suburbs, and one far South of Seattle. The 144 MHz repeater was a wide area repeater. With the dispersed coverage of the repeaters, the users generally didn’t have try to get into a distant repeater. The 440 repeaters were all connected over a wireline backbone (but some folks had radios on multiple repeaters and could do failover routing).

Once you got your system configured… it just worked. We were able to run email between ourselves, and do multi-user emails (bulletins), and we even had list servers, and the aforementioned Internet gateway. We did all the usual Internet activities of the era — file transfers, pings, email, finger, etc.

Really… it… just… worked. We didn’t need any services on the Internet — PSARTN was an Intranet. The Internet connectivity was a convenience, not a necessity.

One of the wildest experiments that we did was a weekend FTP session of some big (for our network) file that was going to take hours. By then our TCP/IP knowledge had advanced to having a sliding window protocol — as long as a transmission got an ack, the sending station would try sending longer and longer packets. For this experiment, two stations were able to access one of the 440 repeaters and the 144 MHz repeaters. They configured the transmitting station to use one repeater, and the receiving station to send the acks via the other repeater. It worked spectacularly well and the sending station transmitted the file. The sending station’s repeater got a bit warm, but it was built for continuous duty.

I’m one of the few folks now that remember the PSARTN, and care about it, and the example that it set for how useful repeaters can be for providing a quality experience of data communications over Amateur Radio to new Amateur Radio users who are interested in digital communications. Connecting to PSARTN repeaters was… challenging… and expensive initially. I got connected initially with a 2 watt crystal controlled radio into an expensive run of low loss coax cable into an 11 element beam. Then our group discovered an easy modification for 9600 on surplus GE MVP UHF and VHF radios and then connections into the repeaters were easy.

But the key point — the repeaters worked to connect us all with high speed data communications over Amateur Radio VHF and UHF channels and thus users could have a modest station that worked well.

Operating TCP/IP on the PSARTN was an incredibly satisfying experience, and once enough of us had enough experience in all the gotchas, getting new users onboard was pretty easy because all they had to do was to pick a repeater, get their equipment built up (we gave lots of advice), and get it configured (we had documentation).

One of the primary successes of data communications using repeaters such as the PSARTN was there were very few collisions of two stations trying to use the repeater simultaneously. Every station using a repeater knew, within a few hundred milliseconds, when a repeater was in use and to not transmit so they wouldn’t cause interference. Operating at 9600 bps meant that most transmissions were only a few seconds, thus there was lots of channel capacity on a repeater to accommodate many users.

As for the plaint that repeaters are a single point of failure and encourage a “user mentality”, there is the same issue with voice repeaters (which are, or were, widely used), which is generally answered by redundancy — being able to access other repeaters.

Having experienced the Puget Sound Amateur Radio TCP/IP Network, with the technologies of that era… I continue to be gobsmacked by how much better those directly relevant technologies are now:

• We have full power, frequency agile, 144 / 440 MHz radios with flat audio connections (no modifications required) such as the Yaesu FTM-6000R that can easily do 9600 bps (and potentially faster).

• We have high quality audio interfaces such as the Masters Communications DRA50M which connect directly to the most common flat audio connection on radios (the “data / 9600” port using a 6-pin MiniDIN connector).

• We have very fast, very cheap dedicated computers such as the Raspberry Pi series that can be dedicated to Amateur Radio activities.

• We have excellent software modem implementations such as DIRE WOLF that can do things that hardware TNCs of the PSARTN era could never do, such as single bit error correction using bit-flipping.

• We have well-understood implementations of TCP/IP in Linux and many applications.

• We have several implementations of Forward Error Correction, which makes a huge difference in overall reliability of data communications.

Thus I think the time is right to try data communications over Amateur Radio repeaters, and see if we can “recreate the magic” of we users of the PSARTN experienced.

I’m grateful to friends Ren Roderick KJ7B and Michael Sterba KG7HQ who reviewed and commented on a very early draft of the section about the PSARTN.

In Zero Retries 0175 — MMDVM-TNC is (Kind of) Real, I wrote:

I plan to do a future article in Zero Retries — Explaining the Use Case for Data Over Repeater, about why such a data capability is important.

Clinton Blackmore VE6CWB recently sent me an email, which I’m reprinting here with his permission. I’m including explanations to his questions because he asked some good questions which help me frame the longer discussion that will be Part 2.

It seems like you’ve mentioned MMDVM repeaters a fair bit recently in Zero Retries — enough that it’s caught my attention.

I’m happy to wait for your forthcoming explanation — “Explaining the Use Case for Data Over Repeater” — promised in ZR 0175. In preparation for that article, I have a few questions and I hope you’ll answer address them.

It sounds like you might add data transmission as a supplement to regular repeater activities. Would it just be using the repeater for data some of the time and voice at other times? Or would you imagine having extra frequencies (or wider bandwidth) for the repeater?

The very easiest mental model for the usage of a mixed mode FM voice / MMDVM-TNC data repeater is that hundreds or thousands of mixed mode repeaters already exist and are deployed and in use…* it’s just that the “data” capability of those repeaters is “fixed formatted” into digital voice (which in the end, is transmitted as data, not conventional FM voice). Example:

• Yaesu kickstarted the widespread deployment of mixed mode repeaters in the US with their DR-1X and now their DR-2X repeaters — FM voice and System Fusion digital voice (with a little bit of data).

• Icom’s current D-Star repeaters are mixed mode — FM voice and D-Star digital voice (with a little bit of data).

• Hytera’s DMR repeaters are mixed mode — FM voice and DMR digital voice.

All of these repeaters listen on the input frequency for FM voice and (their flavor of) digital voice, and then…

• When an FM voice signal is received, the repeater transmits FM voice.

• When a digital voice signal is received, the repeater transmits digital voice.

“Data capability” on D-Star, System Fusion, DMR, and P25

It’s a completely different discussion, perhaps meriting a detailed treatment at some point in the future, but D-Star, System Fusion, DMR, and P25 digital voice modes all incorporate some limited data capabilities. Thus with the installation of a new mixed mode D-Star, System Fusion, DMR, or P25 repeater, there (could have been) a data capability “built in”. Thus you might consider this entire discussion of mixed mode repeaters (FM voice / MMDVM-TNC data) to be moot.

However, in my studies of the data capabilities of all of these systems, the data capability in those systems is an afterthought at best, or only partially (poorly) implemented or in the case of System Fusion, deliberately limited. Not to mention the data capabilities in these systems is now far behind the current state of the art, such as no use of Forward Error Correction (FEC) for data.

Thus, to add good, reasonably fast data capability to a repeater¹, Amateur Radio has to invent new technology, such as adding an MMDVM and incorporating MMDVM-TNC data, or perhaps M17 or Opulent Voice. All of these new systems have a well-implemented data capability in addition to digital voice.

If I receive enough feedback that Zero Retries readers want to read a detailed critique of the limited data capabilities of D-Star, System Fusion, DMR, and P25, let me know.

Thus, what I’m proposing with retrofitting FM voice repeaters, continuing the primary use of an FM repeater’s existing voice operations, and adding an MMDVM to support a secondary use of MMDVM-TNC data is conceptually the same as what we’ve been doing for years on hundreds or thousands of existing mixed mode repeaters.

My observation over decades of promoting the conversion of little-used FM voice repeaters for 9600 bps FSK data (and creation of new such repeaters) is that no repeater owner is willing to do so. The idea of “data repeaters” is just completely foreign and not understood, and there are only historical examples to cite such as [The Puget Sound Amateur Radio TCP/IP Network](https://www.superpacket.org/2021/03/the-puget-sound-amateur-rad io-tcpip-network-circa-1995.html). Thus a scheme like adding MMDVM and the subsequent ability to also use MMDVM-TNC data, that can (theoretically) switch seamlessly from FM voice to MMDVM-TNC data will likely be better received.

Extra frequencies is an interesting idea, especially if the repeater could receive on another input channel, possibly on another band. I’m not currently promoting wider bandwidth, though I’m definitely promoting the preservation of current (non-narrowband) channels.

Also, I’m dying to know, as transmitting data over a repeater in an infrastructure-up situation doesn’t make a lot of sense (vs. downloading files from the internet) unless you hope people skill-up and can use it in an infrastructure-down situation. What sort of data might one transmit? (I seem to recall you saying something about FLArq in this vein, too).

In an “infrastructure up” (neat descriptor!) situation, I’m positing using a repeater in data mode in the same way we currently use a repeater for voice. In “infrastructure up” situations, we certainly don’t need to use repeaters for voice communications — we have mobile phones and mobile phone networks for any conceivable voice needs.

The “broadcast files” capability I referenced is actually flamp² which stands for Fast Light Amateur Multicast Protocol. flamp is one of the fldigi suite of data modes. Literally any useful file can be distributed via flamp — maps, weather bulletins, repeater / node frequencies (code plugs), local Amateur Radio events and nets calendar, club bulletins, photos, even rudimentary web pages — pure HTML is remarkably compact. If you have a receiver on the flamp frequency, even a computer as simple as a Raspberry Pi Zero 2 W is usable for this purpose, and there’s no transmit capability required. Thus Amateur Radio clubs could use flamp to distribute information to those interested in Amateur Radio. The basic point is that with flamp, we can use Amateur Radio to promote Amateur Radio, similar to the way we use nets and HF chats to promote Amateur Radio to folks who might be listening to learn what Amateur Radio is all about. And yes, this capability is “legal” — this sort of one way information transfer is well within the category of Amateur Radio “bulletins”`.

My other question about retrofitting repeaters is, um, well, again about frequencies and slots. It seems you are suggesting we could upgrade to digital radios from analog ones using existing repeaters without leaving behind users with older tech, but how?

Ben Foght N5AMD wrote a great article on the hardware installation of an MMDVM into a repeater — How to make a MMDVM Digital Repeater. For the purposes of adding data capability (MMDVM-TNC), the MMDVM hardware merely has to get updated to add MMDVM-TNC as another mode, just like you would add another digital voice mode to an MMDVM, such as M17. Then the repeater passes FM, or digital voice, or MMDVM-TNC data. MMDVM-TNC is still experimental — it’s not currently in the main software distribution for MMDVM, so actual experience with it in the real world is a work in progress, and this is one of the many projects I have queued up for my test repeaters in N8GNJ Labs to get some real experience using MMDVM-TNC and “mixing and matching” FM voice operations and MMDVM-TNC data so I can advise from actual experience.

I could see a repeater identifying an input signal as analog FM or some digital mode … but then it surely needs to transmit an FM audio signal out. (Admittedly, there’s a couple of people with poor signals, who, if they could use digital modes with forward error correction, would probably come through way better). I assume the digital radios don’t want to listen to analog FM signals, though. I could imagine a repeater using two pairs of frequencies, one for digital radios and one for analog, and mirroring messages between them, but that sounds like extra expense and gear and getting repeater area authorities to allocate additional frequency pairs — which, which not insurmountable is not an easy upgrade.

Speaking of “repeater identifying…”, a longstanding, very widespread “missed opportunity” in Amateur Radio repeaters is that current generation repeater controllers can “ID” and transmit voice bulletins with high quality digitized voice, either recorded human or pleasant sounding artificial (AI) voices. Thus the repeater can, perhaps hourly, provide a rotating set of voice bulletins about the repeater’s various capabilities, upcoming nets, etc. Personally, after four decades now of listening to very low quality artificial voices to ID repeaters, any repeater that I have any influence on will not use such crappy voice, with the subsequent crappy impression to prospective Amateur Radio Operators.

At the moment, we don’t have “digital radios”, at least for practical use (reasonable transmit power). Thus what we’re using is conventional FM radios with data modems (formerly known as TNCs) attached. What we’re calling “data radios” are conventional FM radios with a special input, called, variously, “flat audio”, “data jack”, “9600 input”, etc. that can be used to bypass the conventional voice pre-emphasis and voice de-emphasis stages of the radio that make human voice sound good on an FM radio, but distort higher speed data “audio” that is sent through those stages.

The data modems are endlessly patient, and will happily listen to a mixed mode (FM voice / MMDVM-TNC data) repeater, and only “spring into action” when it actually decodes an MMDVM-TNC transmission.

All of this would be much easier explained in an interactive block diagram, which I intend to do eventually. I have not yet spent time learning how to instruct an AI like ChatGPT to create such diagrams, even videos, but that’s on my long to-do list.

There are endless refinements possible for mixed mode repeaters.

For one, consider the possibility of adding a secondary receiver to a mixed mode repeater. Imagine a 146.76 MHz FM repeater, with its input frequency at 146.16 MHz. This repeater gets retrofitted with an MMDVM and a secondary receiver, perhaps even on a different band like 222.01 MHz. When someone transmits FM voice on the primary input frequency, FM voice is retransmitted like normal. When someone transmits MMDVM-TNC data on 222.01 MHz, that data is routed into the MMDVM and recognized as MMDVM-TNC data, and is retransmitted as MMDVM-TNC data.

Another potential refinement for mixed mode operation is that most current repeaters can be configured to transmit a subaudible tone when the repeater transmits in FM voice mode, and a FM user radio can be configured to mute the audio unless that subaudible tone is received. A digital voice radio will generally automatically switch between FM voice and digital voice.

Part 2 Upcoming - First Principles of Mixed Use FM voice / MMDVM-TNC Data Repeaters

In Part 2 that I’ll publish in the next week or two, I’ll discuss the following ideas (“First Principles”) in more detail:

• Amateur Radio capabilities, during normal (non emergency) times is generally inferior to the capabilities of commercial (and public safety / government) Internet / cellular / satellite services. What Amateur Radio can do better than Internet / cellular / satellite is personal experimentation and training / learning with radio technology, some research, and some recreational activities.

• Why do we in Amateur Radio VHF / UHF operating want to use repeaters at all? Because repeaters allow wide-area communications by limited power / limited antenna stations over widely varying terrain. In a phrase, using a repeater “levels the playing field” for all stations, including portable radios, mobile radios, and base stations. It’s generally easier, more effective, and provides a more satisfying experience to use a repeater instead of simplex communication on VHF / UHF.

• Why do we need data capabilities in Amateur Radio at all, including repeaters? Again, personal experimentation and training / learning with radio technology, some research, and some recreational activities. Some would argue that we need to be able to use data capabilities in Amateur Radio for communications in emergency conditions that increasingly require data capability rather than voice. Example, distributing the occupancy list of a mass casualty shelter when a hurricane has wiped out terrestrial infrastructure including power and mobile networks. It just doesn’t work to try to read hundreds of names, accurately, via voice.

• Why not use (single channel, store and forward) digipeaters for data operations instead of repeaters? Generally, for the same reason we don’t use single channel store and forward (simplex) voice repeaters - the experience is poor in comparison to a realtime full duplex repeater.

• Isn’t using a repeater a “single point of failure”? Yes, but we don’t seem to consider that an issue for voice operations… and we have a lot of (redundant) repeaters to switch to if one fails. Not to mention we know how to quickly stand up a temporary or emergency repeater.

• In the 2020s, increasingly many… arguably most, repeaters have gotten quieter and quieter with less and less usage³. In past decades, we used repeaters a lot, for example, for autopatch (some repeaters were built exclusively for autopatch). But now, we don’t use repeaters nearly as much. Thus these “almost entirely quiet repeaters” aren’t widely used. Thus does it make sense to continue supporting repeaters that are barely / rarely / almost never used? Wouldn’t it be a better use of such a repeater to adapt it for data use with an MMDVM and add MMDVM-TNC data capability? Ditto for repeaters that are unused during certain times such as late evening / early morning? Computers and data communications are “patient” and can wait.

• Why do a disruptive hardware change of installing an MMDVM when a repeat owner could allow an unmodified repeater to be used with VARA FM or conventional 1200 bps AFSK packet radio, or perhaps a newer packet radio mode such as what’s possible with a NinoTNC? That’s a fair point worth discussion… but generally such operations are disruptive to normal operations on the repeater — FM voice users will hear “a lot of garbage” on the repeater when data modes are in use.

• What’s so special about MMDVM-TNC? Why is it worth considering converting a repeater to support it? I posit MMDVM-TNC is worth considering because it is a well-designed, fast data mode that can be run from an MMDVM, and MMDVMs have already been successfully implemented by existing repeaters (initially, for digital voice operations). MMDVM-TNC uses a robust modulation method, uses an advanced Forward Error Correction (FEC) system (IL2P), and offers different speed tiers, with a minimum 9600 bps, and potentially as fast as 38400 bps, exceeding the 25 kbps possible with VARA FM.

• What if we do go through a disruptive upgrade to add an MMDVM and “something better comes along” and we have to do it all over again? That’s quite possible, and it’s likely that “something better will come along” in the next few years. But that “new thing” will likely be a highly capable software modem that will connect the same way as the MMDVM, include all the existing MMDVM functions including MMDVM-TNC, and much more.

• And much more, including some examples of what we can do with a real Amateur Radio data infrastructure. Hint — look at what’s already being done with user-accessible Amateur Radio microwave networks such as HamWAN and AREDN.

Zero Retries Editor’s Note - In the aftermath of Hurricane Helene, there is water, water, everywhere, and not a drop to drink. My thanks to N2IRZ for these very good, basic points about survival in a disaster. Without the basics of life, like safe water, and especially enough safe water, the “nice to haves” like communications won’t make a difference.

Disasters generally affect infrastructure, and it universally takes at least a few days to respond to these effects. Consider [Hurricane] Helene and Western North Carolina: Major flooding significantly damaged power and water infrastructure, and washouts and fallen trees rendered virtually every road impassable, delaying emergency response for a few days. No water, no food, no power and no help on the way. How will you (literally) survive?

In NC, some local supermarkets have generators and so were able to help some residents with food and water, but this disappeared quickly. We all know that we can survive a week or two without food (uncomfortably, perhaps) but more than a day or two without drinking water can be fatal.

The point is that some simple steps can prepare you to survive a disaster. Keep at least 3 days’ worth of water and nonperishable food stashed away. Plan on at least one gallon a day per person for water, and dry or canned food that can be eaten as-is. This is above and beyond what’s in the cupboard right now. Pets have needs too, perhaps less water but more food, as pet supplies may not recover for weeks. And don’t forget some money, in smaller bills and change, since banks and ATMs, as well as credit-card processing, will be unavailable for a week or more.

Don’t forget everyday items you will need, like a can opener, some butane lighters (to light a fire), and other supplies (like plates, cups, toilet paper, soap), since the contents of your home might be completely unusable. Start today by making a list of what you think you’ll need, then start collecting it. Buy products with distant ‘best-by’ dates, sticking to the cheapest brands because you’ll hopefully never eat it, but if you have to, it’ll still keep you alive: Perrier is no better than Costco water in a disaster.

If you take absolutely essential-to-life medications, have a week’s supply (or more) stored separately, replacing it every few months.

Store it all in a plastic tub, safe from floods and somewhat physically protected from damage. Refresh the contents every year on your birthday, perhaps donating the older packages to a local food pantry.

You can probably think of other useful supplies, like a generator and fuel, a deck of cards to pass the time, or insect repellent, but the point is to start helping yourself help yourself and avoid being a victim.

In my opinion, the current “best bang for the buck” of emergency communications capability is to invest in a Starlink Mini and corresponding Starlink service, and an independent source of power for it such as the Ryobi 18V ONE+ 120-Watt Power Source With 12V Output that I discussed in Zero Retries 0169. Turn on your mobile phone’s Voice over Wi-Fi feature, and as long as the Starlink Mini has a clear view of the sky, and power (and your billing is up to date), you have a “works anywhere” mobile phone, text messaging, web browser, apps such as Facebook, weather, etc. With the Starlink Mini’s Wi-Fi, laptops, tablets, even smart televisions have Internet access, with enough bandwidth to be able to share.

But however useful, Starlink Mini and its accessories is a dedicated set of hardware, and a subscription, that may not make sense (or be affordable) to many people. Thus, there really needs to be an alternative for the folks that have “just a phone” as their communications lifeline.

In the impacted area, Hurricane Helene wiped out most terrestrial communications infrastructure — cellular, landline, and most impactful, grid (mains) power. Battery backup systems and generators are just that — backup, not primary power sources. Battery backup systems bridge a gap of a few hours, or a few days at most, of loss of grid power and restoration of grid power. Backup generators can generally carry primary loads, but require regular refueling and periodic maintenance. It’s also notable that restoration of all that destroyed communications infrastructure, especially in Western North Carolina and that region, may have to wait for roads not just to be cleared of debris, but in many cases entirely rebuilt. In some cases, construction of entirely new roads and bridges in a different area as that may be more practical, timely, and cost effective than restoration of destroyed roads and bridges. Thus “regular” communications may take weeks or months to be restored.

Apple iPhone Satellite Messaging

One of the lesser-told stories in these disasters is the availability and use of satellite communication features that were introduced in Apple’s iPhone 14 in 2022, and have since been included in all iPhones since (not just the high end “Pro” models). When this feature was introduced, it was solely for “Emergency SOS”, but Apple has now enabled satellite communications for non-emergency (text message) communications. The built-in iPhone satellite communications capability certainly isn’t as good as a dedicated satellite communications device such as a SPOT dedicated satellite communicator on the Globalstar satellite network (the same satellite system that Apple iPhones use) or a Garmin InReach (uses the Iridium satellite network). But using those networks require the (advance) purchase of a separate device, a separate (continuous) subscription fee, and they have to be with you when you really need it, and of course, having it charged up and accessible.

Thus, the device that you always have with you — your iPhone 14 (or later), that you’ll be certain to grab on the way out the door, is arguably a reasonable compromise between the lower performance of the iPhone satellite communications, and the dedicated but higher performance dedicated satellite communicators.

Starlink and T-Mobile Special Temporary Authority to Activate Satellite to Phone Capability

While the iPhone satellite communications capability depends on new radios (or enhanced capabilities of the “radio chips”) being built into the iPhone 14 models and later to access the dedicated spectrum of the Globalstar satellite network, a partnership between Starlink and T-Mobile in the US is taking a different approach to use normal cellphones to access Starlink satellites¹. This type of service is called Supplemental Coverage from Space (SCS). T-Mobile is the first² partner of SpaceX for this service³, and T-Mobile has reserved some of its terrestrial cellular spectrum — 1910–1915 MHz / 1990–1995 MHz for use on Starlink satellites that have additional radios and large, focused antennas for providing SCS. Starlink has not yet begun commercial service (nor received full permission to do so by the FCC) because not enough satellites are in orbit equipped to provide this service.

But, there are enough such “SCS” Starlink satellites in orbit to begin testing and SpaceX requested Special Temporary Authority (STA) from the FCC to begin that testing on 2024-05-01 for a period of 180 days.

But then Hurricanes Helene and Milton hit, two weeks apart, and without mobile networks being operational, there are a lot of people that do not have any communications. They have phones and perhaps ways to charge them (generators, vehicles, etc.) — just nothing to connect to.

Thus a new Special Temporary Authority (STA) for folks in those affected areas will provide at least minimal services, per this tweet (?) by SpaceX on X dated 2024-10-06:

SpaceX and @TMobile have been given emergency special temporary authority by the @FCC to enable @Starlink satellites with direct-to-cell capability to provide coverage for cell phones in the affected areas of Hurricane Helene.

The satellites have already been enabled and started broadcasting emergency alerts to cell phones on all networks in North Carolina. In addition, we may test basic texting (SMS) capabilities for most cell phones on the T-Mobile network in North Carolina.

SpaceX’s direct-to-cell constellation has not been fully deployed, so all services will be delivered on a best-effort basis.

While the service provided by this STA seems limited to T-Mobile customers, it’s available for use on any recent mobile phone, not just iPhone 14s and later, or phones from other manufacturers.

Satellite to Phone Wireless Emergency Alerts

I was unaware of an even earlier STA that Wireless Emergency Alerts via satellite was being tested:

T‑Mobile Conducts the First Ever Wireless Emergency Alert Via Satellite

September 11, 2024

T-Mobile Starlink satellite-to-smartphone technology to bring critical emergency alerts to 500,000+ square miles of land currently unreachable with earth-based cell towers.

Emergency alerts will work for everyone – even Verizon, AT&T and other wireless provider customers will receive critical emergency alerts.

BELLEVUE, Wash. — September 11, 2024 — Hurricanes, tornadoes, fires — the type of catastrophic events that often trigger a wireless emergency alert — don’t care about wireless coverage zones. Soon it won’t matter.

Today, T-Mobile (NASDAQ: TMUS) announced that it successfully sent and received — for the first time ever in the U.S. — a wireless emergency alert (WEA) via satellite. The breakthrough opens up the 500,000 square miles of lightly populated, mountainous and/or uninhabitable land across the country to critical, life-saving emergency alerts.

“This is one of those days, as the CEO of a wireless company, that makes me pause for a moment and reflect on how technology advancements and the work we’re doing is truly impacting life and death situations,” said Mike Sievert, CEO, T-Mobile.

At 5:13 PM PT on Thursday, September 5th, T-Mobile initiated a test alert for a hypothetical evacuation notice. The alert was sent 217 miles into space where it was received by one of the more than 175 Starlink direct-to-smartphone satellites currently in low earth orbit that effectively function as cell towers in space. The alert was then broadcast to a geographic area impacted by the hypothetical evacuation notice and received by a T-Mobile smartphone.

In total, it took emergency operators just seconds to queue up an emergency message and deliver that message via Starlink satellites to users on the ground.

The life-saving benefits of satellite-enabled WEAs are immense. Take the 2018 Camp Fire in Northern California as an example. The fire, which ultimately burned more than 150,000 acres, forced the evacuation of 52,000 people, destroyed 19,000 structures including most of the city of Paradise and, most devastatingly, took 86 lives, erupted in the rural Sierra Nevada mountains.

Those who lived, worked or played off the cellular network grid - relatively common in lightly populated areas with significant elevation changes — had no access to emergency alerts due to lack of wireless service coverage. The fire also took out a reported 17 cell towers on the first day of the fire and 66 total during the first two weeks of the blaze making communications — with first responders or loved ones — nearly impossible for many.

T-Mobile and Starlink, with more than 175 direct-to-smartphone satellites currently in low-earth orbit, are currently testing satellite-to-smartphone service. Additional SpaceX launches are scheduled over the coming months to add more satellites to the current constellation, further blanketing the country with wireless coverage. As that happens, T-Mobile intends to beta test the service before launching it commercially.

While it’s early days for Satellite to Phone technology, it seems clear that these new capabilities for mobile phone users will be yet another game changing communications technology, especially the Wireless Emergency Alert capability. While there is an existing, robust infrastructure for Emergency Alerts called Emergency Alert System that incorporates notifications via mobile telephone (when mobile phone service is available), broadcast television and radio, direct broadcast satellite, streaming television service, and the dedicated NOAA Weather Radio All Hazards (NWR) network of transmitters that can be received by compatible receiver units… those are all trumped by the one, primary device that everyone has readily at hand in an emergency — one’s mobile phone.

Thus this new capability provided by T-Mobile and Starlink:

[Satellite] Emergency alerts will work for everyone – even Verizon, AT&T and other wireless provider customers will receive critical emergency alerts.

… answers a long standing issue that for many folks, their mobile phone is their only communications device. If that isn’t working because mobile networks are down, they cannot receive emergency notifications. Thus this new capability of Satellite to Phone Wireless Emergency Alerts, once fully operational, seems destined to save many lives by providing reliable emergency alerts, regardless of the state of the terrestrial infrastructure.

I intend no disrespect to all the varied Emergency Communications activities that are performed within Amateur Radio, or those that perform them. The emergence of Starlink as a Broadband Internet Access system with few dependencies other than power has changed the paradigm of emergency communications. But now, but the emergence of the new Starlink Mini has profoundly changed the paradigm of emergency communications.

The photo above tells the story at a glance about how well-suited Starlink Mini is for providing emergency communications when normal communications such as cellular or consumer Internet access are unavailable. Starlink Mini is light enough and compact enough to be carried on one’s back (or in a backpack). It can be powered by any USB-C power source, including compact USB-C battery packs (for at least a few hours) or an AC to USB-C power adapter. Wi-Fi and Ethernet are built-in on the unit. It’s managed by a smartphone app. To set it up, open the app, follow the instructions for orienting it optimally (though it will likely work acceptably by laying flat if there is enough clear sky). Within a few minutes at most you are connected to the Internet at broadband speeds. It can easily be remoted to a rooftop using a simple and inexpensive power extension cord and an Ethernet cable. It will work nearly… anywhere!

The reason I bring this up is that a Zero Retries reader contacted me about an article about a “Go Box” to set up Winlink and noted “things have changed now that Starlink Mini is available”.

Disclaimer — Yes, Starlink is a subscription service, and you have to buy Starlink Mini for a few hundred dollars and keep a service plan active for one’s Starlink Mini to be ready to use at a moment’s notice. To use one’s Starlink unit for emergency communications will likely mean exceeding the “inexpensive” service tier’s maximum data transfer limit of 50 GB. Acknowledged that those are real issues now, but Starlink has exhibited considerable flexibility in adjusting its services in response to changing business conditions. It’s my (optimistic) guess is that in a declared emergency, if one asks, Starlink can temporarily waive data transfer limits or cost penalties for “excessive usage”.

Winlink, Briefly

The goal of using Winlink, of course, is to be able to send Internet email over Amateur Radio spectrum, both HF and VHF / UHF. Using Winlink used to be a bit fraught with peril because of the relatively poor data modes Amateur Radio has traditionally used for Winlink. Formerly the only good option had been the pricey and proprietary Pactor 4 modem for HF. Now there are other options for Winlink, especially VARA - FM for VHF / UHF and VARA HF for HF. The cost of a VARA license and audio adapter to use VARA FM and VARA HF are a fraction of the price of a PACTOR 4 modem, and work comparably on HF, and work great on VHF / UHF (up to 25 kbps).

Starlink Mini / Winlink Go Kit Comparison

But consider the bigger picture here in “Winlink versus Starlink Mini” as a “Go Kit” solution (in approximately the same form factor):

• Winlink is “narrowband” email, with some capability for attached files.
• Starlink Mini is a broadband Internet system, and thus can handle any Internet activity - video cameras, video conferences, viewing streamed video, file transfers, email, Voice Over IP telephone… and can do all of that for multiple client devices such as multiple laptops connected via Ethernet or Wi-Fi.
• A Winlink Go Kit is a complex assemblage of radio(s), modems, computers, software, antennas, power supplies, and integration.
• A Starlink Mini is simple by comparison - power from a USB-C source, and connect to it via Ethernet or Wi-Fi, use normal Internet systems such as web browser.
• A Winlink Go Kit can only be used by an Amateur Radio Operator who is trained / practiced in using the combination of the radio, the modem, the computer, and the software, and all of the procedures on how transmit and receive via Winlink.
• A Starlink Mini can be used by anyone; it’s effectively “unlicensed” wireless Internet. The app is easy to understand, and once it acquires the satellite constellation, it just works when you connect to it with Wi-Fi or Ethernet. The app provides status, devices connected, some management, and diagnostics including a speed test for troubleshooting and it can tell you if there’s an issue with the satellites, or obstruction.
• Winlink requires some infrastructure, especially when using VHF / UHF Radio Mail Servers (RMS).
• A Starlink Mini requires comparatively little infrastructure now (a regional Starlink Ground Station) and in the future will require practically no infrastructure through the use of inter-satellite links.

Analogy - Autopatch

I think there’s a useful analogy in Amateur Radio’s very active use, and then complete disuse, of “Autopatch” on VHF / UHF repeaters. As a new Amateur Radio Operator in the mid-1980s, one of the most popular uses of repeaters was autopatch - “automatic phone connection”. If you wanted to make a phone call from your portable or mobile VHF / UHF radio, you could easily and quickly command the repeater to connect a phone line, dial a call with touch tones from your radio, have your conversation, and then disconnect the phone line. Autopatch was an incredibly popular feature of repeaters… but no one uses autopatch any more. There’s no technical reason not to continue using autopatch; it would work as well in the mid 2020s as it did in the mid-1980s, and there’s only a minor cost issue in having a telephone line connected to a repeater for a monthly fee.

The reason that no one uses autopatch any more is because using one’s own mobile phone is so superior to using autopatch that it’s no longer even a question about using autopatch. Why would you even want to consider using autopatch?

I think that’s the situation we’re now in with Winlink, albeit at the very beginning of the situation where Starlink (Mini) is such a superior solution to the issue of emergency communications. But I believe that the conclusion will eventually be the same as with autopatch; no one will consider using Winlink because using Starlink (and other similar systems now in development) will be a far superior solution for emergency communications.

On Beyond Starlink - Mobile Phone Direct to Satellite

Not to mention… by the end of this decade, we may not even need Starlink to use at least basic satellite connectivity from our mobile phones in an emergency, thanks to:

• Apple Satellite Messaging in its newest generations of iPhones (uses Globalstar,
• SpaceX and T-Mobile providing satellite communications using normal mobile phones,
• AST Space Mobile will also provide satellite communications using normal mobile phones.

It’s amazing to me that Iridium, the one “phone works everywhere on the planet via satellite” service provider, has fallen out of the conversation versus the above developments getting lots of attention.

It’s a brave, interesting, much more communications-rich new world!

About Zero Retries

Zero Retries is an independent newsletter promoting technological innovation that is occurring in Amateur Radio, and Amateur Radio as (literally) a license to experiment with and learn about radio technology. Now in its fourth year of publication, with 1900+ subscribers. Radios are computers - with antennas! ¹

ℹ️ Visit About Zero Retries for more information about this newsletter.

📥 Subscribe to the newsletter to receive it by e-mail.

I’m an accidental Board member of the Mount Baker Amateur Radio Club and thus I’m part of the planning for future meetings of the club, so now I’m paying a bit more attention to the relevance of Amateur Radio clubs and their meetings.

Thus the following two mentions combined in my mind:

Donald Rotolo N2IRZ commenting on Zero Retries 0135 (emphasis added):

You know, when I go to the chess club, we play chess. At the bridge club, we play bridge. At the beer brewing club, we don’t brew, but we do sample and comment on members’ beers. But at the ham club, all we do is talk. A recent topic at the meeting was “how to prepare for being a SK”. Wow, that really brought the kids in.

Ham clubs need to play radio at their meetings. Solder fumes should be common. How to use a multimeter. And so on…

Combine that with this quote I’ve been citing for practically the entire life of Zero Retries:

The Universal Purpose of Ham Radio is to have fun messing around with radios.

Bob Witte K0NR

Very soon, this will be my primary metric of how useful / fun / interesting / relevant an Amateur Radio group or associated meeting is:

Do we… play with RADIOS… at the meeting?

If we aren’t playing with radios at an Amateur Radio meeting, we (Amateur Radio) are not making use of the singular, unique advantage that we have over every other technical hobby and associated meetings - we Amateur Radio Operators can play, knowledgeably, with radios - and have fun with them.

If, as N2IRZ says, we’re just talking, then we’re doing it wrong.

In Zero Retries newsletter 135, Editor Steve Stroh N8GNJ inaugurated a series of articles about the future of Amateur Radio with one titled Techies, Data Communications, and Experimentation with New Radio Technologies Could “Save” Amateur Radio which starts with:

There’s a lot happening in Amateur Radio, and especially Amateur Radio in the US. Taken at face value, there are a number of trends that are troubling, to the point that some folks seem to consider Amateur Radio a lost cause. While this discussion will at first seem like it’s “Amateur Radio Politics” (which I generally try to avoid in Zero Retries"), these preliminary points are presented in service to Zero Retries’ mission:

… promoting technological innovation that is occurring in Amateur Radio, and Amateur Radio as (literally) a license to experiment with and learn about radio technology. ¹

N8GNJ then presents and discusses these points

• The numbers of Amateur Radio Operators (not just in the US) are declining.
• The influence of the ARRL will inevitably decline as a result of disgust at the Board of Directors infighting and dues (subscription) price increases.
• The use of voice VHF / UHF repeaters are significantly reduced, year to year.
• Emergency Communications has subtly become less of a justification for Amateur Radio. Simultaneously …
  • In the 2020s and beyond, First Responders have more, and more reliable communications options than ever before.
  • Communications infrastructure available to most individuals has become more concentrated and fragile.
• COVID-19 and other factors have impacted many Amateur Radio club meetings and other reasons for Amateur Radio in-person events.
• The potential audience of younger folks to become new Amateur Radio Operators just don’t find “talking to strangers” or “operating on shortwave” to be compelling reasons to become Amateur Radio Operators.

In his final point—The “Solution” for the above? Reimagine Amateur Radio to attract techies—N8GNJ presents some potential solutions.

Read the complete article …

About Zero Retries

Zero Retries is

An independent newsletter promoting technological innovation that is occurring in Amateur Radio, and Amateur Radio as (literally) a license to experiment with and learn about radio technology. Now in its third year of publication, with 1200+ subscribers. ²

An Introduction Issue—Zero Retries 0000—was published on July 9, 2021.

The Zero Retries newsletter is currently published on Substack. In Zero Retries 135 the editor announced his intention to rehost the newsletter on another platform during 2024.

Readers can subscribe to Zero Retries at https://www.zeroretries.org/ or click the “not now” link for the newsletter index.

Editor

Steve Stroh, N8GNJ

Website

https://www.zeroretries.org/