An Introduction to Meshtastic(meshtastic.org) |
An Introduction to Meshtastic(meshtastic.org) |
Meshtastic and Meshcore are both cool LoRa-based mesh text messaging that operate in an no-license-required band. While this limits your transmit power, it doesn't prohibit encryption - the inverse of most ham radio rules!
Some cities have thriving communities of Meshtastic and/or Meshcore. You can look at maps of coverage to get a very general idea - in my experience, most Meshtastic nodes are NOT listed, while a good number of Meshcore nodes are.
Meshtastic treats the mesh as dynamic - clients are assumed to always be moving, so transmissions flood between different nodes that are in eachother's reach.
Meshcore has a static layer - repeaters that are assumed to be in fixed positions - and a dynamic layer - companions that move. With fixed and hopefully reliable connections between repeaters, routing paths between two users can be 'cached', which avoid the bandwidth overhead of flood routing.
You can get started with a low cost ($30) transceiver board and an SMA antenna ($10) for the ISM band of your region. Stick it in a box an mount it somewhere high up, and see if you can pick up any other nodes!
I don't know about online tools, but it should be the opposite when actually using it, as by default Meshtastic is much more chatty (and wasteful) than Meshcore.
Do people communicate to distribute prohibited anti-government propaganda or is it a network of people who otherwise be too shy to talk to each other by other means?
What is the use case?
It's primarily just an experimental system. Demonstrating that fixed infrastructure isn't actually necessary to communicate.
Beyond that, it's a mixture of HAM radio for communicating with people outside of your immediate circle, and disaster prep.
The best realistic scenario I can see for using it is after a sever weather event like hurricane, tornado, tsunami, etc. that takes out significant comms equipment. Having an ad-hoc network pop up using battery powered nodes able to setup a secure comms channel to organise aid deliveries would be a powerful tool. But existing infrastructure is resilient enough that it's not actually necessary in modern times.
Beyond that, it's probably more of an IoT type thing. Setup a bunch of nodes across a significant area of land, run machinery, sensors, etc. remotely via a self-healing mesh network.
Five mins from our house suddenly cell service from all the providers gets very spotty. If you live near the top of a hill facing the right direction, you can maybe jury rig a cellular antenna on a pole. There is legacy POTS phone service via 60 year-old copper but few use it because it's only ISDN barely faster than dial-up and >$100/mo. Otherwise, there was no option for reliable residential phone/data/text service until Starlink became available in our area a couple years ago.
So everyone in our entire area has 2M radios to communicate in emergencies because in four years we've had two fires come close enough to close our roads, been snowed in twice (without power) and a small bridge got damaged in flooding blocking vehicle access in and out for four days. We can't even see any of our neighbor's houses from our property yet several times a year we need to get extremely local information from, and coordinate with, people we'd have to hike to visit. And it's usually because something is happening which takes out local power and/or road access. But the old 2M radios have to be monitored in a real-time which feels really antiquated. So, to me, inexpensive LoRa that could enable store-and-forward messaging and conditional unattended alerts suddenly sounds very useful.
A lot of people use it just to chat with friends and family in a fun way.
Of course the preppers and privacy evangelists see it as a means to get ready for living in a hostile environment. Being fair to them, things don't look awesome in the US.
I bet a few criminals use it, but it's still very niche.
Otherwise, you buy a couple, set them up, spend a week or two sending very slow and unreliably forwarded messages that mostly amount to "hi! i have an ACME 32ABC radio! What do you have?", and then put it in a drawer or sell it on.
Just like ham radio, really.
People are very enamored with what you could theoretically do with it, but they never actually do any of it. It's a hardware fetish, it's all about building boxes with solar panels and seeing how many nodes you can light up on the map. Reminds me of another ham radio thing I never got into, "contesting".
More practically, I'm going to try it out while camping this summer. In areas with low or no cell coverage, my phone is useless or dies quickly. Throw a repeater in a tree, and hand your friends nodes.
But also you don’t build these things when you need them (it will be too late), you need to build them before you need them.
Of course if it ever becomes popular, that will quickly change. But for now it is like early IRC.
Sometimes you discuss new Meshtastic gear or setting up a router together.
Advertising services
But only if use of the network becomes popular
(Generally no one thought this would be the eventual use case for the internet)
Something like this might work?
I worked R&D on LoRa project a few years ago. Their use case was a long-range emergency communication system for workers in remote areas(no wifi, lte or LEO at that time). Now I see a bunch of applications in this field that aren't what you describe. :)
95% of it is people just doing ping? Pong! In chat.
The cheapest devices are like $10. Order one and have a go.
More info here if someone's interested: https://www.civtak.org/
Gov/DoD/SpecOps use it to maintain radio connectivity with things like the Android Tactical Assault Kit (ATAK).
The entire thing would fall over in any kind of scenario where you needed to rely on this janky mesh network as a primary means of communications.
It can be fun/useful for very out of the way things where you have a handful of people out camping, or other off-grid situations. But frankly even in those cases there are far better/established ways to keep in sync if you need to (eg: FRS).
This stuff is mostly a solution looking for a problem.
This all works great with no local SIM cards or other subscriptions or infrastructure needed.
We plan to run experiments with Reticulum when we stop for the cyclone season. Reticulum would open a lot more possibilities with both LoRa and internet-based comms. The Columba app seems to do a lot to bridge the usability gap, but work will need to be done to integrate Reticulum with our boat systems the way we have with Meshtastic (alerting, telemetry, digital switching control).
https://blog.noforeignland.com/off-grid-boat-communications-...
Sadly I sail on the other side of the world so we likely won't be connecting.
Many are using GMRS, as it provides easy access to the entire family for $10 over ten years and requires no test. But as does most UHF/VHF or line-of-sight comms, it relies heavily on repeaters.
My handle on meshtastic, LoRa, etc, is still first impression, but I know a lot is going on here, with compelling twists and alternatives in development, eg js8call?. I'm very interested, though haven't had time to learn anything yet.
Being in Florida, which is 1) a power island 2) a hurricane magnet and burgeoning tornado scape among other vulnerabilities, resilient backup comms seems more than prudent.
I've been procrastinating and distracted, but have had the idea of learning markdown and hugo, then making a Florida ham/mesh/LoRa/gmrs/etc website designed to be highly inclusive rather than exclusive, with the hopes of getting many involved.
I don't know much yet, but the whole mesh subject is objectively fascinating and promising. I went from not knowing AM/FM to ham in two weeks of study. I'm still patching and catching up, but seriously interested.
KR4KZI 73
Re: markdown and hugo, that's an excellent combo. You could easily setup a raspberry pi or $5/mo vps to serve it behind caddy/traefik/nginx and have a working multipage app (mpa) in a weekend.
Or a few minutes with AI haha
*That and literally hundreds more equal or worse.
After CopyFail (and apparently a lot more), I am thinking I might force myself to BSD for servers, and maybe even my work station. And there's a new idea! An LLM specifically trained for BSD....
It reminds me of the early internet. In the early 90s the entire list of URLs could fill a notebook. And it was my first exposure to P2P nets. Meshtastic is a bit like that where it doesn't work well until you have a large enough community of nodes and gateways.
It's like trying to get friends to use signal. I've moved to Meshcore recently and it gets me connected to the rest of the UK. But it took me 2 dedicated repeaters.
I love the meshcore homesassitant plugin, managed to setup some alerts to send out a message on a private channel every 5 minutes if I lose power as an example.
These local meshes remind me of BBS days. You have to know a few things to access it and the community is all the better for it.
Mesh topology like this is cool, and the concept is cool. Meshtastic as an organization with one of their leadership being a lawyer and being very litiguous to protect their naming scheme, not so much. They go after so many projects, ideas and other things that say they are 'Meshtastic' powered. There's a whole discord of them. It is disheartening and very jarring. They want to use the moniker of M-Powered or M-PWRD which has no identity or meaning.
From what I could see the general vibe seems to be shifting from meshtastic to meshcore.io in the past months.
They are implemented a bit differently. The chatty nature of Meshtastic works very well in small groups, or unknown area, when you need to talk a bunch of your friends scattered during a trip, to monitor your tracktors on a large field, etc..
Then you try to scale it to a larger city and it just completely breaks. Then Meshcore.io enters the picture. Every larger community that switches says the same - it's a huge reliability difference. It also comes at a cost of some discipline and more infrastructure planning (repeater nodes).
The more I play with both the more I respect both projects.
As for Reticulum, I don't see it competing in the same category at all. It has much higher aspirations, but also it seems at the moment it's much less practical and popular.
If you are going to have the setup and work of repeater nodes though doesnt something that gives actual network use between devices make more sense?
Indeed, but when talking to friends who find Meshtastic / Meshcore interesting we often also talk about Reticulum as it has some overlaps and is an interesting project.
https://www.seeedstudio.com/blog/2026/03/23/meshcore-vs-mesh...
There's a history between Tastic and Core, but it's a different one. Meshtastic doesn't scale that well in urban areas and it seems that some on the Mestastic team didn't see that problem as a priority/ignored the problem/are too stubborn. And then Meshcore is created with a different routing, works much better in practice, proving that the mesh could be much better. In countries like the UK it seems to have replaced Meshtastic in most places.
There's no mention of Meshtastic.
For me the team behind meshtastic needs some help behind their approach to APIs, the app releases frequently break that contract and they probably just need a little help in that area to improve it.
Meshcore sounds compelling to me because it has that fixed vs dynamic target approach which I suspect is more true to the real world given folks are standing up solar powered radios attached to fixed points and then trying to send messages from their phones.
Edit: I guess meshcore isn't really a real project.
Put it on your roof with a cheap solar panel meant for a security camera, and join it to your home WiFi.
Just use the little plastic case it came in as an enclosure. Cut a hole for the antenna and USB.
A slightly larger antenna will help. There are many on Amazon, and they’re cheap.
(I have tried lots of boards and this has been the best setup for me)
The edges drawn are between nodes that have been able to hear each other in the last 24 hours, based on observed traceroute packets.
(Even then, it’s only a subset of the actually-connected nodes: the map only shows nodes that have published their position on the public channel, and have set a flag that their data is okay to uplink to a server over MQTT.)
Unless an intermediate node lies and doesn't decrement and retransmits anyway.
6 degrees of separation is probably the intended design constraint, assuming there are sufficient nodes to do long range propagation it would work, so 3 bits should be enough in theory. Or passive repeaters as you suggest to go even further. But it seems in practice to be insufficient.
Perhaps this is the reason Reticulum works so well? hop limit of 255, support for any transport mechnism so a fragmented internet is still suitable for long range propagation.
Perhaps a per node hash of known recent routes to avoid flooding every single time and using flooding as a backup.
This seems like a horrible default setting or configuration. Why public channel isn't separated from a sort of control channel for those kind of station keeping messages is kind of mind boggling.
In the PNW there are two very successful meshcore meshes, cascadimesh and psmesh. The former stretches all the way from Oregon to BC and the latter focuses more on the sound.
I just switched over to the mesh core version of psmesh and instantly I was able to get chats from folk across the state. With the Meshtastic version I couldn’t see my friends nodes once we left the pub. And I never got a ping back from my tracker the next town over despite futzing with the channel settings for a couple days
LoRA is also used extensively for hobby size UAV handheld controller/ground control station to air unit controls, and in its narrower channel sizes can be very long range. The well known TBS crossfire serial bridge radio system which predates LoRA by a number of years uses a chipset that is sort of an ancestor of current-gen LoRA stuff.
But it’s the only radio-based mesh network I’ve ever “seen” anyone else on.
It's a fundamental flaw of most wireless communication. Are you referring to something more specific?
Light up MQTT-explorer and explore the default topics for a good laugh
https://reticulum.network/start.html has an overview and how to connect.
There is a manual with a lot more information on how it works and the ideas behind it at https://reticulum.network/manual/ however it's quite large and not really a user friendly guide
If you just want to play with it https://reticulum.network/manual/software.html has a list of clients and software using it.
Chit-chat is one thing.
You can more or less only communicate with people agreeing with one specific tradeoff and the majority have spoken and they want slow long range texting, so you'll only be compatible with their mesh if you're compatible with that use case.
The software works fine if you set up super high bit rate, although you will not like the short range. There is no such thing as a free lunch and something like Y2K wifi speeds will have Y2K wifi ranges and battery use LOL. Also you will not mesh with other people's infrastructure if you intentionally use a modulation scheme they don't use. Its quite capable, really. But if most people want to max out their possible range and secondarily max out their battery life, it isn't going to be very fast...
This us vs them/there must be a winner attitude that I see in both communities is really toxic and unnecessary. Look at ham radio: some people use CW, some people use SSB, some people use SSTV, some people use FT8 (but not everyone! There are still hams using other digital modes), many operators dabble in a mix of the above. There are a variety of options and nobody is pressuring other operators to use a particular mode or band.
* MeshCore has a lot more reach than Meshtastic. Often 100+ kilometers compared to just a few kilometers. Even if MT is more popular in your area, there's a good chance that MC will give you far more actual range.
* The online node maps for both are unreliable. I don't recommend relying on them for anything.
* Meshtastic uses a basic flood algorithm, up to 3 hops by default and with a hard limit of 7. Every device works as a repeater.
* MeshCore distinguishes between Companions and Repeaters. It uses flood routing by default, and attempts to establish smarter direct routes where possible. Companions are end-user devices for sending and receiving messages. Repeaters are ideally mounted high up in a static location, and they forward packets they receive. Companions normally don't act as repeaters, but can do so if needed in off-grid situations using the "off-grid repeat" setting.
* Some are concerned about whether MeshCore is open-source. The firmware contains everything important, and is fully open-source. The official companion client app is closed-source freemium. But it's simply a GUI that talks to an API over Bluetooth, TCP, or Serial. The official CLI client is open-source, and you can use any client app you want, including the popular MeshCore-open app.
And I've never spent time learning about it, but I'm under the impression Meshtastic is all about open-source and closer to ham radio philosophy, while Meshcore is backed by some for-profit organization?
https://reticulum.network/manual/whatis.html
Reticulum is an open-source, cryptography-based networking stack designed to build secure, resilient, and decentralized low-bandwidth wide-area networks using readily available hardware. It operates efficiently over various mediums like LoRa, WiFi, Ethernet, and serial connections, enabling autonomous communication that cannot be easily censored, surveilled, or controlled
That means my awful, underpowered and suboptimally placed gw tries to take part in the network - I can't even make it repeat late (prefer other repeaters), so it jusg mostly adds to the noise (and increases power use).
That's with Meshtastic
With Meshcore? I was unable to hear even one transmission, and I love in the densely populated region.
This doesn't really work, maybe as an impromptu off the grid chat platform for a forest walk, provided no one strays too far. Even 0.5W transceivers work better.
https://www.meshcoretel.ru/en/MOW/map
Maybe other maps too...
I thing Russia's main problem (not only with mesh) is that you have millions of people living in or near few cities and very few inbetween.
And those living "inbetween" typically have no money or time for things like mesh, they are struggling with simplier things.
You could also use their site planner to plan out optimal placement: https://site.meshtastic.org/
First, the biggest issue r/n is the concern that external internet will be limited to a point of no return, for this meshtastic is quite useless because to go across the border you need powerful transmitters and risk of placing and maintaining them near the border. In russia this is not only risk of going to prison but also being literally shot if border patrol/FSB overreacts. Even if you're successful bandwidth is miniscule compared to what a modern country needs to communicate internationally.
Second, due to Ukraine piggybacking on cellular networks for drone targeting/control cell service is frequently disrupted by authorities in the areas of a likely attack (it's obviously as effective as this sounds compounded by general incompetence of the government). While they cannot shut it down completely because russia still doesn't want to go back to the stone age, this concern is largely non-existent for meshtastic though. If it becomes widely popular and coverage expands, it also could be used by Ukraine as a control network, and in this case I would expect russian authorities to just jam the whole frequency range and be done with it. So the moment it becomes viable alternative is the moment it will be shut down.
Since wired internal connection inside Russia is not limited, so _for now_ there is no need in dense LoRa meshes - Meshtastic mqtt transport will work just fine.
With Meshtastic/Meshcore you are open to scenarios like "a repeater on a boat, large group scattered across large marina, nearby town and hiking areas". Or "bigger family skiing in a large resort". You can get insane distances with great reliability if you place a node strategically on the highest point around. And it costs around $100 to buy an (off the shelf) solar node that can run close to 24/7. Some folks out there have amazing use cases in the mountains.
> something that gives actual network use between devices make more sense
There are plenty of Meshcore/Meshtastic independent devices. With keyboards or touch controls and all.
So the split is just a "the development team has nothing to do with the .co.uk site, his youtube channel and discord group", not a fork.
I played with Meshcore, wasn't able to use Meshtastic.
Still wouldn't rely on this for anything serious because it's not reliable at all, but it's nice to be able to send a message across the city or to the other side of the country.
We had to create different channels for different regions (connecticut, new hampshire, massachusetts, rhode island) just to keep from having hundreds of messages in a single public channel when posts are generally about local things.
Unfortunately it doesn't fix congestion issues completely, repeaters still repeat everything (and can't listen while transmitting) so packets from connecticut still put traffic onto the entire mesh all the way to new hampshire, but at least it's better organized.
It seems like there are some ideas to help with congestion, maybe making channels region specific (repeaters can be programmed to only repeat traffic with a certain region code) but for now it's shocking to me how far I can reach with a 0.125W radio.
But yeah, it's not "reliable", that's what TCP/IP is for =) Definitely a toy network for now, and a single malicious bot on the frequency band would absolutely wreck it. Or a single high power jammer sending out noise at just the wrong frequency. Definitely a fun project for people who would do ham radio if they had any interest in taking a test...
I wouldn't use it for emergencies though, it's theoretically a backup if the cell network and internet go down but the reliability just isn't there and I suspect never will be.
Would be great if it also ran some type of Topo mapping software!
To give more info here, it uses Heltec Wireless Paper, which is a self-contained unit and just needs a battery added.
One guy associated with the project, but never a contributor to the firmware, decided to create his own vibe-coded client/app and to trademark the name of the project. The rest of the team said "no" to that and continued doing what they were doing before.
The reason Meshcore works better for texting is the different routing. It has nothing to do with the recent drama.
I'd argue at this point there's more MeshCore networks in the world than Meshtastic
It's just a personal opinion, but I really think this is not the case in reality.
Those guys in the middle of nowhere are the biggest geeks in LoRa networks. They have more practical scenarios, more to gain and better conditions for great distances.
A friend of mine lives far away from me, barely populated area, a big city in between of us. He struggled to get any network traffic, but now he uses narrow antennas to point to particular repeaters and suddenly the whole metropolis is open to him. We talk with acceptable delivery rates (I'm guessing 70%, which is actually very decent in dense area like mine). He is currently trying to expand his local network. His neighbors are less technical, but they have frequent power failures and need alternative way to reach each other.
On the other hand there is A LOT of client nodes and repeaters in my city. Many struggle to reach even a single repeater - hard to access roofs, high buildings, crowded network with plenty of conflicts. This kills motivation for many.
Population dynamics in Russia are vastly different than in the West.
Geeks don't exist or survive in the countryside.
Not only it's abject poverty, it's also the culture that penalizes anyone who sticks out.
Russian countryside is kept poor and uneducated with no opportunities other than signing a military contract. That's how Russia was able to fight its 3-day invasion of Ukraine for 4 years without doing a full scale mobilization.
Anyone who even knows what a "mesh network" is would be in a city.
That's one aspect in which Ukraine and Russia are different.
I just picked up a T1000-E to take on holiday next week, interested if I see anything in the air of in Europe.
These people should not be making a short range text solution, they should be building a low bandwidth internet extension with gateways to the real internet. Most of the information content of the internet can readily fit over 56kb lines once you strip out all the fluff. And in an emergency where you need or want a decentralized mesh network, that's more important that being able to text, who exactly?
Nevertheless IMO the mentioned demographics/use-cases want something turnkey ready, ~single button, and currently LORA based stuff is not like that, and is clumsier and needs a certain level of... tinkerer mindset, compared to a commercial license free radio system (which might have been set up with a minimal effort), where learning curve is basically volume knob and PTT button.
I have handed out simple FM radios for total tech-illiterate people (mothers, children, elderly physical workers) in an outdoor event with 0 cell coverage, and comms worked perfectly over the dozen square kilometer area. Biggest incident was a lost radio handset.
Also, DMR isn’t all about Ham Radio. It is used commercially.
Go somewhere properly remote such as the high north. There is no cell network outside of town. And the satellite coverage is spotty at best. Say you need to go look for someone. Meshtastic relays can be up and working in minutes. A chain of rescuers can spread out along a path, and remain talking to each other, as fast as they can move. Sure, radios can do this too, but long range voice radio require serious power and are still largely line-of-sight. Radio relays are an entirely other expensive thing.
Think also of remote camps (logging/planting/fishing/climbing etc). Toss a lora relay on every vehicle and every work party can talk via the app installed on their normal phones. Use GPS-enabled devices and you can passively keep track of every vehicle. Need to operate two valleys over? As the first crew deploys out they can plop down relays at key points. Years ago I setup something like this using wifi relays. It was hell. It never worked right. The range and lower power demands of lora would have been infinitely easier.
Not to say they don't fill a niche, but bandwidth and range limit it's viability to small operations; even an optimal network cant handle more than a few hundred tweets per hour.
I've done a number of projects with commercial radios operating in the 902-928MHz unlicensed band and typically we target 1-10 mi (roughly 1-20km). Elevated antennas with enough gain to get you to the legal limit (4W EIRP) can get you a heck of a lot of range, even without line of sight.
With line of sight, communication to the horizon is possible.
If you're talking about the EU 869 MHz unlicensed band used by LoraWAN, thats quite a bit different and I'm less familiar.
Something that can also use devices we already have seems like a better solution. I'm surprised BitChat has not seen more popularity. Something that combined that + a dedicated hardware mesh transmitter (for longer range when needed) and allowed adhocactual network use between devices would be pretty damn cool.
Also arent LoRA systems mostly still line of sight?
They already have it because they ready HN! Only somewhat fascitious... A few motivated individuals can provide connectivity to thousands. There's probably a few such folks in your city already!
> Something that combined that + a dedicated hardware mesh transmitter (for longer range when needed) and allowed adhocactual network use between devices would be pretty damn cool.
This is exactly how these systems tend to work in practice. Just connect your phone via Bluetooth then use the Meshtastic app. The app can function on it's own to send messages to other phones without cell service, you just won't get the range.
Never had a peep from it. 0 connected peers, always -- in every environment.
This leads me to believe that it either does not work, or that I am the only one using it (in which case it also does not work).
This is my use case. There are no other options that don't demand something like cell phone coverage (LOL no) and cost $1K/node and maybe $50/month until you give up on the project. For hobbyists this is a total non-starter. Do you have a budget for satellite telemetry, LOL no.
With meshcore its more like $100/node and $0/month forever which is somewhat more affordable than the next cheapest possible competitor.
For a scientific-ish environmental sensor that generates a couple integer data points every hour or so, its vastly more than fast enough. My project is not far enough along to discuss beyond that and I may end up scrapping it for reasons totally unrelated to networking LOL. At least the networking side was beyond trivial to set up.
Very handwavy meshcore is meshtastic plus the lessons about digipeating etc that ham radio guys (like me) figured out the hard way back in the 80s/90s. So I have zero interest in reinventing the wheel with meshtastic and reliving the problems of the 80s in the 20s. I don't think meshtastic devs looked at prior art when they initially designed it, or they would have done things quite a bit more like meshcore.
There is approximately infinite interpersonal drama in both projects with all kinds of legal wrangling and much bike shedding apparently to prevent progress. I would not be surprised if TPTB are using funded LLM bots to disrupt those projects. Its pretty bad but as long as you stay away from the communities and socializing, both projects produce good code. I suppose for people that like that sort of thing, either project would be very attractive.
For instance, sprinkling a bunch of nodes + sensors in hostile territory should allow for gathering intelligence, guiding drones, setting of fuses...
With the obvious solution of putting mesh nodes on the drones, foregoing the scattering part.
For everyone else, IoT is already there. And good old paying people to do things (the supply of dumb people never gets depleted).