Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

We're making an open-source $30 GPS/mesh radio, would like advice

meshtastic.org

Post reply on HN
Print view View original

171–176 of 176 posts

Re: We're making an open-source $30 GPS/mesh radio, would like advice

#171

Post by l33tman »

punkgeek wrote: post #2

Hi ya'll, so we've spent a few weeks making device software and an android app to convert a $30 chinese gps/radio into a long range mesh radio for hikers. It is working fairly well now (though still an alpha) and we'd love more people to try it and/or have fun coding on it. We got about 30 users (and some devs from this hackaday article) but we'd love more users and devs to join this happy-happy project: https://hack…

The LORA stuff is pretty cool in that the radio stack (if you can call it so, it's super little code) is open source and you can get complete modules with an STM32 and the associated LoRa radio circuitry and just add power and an antenna.

Re: We're making an open-source $30 GPS/mesh radio, would like advice

#172

Post by ourlordcaffeine »

myself248 wrote: post #127
punkgeek wrote: post #35

Earlier quoted context omitted.

Oh yes, totally agree. We have a rough duty cycle spreadsheet now and we will be careful to obey the rules ;-)

That's the nice thing about 915 in the US, it's basically do-whatever. Not the case with 315/433, though! I'm a huge fan of 900MHz meshes, and have a basement full of Ricochet hardware. Starmode chat was the coolest thing in the world, but GPS's were expensive back then, so mobile nodes didn't know where they were. Stationary routers were programmed with their lat/lon when they were installed, and the MCDN mesh did g…

>Is there support for arbitrary data transport over the Lora layer

Seconding this, I am thinking about making ultra-remote seismometers that send back (highly compressed and carefully selected) readings via something like LoRa. I could use this as the transport layer and only have to worry about the sampling and processing.

(Yes - I know about the raspberry shake. It has a lot of flaws with the design which is why I'm going for the redesign)

Re: We're making an open-source $30 GPS/mesh radio, would like advice

#173

Post by airbreather »

rapjr9 wrote: post #50

Note that there are limits to how large a mesh network can scale, it's likely to start running into problems with several hundred users. You can scale larger by limiting data to short text messages, but you'll probably never be able to handle more than 10,000 users, unless you can find a way to create high bandwidth backbone links between clusters of users. There has been a lot of research on this in the wireless sen…

I have looked into this in some detail doing deployment studies for wireless HART instruments (industrial processing plant gear) which can talk to a base station or store and forward for a friend.

So depending whether you want broadcast or point to point, or in between, performance is dominated by "pinch points" which is such an obvious term that I wont describe it.

This means that a groups situation in this context could be dramatically improved by a unit or two in a balloon or drone above the group or similar.

I suspect in practice when deployed and medium sized groups use these that a "signals master" or two will have primary responsibiliy to maintain such nodes to keep effective comms up and might use kites, balloons, drones, terrain, whatever and a particular set of skills and equipment would develop.

Also, if any given unit can impose a "cost" relative to business, amount of battery used etc then self levelling and incentive to be terser with messaging can occur. Lets face it, when you are down to 3 bits a second the use of emoticons will drop off pretty quickly.

Re: We're making an open-source $30 GPS/mesh radio, would like advice

#174

Post by airbreather »

PVHSAlex wrote: post #61

15 year old High School sophomore noob here. We (crazily) are attempting to put up a CubeSat (shameless plug www.seakingspace.com). Using the Lora SX1268 433Mhz we tested the Semtech transceiver across the Catalina channel. 25+ Miles SF factors 7,9,11 at 125 BW. Worked great. We did a link budget and should work over 500KM line of sight no prob. The SX126X is the new improved version of the Lora chip. Im sure a verio…

LoRaWAN® distance world record broken. 766 km (476 miles) using 25mW transmission power. After almost 2 years, the world record of 702 km (436 miles) has been broken.Aug 14, 2019

25mW, imagine what you could do with 1/2 a watt.

Re: We're making an open-source $30 GPS/mesh radio, would like advice

#175

Post by airbreather »

punkgeek wrote: post #71
exabrial wrote: post #54

XBee it's already at an $30 and $15 price point, I think to 'disrupt' you would need to be profitable at a -10x price point. This is my opinion. I'm usually wrong. But you asked.

Thank you. This is a full circuitboard with GPS+battery+lora+CPU+oled screen for $30. So I'm not trying to compete with a $15 radio devboard that someone has to solder into their project, but more something that 'just works'. And open source, for fun, so not selling it at all ;-)

Maybe you could persuade M5Stack to do a version of hardware that suits.

Looks like maybe they do already for $40 if you combine a base and LORA unit -

https://m5stack.com/collections/m5-core

https://m5stack.com/collections/m5-module

Re: We're making an open-source $30 GPS/mesh radio, would like advice

#176

Post by ChrisMarshallNY »

punkgeek wrote: post #149
ChrisMarshallNY wrote: post #89

I’m interested, but we’ll see how it goes. This is a non-trivial endeavor. I’ve been trying to write an open-source Apple driver for goTenna units, and have been unable to get them to help me out by providing their BLE spec. That’s a shame, because I write really nice SDKs. Even if my driver was open and free, it would provide significant brand reinforcement. Win-win. I believe that this tech is important, as it can…

The current alpha works pretty well. (I'm sort of retired now, but my old job included writing software for these sorts of radios). If you'd like to play with this project we'd love your help (and we have a documented ble api). Here's an issue where various iOS types are ideating on possibly making an iOS app: https://github.com/meshtastic/Meshtastic-esp32/issues/14

Just FYI. I ordered a couple of units, and will work on a native Swift SDK for the project, once I get them (AliExpress said "20-40 days").

In the meantime, I have my work cut out for me.

Post reply on HN