A distance record for LoRa transmission has been set that you probably won’t be able to beat. Pack up your gear and go home, nothing more to achieve here. At a superficial reading having a figure of 71,572 km (44,473 miles) seems an impossible figure for one of the little LoRa radio modules many of us have hooked up to our microcontrollers, but the story isn’t quite what you’d expect and contains within it some extremely interesting use of technology.
So the folks at Outernet have sent data over LoRa for that incredible distance, but they did so not through the little ISM band modules we’re used to but over a suitably powerful Ku-band uplink to a geostationary satellite. They are also not using the LoRaWAN protocols of the earthbound systems, but simply the LoRa modulation scheme. So it’s not directly comparable to terrestrial records such as the 702 km we reported on last year, and they are the first to admit that.
Where their achievement becomes especially interesting though is in their choice of receiver. We are all used to Ku-band receivers, you may even have one on your house somewhere for satellite TV. It will probably involve a parabolic dish with a narrow beam width and an LNB whose horn antenna is placed at its focus. It would have required some skill and effort to set up, because it has to be pointed very carefully at the satellite’s position in the sky. Outernet’s mission of delivering an information service with the lowest possible barrier to entry precludes the extra expense of shipping a dish and providing trained staff to align it, so they take a very different approach. Their receiver uses either an LNB horn or a small patch antenna pointing at the satellite, with none of the dishes or phased arrays you might be used to in a Ku-band installation.
You might wonder how such a receiver could possibly work with such a meagre antenna, but the secret lies in LoRa’s relatively tiny bandwidth as well as the resistance to co-channel interference that is a built-in feature of the LoRa modulation scheme. Even though the receiver will be illuminated by multiple satellites at once it is able to retrieve the signal and achieve a 30 kb/s data rate that they hope with technical refinements to increase to 100 kb/s. This rate will be enough over which to push an SD video stream to name just one of the several examples of the type of content they hope to deliver.
It’s likely that the average Hackaday reader will not be hiring satellite uplink time upon which to place their LoRa traffic. But this story does provide a demonstration of LoRa’s impressive capabilities, and will make us look upon our humble LNBs with new eyes.
Like in so many other things, the ability to write a valid big check helps in record setting attempts.
I fed that into google translate and got. “You can do cool things with lots of money”. Which makes “get lots of money” a goal rather than a dismissive comment.
Get lots of money, check! I’m working on it. Can I borrow a fiver?
In fact, if everybody lent me a fiver I’d be able to generate returns for older lendees through revenue paid by new lendees. I could make everyone rich.
Please lend me a fiver to: P.O. Box Retire to the Bahamas 2018.
Nice try Ponzi. You won’t get me. I’ve created a legitimate business. I’m selling money trees for $100. Now to guarantee this isn’t a scam I’m only selling established trees that are flowering and already come with 5x $10 notes on them.
Uh huh. I’m having none of THAT scam. On the other hand, can I interest you in a new cryptocurrency?
I put my fiver in the CD tray, should get to you soon over the interthingy.
Now that is strange, my teleport CD project has started working, fivers coming out, anyone want to invest.
Pikers! I am launching a pamphlet in which I describe a sure fire way to win at Craps. Only $50 and you can win that back in 90 seconds.
Come up with a coherent and believable plan for something cool, assemble a team with the skills to make it happen, and start a crowdfunding campaign.
Articulate your plan. Show the research you’ve done that makes you think you can do the cool thing. Explain which parts will be open-sourced or otherwise returned to the community when you’re done.
If it sounds like you and your team might have a reasonable chance of achieving that goal, I’ll throw you more than a fiver.
That’s nice but, my method sidesteps all the troublesome R&D and product. Cuts straight to the UNTOLD RICHES! Don’t you like untold riches? 2878% Return guaranteed! Ask any Nigerian Prince.
Or we can just work together. You give me 5. I give you 10, which only costs me 5 and you gave me a five. Then you give me 15 which only costs you 5, but wait. I GAVE you 5 more than you gave me, so it costs you nothing! Just keep going and we can make a hellofa living.
If there’s a repeater involved, shouldn’t we talk about 35,786 km record? And mention the 90 watt amplifier on the satellite? Less clickbaity but still pretty freaking impressive.
Yeah, I’d be more impressed had it been passive EME (Moon bounce) instead of using a commercial satellite transponder like to the ones anyone gets a zillion 60Mbps HD TV channels with a 20 bucks receiver…
Sounds like you are challenging hams to do an EME bounce with LoRa modulating a frequency of choice… likely to refit a home station that is already set up on an EME for lots of ERP and a low noise preamp to listen to for the echo. Lots of variations on this. But it won’t be a low power gig with a passive repeater like the moon.
90w on 10Ghz plus a giant dish like that is more than the necessary to Moon-Bounce. I would like to know if they tried this.
Now I wonder if you can bounce off something even further, like Mars.
Check out the attempt from Andreas Spiess:
https://www.thethingsnetwork.org/community/Bern/post/andreas-spiess-cut-the-edge-201km-connect-to-the-ttn-gw-weissenstein
Sounds a little fishy, considering the circumference of the earth (measuring along the equator) is only 40,075 km…
I don’t know what you mean by this, Geostationary satellites are roughly 36,000km away in orbit. So figure in a round trip and you easily have 71,572km.
Maybe it is a matter of undercaffeination between chair an keyboard, but why is the above better than:
https://www.youtube.com/watch?v=adhWIo-7gr4
So they claim the longest LoRa distance record without actually using any Lora equipment and only the modulation scheme…
So if they use BPSK or QPSK modulation it would have been the logest distance record of NB-IOT?
Call me not impressed, they just use the LoRa buzzword.
LoRa and LoRaWAN are not the same thing. Chips like the SX1276 family contain a LoRa modem, which use a chirp spread spectrum modulation (which is no more efficient than FSK for the same bit rate, but makes very low bit rates accessible by being tolerant of frequency error) and specific forward error correction and packet format configurations. LoRaWAN is a network protocol intended for use on top of LoRa radio modems.
Actually, it looks like they’re using the SX1280 Lora chip, so this is certainly “Lora equipment”. From what I can see, the 2.4GHz signal is then upconverted into the Ku band (and later downconverted, using an LNB). Yes, “only the modulation scheme”, but Lora *IS* the modulation scheme :)
If I understand correctly, the LoRa and LoRaWAN protocols are multiplexing and not modulation. Though technically multiplexing is a form of digital modulation I guess where spread spectrum is included. If someone knows a better way to explain, please clarify.
My guess is modulation is performed first, then multiplexing… so technically more a multiplexing method and in this case more an embedded system unless there is a software coder/decoder algorithm coded somewhere.
I don’t see where they say how they got the satellite bandwith to use for this. Was it just slipped into a guard band somewhere, or did they actually pay for it? There is (or, at least used to be) a lot of low bandwidth traffic on geosynch satellites that is not supposed to be there.
There’s more about this new Ku-Band service here:
https://store.outernet.is/blogs/the-official-outernet-blog/world-record
And here:
http://forums.outernet.is/t/lora-world-record-71-572km-to-space-and-back/4904
Specifically the cost is addressed here (it seems they’re leasing the bandwidth and the costs are recovered through sales through distributors, hardware and (maybe?) services bandwidth.
http://forums.outernet.is/t/lora-world-record-71-572km-to-space-and-back/4904/14
Exerpting from the link immediately above:
Syed 2018-02-17 16:42:19 UTC #9
The market rate for one megahertz of Ku is about $5000. That’s leasing just one or two MHz at a time. It’s much lower when acquiring a full transponder. We advise a channel that is 1.6 MHz, which means a 2 MHz lease. And then another couple thousand for equipment hosting and uplinking at the teleport. For roughly $12,000 per month, we can get as high as 40 kbps with standard beams. This is using just an LNB as an antenna. 40kbps is a streaming audio service alongside file delivery.
Not supposed to be there?!
This isn’t an analog transponder on fltsatcom, I am interested what you mean, need more info.
I sincerely wish I had a way to get some low bandwidth satcom going beyond LEO amateur birds where I get 5min on a great pass maybe an opportunity for a few moments twice a day with a small footprint for real comms.
https://www.sony.net/SonyInfo/technology/stories/ELTRES/
Sony LPWA – 321 kilometers
Enjoy :))))