Hacker News new | past | comments | ask | show | jobs | submit login
Crazy LORA ranges today (rent-a-pilot.nl)
106 points by wienke on July 28, 2019 | hide | past | favorite | 36 comments

Some context: LoRa is a new, private and spread-spectrum modulation technique which allows sending data at extremely low data-rates to extremely long ranges. The low data-rate (down to few bytes per second) and LoRa modulation lead to very low receiver sensitivity (down to -134 dBm), which combined to an output power of +14 dBm means extremely large link budgets: up to 148 dB., what means more than 22km (13.6 miles) in LOS links and up to 2km (1.2miles) in NLOS links in urban environment (going through buildings).

From https://www.cooking-hacks.com/documentation/tutorials/extrem...

And all of these range examples are completely irrelevant to the distances you'll achieve using lora in practice. But while we're on the subject there's been lora from/to geosynchronous orbit. So I think that's the distance record: https://hackaday.com/2018/02/22/at-71572-km-you-wont-beat-th...

At VHF and up the only thing that really matters is line of sight. Everything else modulation wise is just frosting. Unless you're in a plane/balloon/etc or paying for power and location on a very tall building Lora isn't going to go any further than your 2.4 GHz wifi router.

These contrived setups for distant records (and using transient tropo ducting) are just for fun.

While I agree the sentiment that the silly range ones are just for fun, I have in practice seen good links over ranges with no line of sight.

I've personally observed links through many buildings, over hills, and so forth that easily dwarf the range of the average 2.4Ghz link.

950Mhz/860Mhz is fine, 315/433Mhz is even better for longer distance but it seems Europe/USA are mostly 950/860Mhz only.

I see your 71572km and raise you a 3.5W radio signal received on Earth from the distance of 1.2 billion km: http://www.planetary.org/blogs/emily-lakdawalla/2005/2005020...

Impressive, but the linked article didn't uses a high gain antenna array like was used to communicate with deep space probes.

They specifically wanted something with a low barrier to entry. Take a look at the antenna they used. Pretty impressive imo.

If I read that right, they used a GEO satellite as a relay, which is nothing like a low-power device, and GP's point was the asymmetrical distance record. Similarly, Hyugens experiment used a simple patch antenna at Titan (that's why it's notable), and a high-gain antenna on Earth.

Guess I should have read further in.

That's not lora modulated. My example was.

To me, this is not extremely long range, in comparison to a Pactor modem over HF, which can reach over 1000 miles, as I've seen on a sailboat passage from Hawaii to California. It requires more power - 150 watts or so, if I recall correctly, and bit rates can be low as a few bits (not bytes) per second with poor propagation. Being able to stay in contact with family in the middle of the Pacific Ocean is pretty awesome.

It's extremely long range for low-power devices. LORA is mostly used for sensors, trackers and other miniature, usually battery/solar powered devices that need to run for months or sometimes years unmaintained.

Also, very small antennas, especially compared to typical HF antennas (sometimes the entire 60+ ft backstay on a sailboat, for example)

For those interested in playing around PyCom (MicroPython devices) are great. As an example the LoPy 4 supports LoRa, Sigfox, Bluetooth and Wifi.


I got one of those but only having one. I can think of some uses for it if I had two but with only one, couldn't do much with it.

Keep in mind that's not exactly true. There are gateways that can be used such as "The Things Network": https://www.thethingsnetwork.org/

150 watts is 5000x more power than what LORA requires, which is why this is so impressive.

Right but this is for edge devices that can be battery powered. You can't use 150 W.

Not just battery powered but battery powered for months+. The way that receive windows are structured into the protocol it allows the device to go into deep sleep between windows.

"Not just battery powered but battery powered for months+"

True, and solar powered too, that is, put one on a pole and forget about it; not even the cost of changing the battery, just wiping the solar cell now and then if wind and rain don't do their job. The rechargeable one can be a supercap or an oversized cell so that when time will inevitably degrade its chemistry to an unusable state, the sensor will probably have ended its planned use life already.

I believe the selling point is that it can run on battery for years, 10+ years sometimes limited by battery's shelf life time.

You also have to have appropriate licensure at that power output.

503'ing here for me, so here's an cache link: http://webcache.googleusercontent.com/search?q=cache:http://...

Also, I've been part of the group looking at LoRa at Lancaster University for some time, and am currently working on part of a commercial deployment too. Here are a few papers published on the topic by Dr. Martin Bor, and occasionally myself (all the top ones are LoRa): https://scholar.google.co.uk/citations?user=WhN1gGwAAAAJ&hl=...

I'm soldering a a LoRA-based board right now. During design process we calculated the power budget and found out the power consumption is so absurdly low the main factor is battery shelf life.

We live in the future.

I’ve done Long Island to Virginia with 3W on a portable at 146 MHz, but the range being seen on 800-900 MHz around Netherlands is quite amazing.

I would agree it must be tropospheric ducting; I’m not aware of any other propagation mode that could be at work here.

That's impressive. It must have felt pretty amazing!

There was a guy in one of the FM DXing groups on Facebook recently who posted a video of a French-language Canadian FM broadcast station he picked up from the Florida panhandle.

Agreed. Tropo ducting is the only propagation that seems feasible for microwave. Fun!

LoRa has become the tool of choice for the high altitude balloon community. At least in the UK, you're restricted to 10mW in the 434MHz band when airborne. People have been pretty successful at getting (very) slow-scan video from balloons at 30km altitude.

https://github.com/projecthorus/wenet is an example of using a RFM98W lora module with a raspberry pi/RTLSDR reciever (albeit at 115kbaud in FSK mode, but 50mW of tx power gets ranges of over 100km)

That is freaking interesting, 115kbps may seem low compared to usual broadband connections, but it's a lot in the embedded world. Also it would be more than enough to get very high quality voice communications. Codec2 which already does wonders with a fraction of a fraction of a fraction of that rate comes to mind. How long before someone puts LoRa encrypted walkie talkies on the market?

never thought LoRa can transmit video, isn't the bandwidth too low for that, I can not even imagine 320x240 video/mjpg using LoRA.

“Slow scan” TV is... slow.

> It is a much slower method of still picture transmission, usually taking from about eight seconds to a couple of minutes, depending on the mode used, to transmit one image frame.


Is it me or maybe you also have doubds about LoRaWAN use cases? (LoRa is physical layer, way of encoding bits with rf, but you need protocol. LoRaWAN is open standard, which is being implemented and used by TTN community)

So for me main drawback of LoRaWAN is limited number of ACKed packets (gateway which can talk to thousands devices can use only 1% of airtime (simplified) and you have to divide it by number of users https://www.thethingsnetwork.org/docs/lorawan/duty-cycle.htm...), so you are allowed to have 10 packets per day. Any other data can be send in way called " send and pray"...

Would you build industrial device for telemetry while you can not be sure about your data? For example energy monitoring with 15 min periods. You can't be sure that your data arrives.

So ranges presented here is matter of luck but not reliable link.

Maybe ISM band limitations should adopt to LoRa which is more resistant to errors.

Or protocol should be changed (symphonylink has protocol for unlimited ACKed frames, OTA etc)

Just my 2 cents.

That article should probably start with a couple of sentences answering "What is LORA?"

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact