Locator beacon idea using LoRa, comments please

Go To Last Post
35 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Morning all. I have an idea I'd like to discuss for a new open source hardware device.

 

Yesterday my cat didn't come home all day, and I was worried. He's back now, but I'd love to attach a tracker to him. Actually there are lots of things we might want to track... Tools, equipment, livestock, drones etc.

 

Problem is, the load capacity of a cat is pretty low, and to be useful the battery will need to last a long time. GPS and cellular modems are right out, and in any case would be expensive. Fortunately there are new wireless technologies that fill this need, in particular LoRa.

 

LoRa is a low power, long range wireless networking system that anyone can set up. You can install your own LoRa base stations in your house, or use commercial city wide ones. It's very low bandwidth, but you don't need much for this application.

 

So my idea is to build a tag that can be attached to a collar or a tool, about the size of the CR2032 battery. In fact it would have a 2032 coin cell in it. I estimate that transmitting once every 15 minutes on LoRa would give around 2 years battery life if carefully optimised. I'd probably look at something like an SX1272 transceiver, although a cheaper option might exist since it only needs to transmit (no RX). Some kind of micro will be needed to manage transmissions. The main difficulties will be power management (CR2032 will need capacitor backup I think, probably something specialist to fit the form factor) and designing a suitable antenna.

 

Users of the system will have their own LoRa base stations in their homes and businesses, connected to the internet. Commercial network use might be an option too. The base stations will simply listen for pings from tags, which will be nothing more than a serial number, and report them to a cloud service along with the signal strength. That won't give a precise location, but it will give an idea of roughly where something is and give the owner some confidence that it hasn't got lost or stolen, or somewhere to start looking.

 

I'd love to hear some feedback and comments on this idea. I think the main problem will be mechanical, making something that is IP68 and performs reasonably well. It might be a question of having disposable tags so that they can be sealed permanently, rather than replaceable batteries, but then you have shelf life issues or a need for an activation switch.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You might want to look at this:

 

http://metirionic.com/en/

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sure, there are commercial solutions, but they are quite physically large and heavy, and are designed to do more than is required for this simple application.

 

Actually, I think the closest thing would be that Bluetooth locator beacon scam on Kickstarter. Obviously Bluetooth is not suitable for this kind of application at all, due to power consumption and range. LoRa makes much more sense.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

perhaps just have a 433MHz id ping. 

and then a lot of receivers (and/or directional antennas), that messure RSSI.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
Sure, there are commercial solutions, but they are quite physically large and heavy

Eh?!

 

The one linked just needs a SAMR21 - you could do that in something that's barely larger than the CR2032 which powers it.

 

Obviously Bluetooth is not suitable for this kind of application at all, due to power consumption

Eh?!

 

BLE has extremely low power consumption - that is its entire raison d'etre!

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm looking at the demo kits on their web site, they are huge. Also, 2.4GHz is no good, range and penetrate are terrible even at high power. Believe me, I've done a lot of testing in this area for my work. There is a reason Zigbee on 2.4GHz failed to take off as a standard for industrial applications and smart cities - it tired to use short range radios and many many repeaters.

 

BLE's power consumption is good by Bluetooth and rechargeable consumer device standards, but it's not going to run for 2 years with useful range and polling rate from a CR2032. Have you actually looked at the datasheets and link power budgets? Even over the typical sub 10m range it's not great.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

sparrow2 wrote:

perhaps just have a 433MHz id ping. 

and then a lot of receivers (and/or directional antennas), that messure RSSI.

 

That's basically the plan. LoRa on 168/433/868/915 MHz, licence free. LoRa is cheap and the transceivers are very good in terms of sensitivity and cost, much better than older parts like the Texas CC11xx series and RFM12B. 868MHz is a good combination of range/penetration and reasonably small PCB/chip antennas.

 

To be an easy to use consumer device I'm not thinking about doing direction finding or anything like that. Just a simple receiver box, say based on a $5 Raspberry Pi board, and a single antenna. Each one would report the serial number and RSSI back to a cloud service, and would report for all devices not just the owner's.

 

Even with just one it would give an indication of when the object/pet being tracked was near by and give a distance/time graph. Lots of cat owners would love to know when their animal comes and goes when they are out. As more people in an area got them, the system would be able to see the tag moving around with a very low level of fidelity, but still giving a useful indication. There are also commercial LoRa networks that are likely to be very cheap to use. I've heard $1.50 for a transceiver and $1/year for service by 2020 for Japan.

 

2 year battery life would be reasonable for a disposable tag if it were not too expensive. It would need to be cheap because the tags will inevitably get lost or damaged. 2 years is a reasonable replacement lifecycle for both consumers and businesses.

 

The only other issue is registering tags with owners. RFID would be ideal but there might not be space for it. Perhaps a barcode printed on, but it's likely to rub off over time.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
BLE's power consumption is good by Bluetooth and rechargeable consumer device standards, but it's not going to run for 2 years with useful range and polling rate from a CR2032.

Ummm.....

 

I attended an engineering seminar hosted by Cypress for their new line of BLE enabled PS0C's and they are quoting 5+ years cell life with a 20 microampere TX power in burst mode. Unit wakes up for a few milliseconds to connect to the host, sends its payload, then goes back to sleep. I would have to find the documents I was given with this info, they are around here somewhere....

 

http://www.cypress.com/products/...

 

 

I have a couple of the demo kits here and I fired a kit up and placed the one unit in a neighbors mailbox all the way down at the end of the block, and set up the other unit to my PC in the house....worked just fine.  Moved the remote unit to a friend that lives on the next block and the comms were spot on.

 

mojo-chan wrote:
I'm looking at the demo kits on their web site, they are huge.

Well no fooling they are huge!!  the idea is to allow you to get scope probes and make connections to the thing!

 

Jim

If you want a career with a known path - become an undertaker. Dead people don't sue! - Kartman

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Perhaps take a look at how an outdoor thermometer are made (that was why I said 433MHz), because other that size it do what you want, (instead of temperature you need to send an ID).

 

I know that some geo-fencing systems without GPS, use 128KHz for pinging (I'm still here), and VHF/UHF for real data.   

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
I'm looking at the demo kits on their web site, they are huge.

They have a download for the SAMR21 XPlained Pro

 

Also, 2.4GHz is no good, range and penetrate are terrible

You set up a "mesh" of beacons - so you don't actually need long range.

 

I worked on such a project recently; we were 20m range with the SAMR21 XPlained Pro, IIRC.

 

Cats don't actually tend to wander very far

 

EDIT

 

http://metirionic.com/en/ATSAMR2...

 

Last Edited: Thu. Feb 16, 2017 - 02:19 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Cats don't actually tend to wander very far

What!!!

I often see one of our cats 500m away, but that could easy be about the limit. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There's been some interesting documentaries on the telly here recently on domestic cat behaviour - including fitting cameras to them to see what they get up to ...

 

http://www.itv.com/presscentre/e...

 

http://www.bbc.co.uk/programmes/...

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jgmdesign wrote:

I attended an engineering seminar hosted by Cypress for their new line of BLE enabled PS0C's and they are quoting 5+ years cell life with a 20 microampere TX power in burst mode. Unit wakes up for a few milliseconds to connect to the host, sends its payload, then goes back to sleep. I would have to find the documents I was given with this info, they are around here somewhere....

 

http://www.cypress.com/products/...

 

 

I have a couple of the demo kits here and I fired a kit up and placed the one unit in a neighbors mailbox all the way down at the end of the block, and set up the other unit to my PC in the house....worked just fine.  Moved the remote unit to a friend that lives on the next block and the comms were spot on.

 

That beggars belief. Are you really claiming that 20uA TX current at 2.4GHz was able to transmit over, say 1000m, from inside a Faraday cage? Using Bluetooth?

 

By the way, I followed you link, checked the recommended part for beacons, and the datasheet (http://www.cypress.com/file/2984...) says 9.2mA TX current and sub 10m range. The RX sensitivity is a mere -90dBm, compared to -156dBm for a LoRa transceiver.

 

Sorry, you must be confused about something. There is simply no way what you claim is true.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:

You set up a "mesh" of beacons - so you don't actually need long range.

 

I worked on such a project recently; we were 20m range with the SAMR21 XPlained Pro, IIRC.

 

Cats don't actually tend to wander very far

 

I don't think that will work, unfortunately. A mesh network is impractical for consumers to set up, it needs to work with a single base station on their property but allow for expansion. 20m range is too low, for low density networks you need 1000m+ which is possible with LoRa.

 

Cats actually go quite far from home. It depends where they live, but experiments attaching trackers to them have shown a radius of a kilometer or two is not unusual. It's surprising what they get up to on the other side of the cat flap!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:

jgmdesign wrote:

I attended an engineering seminar hosted by Cypress for their new line of BLE enabled PS0C's and they are quoting 5+ years cell life with a 20 microampere TX power in burst mode. Unit wakes up for a few milliseconds to connect to the host, sends its payload, then goes back to sleep. I would have to find the documents I was given with this info, they are around here somewhere....

 

http://www.cypress.com/products/...

 

 

I have a couple of the demo kits here and I fired a kit up and placed the one unit in a neighbors mailbox all the way down at the end of the block, and set up the other unit to my PC in the house....worked just fine.  Moved the remote unit to a friend that lives on the next block and the comms were spot on.

 

That beggars belief. Are you really claiming that 20uA TX current at 2.4GHz was able to transmit over, say 1000m, from inside a Faraday cage? Using Bluetooth?

 

By the way, I followed you link, checked the recommended part for beacons, and the datasheet (http://www.cypress.com/file/2984...) says 9.2mA TX current and sub 10m range. The RX sensitivity is a mere -90dBm, compared to -156dBm for a LoRa transceiver.

 

Sorry, you must be confused about something. There is simply no way what you claim is true.

Where did I say anything about a METAL mailbox(faraday cage)?  I admit I find the claim hard to believe as well, but That is what they posted as their consumption rating.  I am not confused about anything.  The applications engineer made a point of this when I made the same points you have.

 

Jim

 

EDIT:  I don't know where you get 1000m from.  I would say the distance from my home to the two mailboxes is maybe 100metres at most.

 

If you want a career with a known path - become an undertaker. Dead people don't sue! - Kartman

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB user

Last Edited: Thu. Feb 16, 2017 - 04:02 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:

 

Yep, that's about the state of the art in cat wearable technology. It's good for short term experiments where you want photos and accurate GPS tracking, but for cat servants (sorry, "owners") something the size of a coin cell and 2 year battery life would be better. You can actually get coin cell size RFID tags for collars to open those RFID cat doors, so it's a proven form factor for long term use.

 

As well as cats I think there could be a lot of other uses. People with dementia and other disabilities could carry a tag, for example. I've seen a kettle for sale in Japan that sends a text ever time it is used, to reassure children of elderly parents that they are active during the day.

 

The more I think about it the more I like it. There are privacy implications, but they are lessened by the fact that the location data will be quite general and imprecise.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Are you really claiming that 20uA TX current at 2.4GHz 

Where does 20uA Tx current come from? Isn't that 20uA average current?

 

A mesh network is impractical for consumers to set up

You didn't say this was for a consumer product - I would probably agree for that case

 

cat servants (sorry, "owners") 

Yes - they say dogs see their humans as family, but cats see their humans as staff ... !

 

People with dementia and other disabilities could carry a tag, for example

That's one of the applications for the project I mentioned earlier. There is a great deal of activity - existing products & new developments - in this area ...

 

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 You can actually get coin cell size RFID tags for collars to open those RFID cat doors

Not that I have used one, but have made HW for id tags and I don't see any reason for a battery on the cat at all.

 

If there are room to a good ant. some tags can go 50cm+ , passive.

 

And just for fun , and yes it's not according to atmel spec but take a look here.:

 

http://hackaday.com/2009/06/27/a...  

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jgmdesign wrote:
Where did I say anything about a METAL mailbox(faraday cage)?  I admit I find the claim hard to believe as well, but That is what they posted as their consumption rating.  I am not confused about anything.  The applications engineer made a point of this when I made the same points you have.

 

Do you have a link to this specific claim?

 

It sounds like when you wrote "they are quoting 5+ years cell life with a 20 microampere TX power in burst mode" you actually meant "they are quoting 5+ years cell life with a 20 microampere AVERAGE current with TX at some unspecified power and interval". That is more believable but without knowing what TX power and what RX sensitivity they have it's not very useful. My guess would be minimum TX power for a personal area network of wearable devices linked to a phone.

 

The datasheet I linked to gives more useful information.

 

Quote:
EDIT:  I don't know where you get 1000m from.  I would say the distance from my home to the two mailboxes is maybe 100metres at most.

 

Okay, but even that is improbable with a 20uA average current. What was the TX power, what was the polling rate? Note that wifi APs, mains powered, maximum allowed TX power, better sensitivity than the Cypress parts from your link and a good size antenna only get about 30m in urban environments, on a good day. You can't really assume line of sight etc.

 

Bottom line, BTLE isn't in the same league as LoRa or other sub-1GHz systems for this kind of application.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
BTLE isn't in the same league as LoRa

Absolutely.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
Bottom line, BTLE isn't in the same league as LoRa or other sub-1GHz systems for this kind of application.

Again Where did I say it was?  I purely put out another option to look into.  LoRa is indeed an entirely different animal than BLE and has a greater range....As posted by another Freak, how far does the cat go?  If within 100 metres then BLE might work, but might not depending on where the cat is. 

 

mojo-chan wrote:
Okay, but even that is improbable with a 20uA average current. What was the TX power, what was the polling rate? Note that wifi APs, mains powered, maximum allowed TX power, better sensitivity than the Cypress parts from your link and a good size antenna only get about 30m in urban environments, on a good day. You can't really assume line of sight etc.

Indeed line of sight.  Its a straight view of the mailbox from where I had the Host.  All wireless specs are line of sight to begin with.

 

Sure your Mains powered AP's are going to perform better, when you have 110/220vac from a wall outlet power is not an issue so you go B.O. on TX power.

 

As Andy said, the 20uA is average power in a very short period of time, I know bloody well a CR2032 will not last 3 years with the TX running full time.  I don't remember the total awake time, but the sleep/hibernate was in the realm of 30 seconds to a minute. 

 

I'd look for that folder right now, but I have to earn a paycheck.

 

Jim

If you want a career with a known path - become an undertaker. Dead people don't sue! - Kartman

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A network is really not suitable for location finding based on time of flight. The latency for messages passing through nodes kills you, both on outgoing and incoming pings. LoRaWAN has some sort of position reckoning built into the protocol, as I understand it.

 

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:
A network is really not suitable for location finding based on time of flight. The latency for messages passing through nodes kills you

But you don't do the timing through nodes.

 

The network part is just for communication - not the ranging part.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A good search term would be "Real Time Location Service" or "RTLS"

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

To keep the system simple and cheap, I think simple proximity via RSSI would be enough. Even just changing RSSI within a building indicates movement, which will provide reassurance.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
Actually, I think the closest thing would be that Bluetooth locator beacon scam on Kickstarter.
TrackR?

mojo-chan wrote:
Obviously Bluetooth is not suitable for this kind of application at all, due to power consumption and range.
TrackR's effective range is improved by GNSS crowd-assist by service subscribers.

Otherwise, it's educated guesses where the cat could be and walk your smart phone around.

Cats hide well; ideally Bluetooth 4 would penetrate some vegetation.

Has a water resistant case as an accessory.

Might not work if the cat is a tiger wink

TrackR

Technical Specifications

https://thetrackr.co/tech-specs

Tracking by Bluetooth 4 with GPS crowd assist from subscribers mobile phones

TrackR bravo details

Diameter: 31mm

Thickness: 3.5mm

Battery Type: CR1620

Max Battery Life: 1 year

Bluetooth Type: 4.0

Bluetooth Range: Up to 100ft [30m]

Crowd Locate Range: World Wide

Device Ringer Volume: Up to 92dB

Materials: Anodized Aluminum, Plastic

Safe for Pets: Yes

Works World Wide: Yes

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
LoRa is cheap and the transceivers are very good in terms of sensitivity and cost, ...
Inexpensive instead of cheap for LoRa has value

For Europe the Microchip RN2483 shows 14mA receive with about the same for minimal power transmit at 433MHz; that's pushing a coin cell (circular lithium polymer as an alternative?)

IIRC, LoRa has two receive windows after a transmit then the transceiver can power down.

A plus on price and cost is the Semtech LoRa stack is FOSS and it's on mbed.

Pycom makes and stocks an ESP32 micro gateway module (plug-in) for LoRa and has teamed with Canonical (embedded Linux, PC, servers)

The Pycom OEM LoRa modules are Q1 2017.

Microchip has a LoRa gateway demonstrator; that might be more for neighborhoods or a village.

 

Note : For North America the RN2903 has 2 to 3 times the transmit current as RN2483; not surprising because we need the range here and it's only at about 900MHz.

 


http://www.microchip.com/wwwproducts/en/RN2483

http://www.microchip.com/wwwproducts/en/RN2903

via

http://www.microchip.com/design-centers/wireless-connectivity/embedded-wireless/lora-technology

http://www.powerstream.com/round-li-polymer.htm

https://developer.mbed.org/teams/Semtech/

https://www.pycom.io/news/pycom-canonical-enable-lpwan-communications-via-snaps/

 

"Dare to be naïve." - Buckminster Fuller

Last Edited: Thu. Feb 16, 2017 - 08:42 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

gchapman wrote:
TrackR?

 

Yeah, and there are a few others, all using BTLE. They all have the same issues, 30m range with line of sight, extremely optimistic 1 year battery life, relatively high cost... But you do get more accurate location information since generally you will be within 5-10m of the object. But of course, getting within 5-10m of a cat when you have no idea where it is, is pretty much impossible except by luck.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

gchapman wrote:

For Europe the Microchip RN2483 shows 14mA receive with about the same for minimal power transmit at 433MHz; that's pushing a coin cell (circular lithium polymer as an alternative?)

 

Yes, it will need capacitor backup.

 

Quote:
IIRC, LoRa has two receive windows after a transmit then the transceiver can power down.

 

Yes, and it seems like most service providers will offer a cheap TX only package and a more expensive TX/RX one. A lot of products will be TX only, things like water meters where there is nothing to configure and a daily reading is enough. The cost I've seen quoted was $1/year for 50 messages a day, i.e. half hourly transmission, but I don't recall if it included RX as well.

 

Quote:
Note : For North America the RN2903 has 2 to 3 times the transmit current as RN2483; not surprising because we need the range here and it's only at about 900MHz.

 

Indeed... Although you can just transmit at lower power instead. LoRa is incredibly sensitive, well down below the noise floor.

 

A Pi Zero looks like it would make a good base station. Low cost, you can add wifi or wired LAN as desired via USB. It's got to be easy to set up. In fact I'd imagine there would be a market for a Pi Zero based LoRa base station for hobbyists if AdaFruit or someone like that made one.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
I'd probably look at something like an SX1272 transceiver, although a cheaper option might exist since it only needs to transmit (no RX). Some kind of micro will be needed to manage transmissions.
In addition to Semtech, Hope Microelectronics (HopeRF) has a LoRa transceiver.

HopeRF LoRa has Arduino and PIC16 support; this likely eases effort on the proof-of-concept.

HopeRF LoRa modules might be small enough for the first prototype; might be able to source HopeRF LoRa transceiver chips for subsequent prototypes.

mojo-chan wrote:
I think the main problem will be mechanical, making something that is IP68 and performs reasonably well.
Creating a consistent and producible case at IP68 will take significant effort and work.

You'll be going back and forth with the injection molding machine operator a few times.

A functional prototype case will ease creation of the injection molds.

Consider that some mobile phones have sealant applied during manufacture or re-work or re-furbish; these are water resistant to a few cm for about half an hour.

To recover a cat's tag, open cell cover, remove cell, reasonably bake the tag for a duration, cool to room temperature, wipe cell contacts, insert new cell, self test, self test results at base are OK, close cell cover.

heart applied to recover the cat

 


http://www.semtech.com/apps/product.php?pn=SX1272

http://www.hoperf.com/rf%5Ftransceiver/lora/

http://www.engineeringtoolbox.com/ip-ingress-protection-d_452.html

https://www.nitrokey.com/news/2016/delivery-date-nitrokey-storage-and-other-news

13.5.2016

...

... the time-critical aspect of the project is the development of the plastic casing. We have already made three revisions to the design of the casing and this week we received the latest sample of the casing. Unfortunately ...

...

http://www.powerstream.com/ppp.htm (PowerStream, Polymer Molding)

http://www.neverwet.com/ (NeverWet)

 

Edit : IP URL

 

"Dare to be naïve." - Buckminster Fuller

Last Edited: Sat. Feb 18, 2017 - 12:01 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
A Pi Zero looks like it would make a good base station. Low cost, you can add wifi or wired LAN as desired via USB. It's got to be easy to set up.

Pycom

LoPy

https://www.pycom.io/product/lopy/

...

The LoPy can act as both a LoRa Nano Gateway and a multi-bearer (LoRa, WiFi and BLE) development platform.

It is programmable with MicroPython and the Pymakr IDE for fast IoT application development, easy programming in-field and extra resilience with network failover.

...

– Can also double up as Nano LoRa gateway

...

– Espressif ESP32 chipset

...

– 868 MHz (Europe) at +14dBm maximum
– 915 MHz (North and South America, Australia and New Zealand) at +20dBm maximum

...

– Nano-gateway: Up to 22km
– Nano-gateway capacity: Up to 100 nodes

...

(certification)

– CE 0700

...

– RAM: 512KB
– External flash: 4MB

...

SHA, MD5, DES, AES

About double the RAM and 4 times the flash as a typical ESP32 module.

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I build products that are permanently submersible for a living, so I'm well aware of the challenges in creating IP68 enclosures. Metal is usually preferred but I think in this case the right plastic and the right welding should do the trick.

 

I've contacted Semtech for some evaluation kit. Having used HopeRF modules in the past I was not very impressed with quality or functionality. Maybe they are better now, I should have another look. Funnily enough I've had the Pycom guy in at work recently too, although their modules are rather expensive and bulky due to trying to do everything. Their idea of "low power" is at least an order of magnitude too high as well.

Last Edited: Tue. Sep 19, 2017 - 07:43 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Microchip has a LoRa gateway demonstrator; that might be more for neighborhoods or a village.

A new arrival at Mouser :

Seeed Studio LoRa/LoRaWAN 868MHz & 915MHz Gateways

Seeed Studio LoRa/LoRaWAN 868MHz & 915MHz Gateways

http://www.mouser.com/new/seeedstudio/seeed-lora-lorawan-kits/

Seeed Studio LoRa/LoRaWAN 868MHz & 915MHz Gateways for Raspberry Pi 3 are long-range wireless solutions to create low-power, wide area networks.

...

To build a LoRa network, 3 things are needed. A gateway, at least one Node and a local server to monitor all devices.

...

Seeed Studio LoRa Gateway for Raspberry Pi 3

...

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Electronic Design

Electronic Design

Build a LoRA Gateway with a Gumstix Module

Technology Editor Bill Wong builds a LoRA gateway using components from Gumstix, Microchip, and RisingHF.

William Wong | Sep 21, 2017

http://www.electronicdesign.com/embedded-revolution/build-lora-gateway-gumstix-module

...

Gumstix also has modular LoRA solutions, including its family of LoRA Gateways. I tried my hand at LoRA using the Overo Conduit LoRA Gateway (Fig.1) based on RisingHF’s RHF0M301 module.

...

I tested the gateway using two LoRA nodes, including the Gumstix Strata Weather Station Node and the Sodaq ExpLoRer with a built-in PC antenna (Fig. 4). The Weather Station has a barometric sensor and temperature sensor along with a GPS jack.

...

The weather station has about 50% headroom for dealing with a custom application and is essentially limited by the AVR micro.

...

The Sodaq ExpLoRer has a heftier microprocessor with more memory.

...

There is also a Gumstix carrier board for the Raspberry Pi Compute Module.

...

There are a number of LoRA gateways built around the standard Raspberry Pi, as well including one from Seeed that also uses the RisingHF RHF0M301 module.

...

If you are looking for an open source LoRA server, check out loraserver.io sponsored by Cable Labs. The Overo or Raspberry Pi could easily handle this server. There are commercial LoRA servers and combined gateways and servers, as well. These allow private LoRA networks to be created and managed. These can coexist with other LoRA networks within the same area.


Gumstix

Gumstix

LoRa

https://store.gumstix.com/catalogsearch/result/?q=Lora

Gumstix

Blog

Create a Strata Weather Station node and more with Microchip Atmega32U4

August 7, 2017 | Karen Schultz

https://www.gumstix.com/blog/geppetto-d2o/atmega32u4/

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Microchip has a LoRa gateway demonstrator; that might be more for neighborhoods or a village.

An alternative is a new arrival at Mouser that has a SAMA5 running Linux with a Semtech radio :

Mouser Electronics

Laird Technologies Sentrius RG1 LoRa-Enabled Gateway

Laird Technologies Sentrius RG1 LoRa-Enabled Gateway

http://www.mouser.com/new/laird/laird-sentrius-rg1-lora-gateway/

... private LoRaWAN (Low Power Wide Area Network) networks.

...

Based on Semtech chipset designs, The RG1 Gateway offers a LoRa range up to 10 miles and pre-loaded LoRa Packet Forwarder software, ...

...

Full Linux operating system: Kernel v4.x running on Atmel-A5 Core @ 536MHz

...

Industrial temperature range: -30ºC to 70ºC

...

"Dare to be naïve." - Buckminster Fuller