Hi, first topic/post but I’ve been enjoying the setup phase here in a very rural location with lots of inconvenient hills. I’ve got two gateway’s running 24/7, one at 217m most of the way up a hill and one down in a valley, so I’m looking to map out reception.
I’ve read the TTN Mapper FAQ and acknowledge the wisdom of worst case scenario use of SF7, but the reality is that stops reception dead in its tracks with the wide variety of solid stone buildings we have. Particularly if my device just has a small helical antenna. Or I drive around the back of a hill.
I can drive around slowly with something set to SF9 or 10 with a more representative stubby aerial or even an external magmount whip antenna or put what we would imagine to be a typical outdoor config for more remote parts (IP67 enclosure with stubby on the outside) on the top of the car.
I’m not looking to get some insanely accurate results, just enough so we can setup some devices without having to constantly return to change them - much as I love hill walking.
It did occur to me that I could code a device so that if I don’t get a ping on TTNMapper app that I can increase the SF at a location until I do.
As I’m doing this mapping exercise with a purpose, I’m only transmitting one byte to keep the tx time to a minimum.
Any pointers or experiences to make my war-drive as representative as possible would be much appreciated.
Thanks Nick for starting this topic on rural best practices. Formulating them can be a very useful exercise for us. I am perhaps living in a similar environment as the one you describe. Running hills up and down can be good physical exercise, but I think we should be systematic and develop procedures which can be applied in many environments.
The purpose of mapping is to achieve best possible coverage. I would like to learn how to do that.
I’ll have to draw from our local experience. Our village is also located in a scarcely populated rural area with quite a number of hills. Our national organizations may differ from what you have in your country, but maybe a reasonable number of individual cases could be a good starting point in trying to generalize.
You are calling the hills ‘inconvenient’, but are they always a challenge or could they be an asset? We do have households on top of every hill which makes it convenient to place gateways on some of them. That could be a good way to ensure visual contact with the households lower down? - I would guess that one good practice could be to map the hills of the area to be covered by the LoRa community.
Mapping the relevant buildings is another step to be taken. In a small village, of let’s say 40-60 buildings it is something which can be achieved manually. We collected Lat/Lon/Alt information of every major building of the village with the help of a handheld GPS device.
For bigger villages and larger areas this can become quite a burden. The national statistics bureau called Statistics Finland maintains data for every 100 m x 100 m square of our country. This data describing active households, cottages and company buildings can be purchased for a reasonable amount of money.
With GIS tools this information can be portrayed on the free maps of the National Land Survey of Finland. This gives us an opportunity to collect the above Lat/Lon/Alt information from the geodata of NLS.
I’ll stop here. Let’s see what kind of following your topic will gain. I wish us good luck.
RMO has a number of interesting features and it can be adapted for LoRaWan developers. Note that the tool was mentioned in 2017 together with some other utilities in the topic Radio network planning for LoRa, but time has passed since then.
The hills round here are inconvenient because there are rather a lot of them resulting in lots of little valleys and nothing is on the top of them, not even lost sheep, due to the 100+mph winds we get in winter - so no gateways where it would make sense.
I had a quick run round with a device on SF7B125 with a PCB mounted helical antenna and then again with it set to SF10 to get a feel for what might make sense. I need to walk down in the village shortly so I’ll be trying that on SF7 but overall, I suspect a proper antenna on the mote may be needed to see some indicative results before moving on to some test deployments.
i can recommend seriosly radio mobile (link above).
let me add here some sample calculations of my office gateway and of a very good gateway in my city. each time a TTNMAPPER foodprint and my radio mobile calculation. I also add my personal config how i configure radio mobile for lorawan calculations (which are nearly the same as in reality).
Hi @hilltronic I can confirm some of your coverage as when I drove up to Hamburg to collect my son from DESY on the west side back in Sept I was running several gps trackers as posted in other threads at the time. As I came up the A1 from Bremen I think I got a hit through your gw at the top end then several hits whilst queuing in traffic for the tunnel on the A7 as ever broadly similar on way back down next day with less hits on the A7 as moving quicker and wrong side of carriage way. Got some hits that night around the hotel I was at but couldn’t check which gw…there was one I think closer to the Notkestrasse/DESY area My phone lost power and so Inlost gps guidance as I came close to Hamburg and ended up going wrong way for a time into words city centre before diverting out and up over the very high viaduct to rejoin the A7. Again multiple hits due to extreme height but didn’t get to investigate if your or other GW’s…likely yours
I’ve only just got started with mapping - in effect the first couple of attempts were mapping out how to do the mapping!
I’ve done some preliminary calculations with the Radio Mobile Online link you provided, thank you.
I’ve now got a small mag mount for the car for me to test to see how representative it is to a stubby aerial on a static node. I also need to put a decent aerial on my other gateway to get the most out of driving slowly around the area!
Mid-term, I’m going to hook up a GPS to a device so I can run TTN Mapper any time the car is on the move automagically as I suspect it will take some time for me to accumulate some decent coverage mapping.
Yesterday I ordered a Adeunis Field Test Device LoRaWAN 868 MHz (OTAA), as a complement to the software toolbox discussed here. Do you have any user experience from this kind of tools?
I’ve had to do some real work this last fortnight but I now have all the bits for a second test drive.
That would be a very slow process for me - so I’m using ABP initially on SF9 and once I can see some gaps, amend accordingly. I know some would prefer SF7 for TTNMapper, but that totally fails as soon as I go behind a solid stone building - which is pretty much all of them around here.