So i have been using a Mikrotik lora gateway for an outdoor GPS tracking solution.
The gateway coverage seems to be very patchy and not as wide as i expected.
People have told me that trying out a different gateway like the Lorix One would provide a better and wider coverage because the Rx threshold is -140 dbi vs the mikrotik which is about -120 dbi.
I would just like to know if anyone has done this test/comparison and if the statement is true? (What else can anyone suggest i look out for aswell?)
your suggestions would be highly appreciated.
below i have included pictures of what a network simulation tool provided of the two different gateways:
imho the coverage of a gateway is mostly determined by its antenna and the position of the antenna. AFAIK all gateways use the Semtech-chipset and there is no huge difference in sensitivity like 20dB.
Where is your antenna (inside, outside, above the roofâŚ)?
I call out Bull! Either comments from people not understanding what they are saying and what they see in real life and on console vs data sheet/gw spec⌠or someone selling something!
Reality is there are 3 principle chipsets, with 4th emerging into market, all based on SMTC Si and ref designs and with similar performance when it comes to basic RF reception/performance bar a couple of dbâ and operating temp. No viable commercial offering that I have seen has such drastically bad performance c/w the market/competition - they wouldnât be around for long otherwise.
-120 db is typical of min you will see in the TTN console (occasionally might see -121 to -124 in my experience, with each SF allowing for additional SNR⌠the other limiter, ranging from a min SNR of around -7/-8 at SF7 to -18â20 at SF12. If you see bottom end at -120 with one on console that is good typical,performance and though data sheet for another may say -138, -140 or even -143 as I have noted on occasion, you wonât see that in the console and in real life!
As @wolfp called out above given the similarity in basic performance itâs the the height, placement and type of antenna and any connecting feed that then really differentiates.
You can chose amongst the Gateways and underlying chipsets based on placement - benign indoor vs extreme temp (and weatherproofing) outdoor, power consumption, price, gw features - backhaul type(s) supported, if nicento haves like POEvsupport, support SF5/6 (not on TTN or main market LoRaWAN spec deployments yet). Iâve use and evaluated many gwâs so far and have Mikrotik, found it works well but a bit over complicated in setup and in Network management capabilities - its basically a very competent router with LoRaWAN capability grafted on, Lorix never got round the sending me the twice promised test/eval units so cant comment personally but I know folk who use them and find them very competent also. Which ever you choose you will not see huge differences in RF behaviour.
Learn more about RF coverage - three gateways with overlapping coverage = good, one gateway = misshapen coverage & a single point of failure.
Weâd be able to figure out something more useful if you hadnât blocked out the two towers names that are somewhere near Pretoria, so itâs a bit hard to tell from the terrain.
I am using a 6 dbi antenna, which is placed on a 25m tower. the area im trying to cover is about 15 km away (rural area) however, it doesnt seem like there is coverage over the entire farm.
haha you are spot on. the person who told me this was the supplier of these Lorix One gateways. Thank you for the feedback though, this is very useful!
I have bought one though and intend on testing this for clarity, so i will definitely keep you updated on the results if you like.
Thank you for this. I suppose we will have to resort to implementing an additional gateway on the farm site. I was trying to avoid having to deploy additional infrastructure for the network, but it seems to get good coverage more than one gateway will generally be required for these big farms.
apologies for blocking it out, the names dont really give indication to the tower locations so didnt think it would be useful.
Um, one gateway = patchy coverage with single point of failure, two or more, better coverage, redundancy. Wondering who gave you the idea you can cover a large rural area with a single gateway.
Perhaps you should investigate point to point mini-microwave to link the gateways back to the backhaul or local LoRaWAN server.
Youâll be surprised what can be done with Google searches.
You appear reluctant to reveal the location, consequently no one can look at the terrain, so not a serious enquiry then âŚ
All to do with location (gateway and node), antenna (gateway and node) and terrain (what is blocking the RF), there are other factors as well.
This gateway (Mikrotik) have a 6db antenna and is at 635m and it is 35+Km to Durban, clear LOS to the see, it picks up reliably nodes in Durban (recently migrated to V3 so all V2 data missing). You will notice hardly any points to the SE, this is due to a tin roof 25m away blocking the RF. I cant lift it above the roof so I have to live with it.
This gateway is at 950m (clear LOS, 360 for +50KM) and if I recall correctly only the internal 3db Mikrotik antenna, longest distance between node and gateway Distance: 105698m me travelling at 120km/h on the freeway,
This gateway is actually place indoors (Multiteck) with a 3db antenna, max range 1300m.
So terrain blocking plays a big part in area you can cover.
So adding 2 gateways at R3000 each, what % of your investment is it? What does a single tracker cost, in comparison to a gateway? I know you need to keep cost low for a good IRR.
Keeping costs low is not important at all if you end up with a product/service that is crippled by technical inadequacies. Particularly when you have received funding to create a product - you have no excuses not to do proper RnD so you can be sure of delivering the benefits your potential customers expect of your device.
In the UK, R3000 is a fraction of the value of one bovine.
Once you know enough about RF, antennas, gateway coverage, atmospheric effects, interference you will be able to confidently deliver. Otherwise you will sell single gateway solutions to farmers who will not be happy when their livestock âdisappearsâ off the map for days on end because they donât have the coverage.
At a minimum, you will be able to authoritatively advise your customers about RF coverage and why they may benefit from two or more gateways.
You should put all your efforts in to the device, itâs durability, battery life and RF coverage. Even if you have to write down the received co-ordinates on a PostIt and look it up on Google Maps, at least the purpose is working, the fancy website can come later.
âItâs unwise to pay too much, but itâs worse to pay too little. When you pay too much, you lose a little money - thatâs all. When you pay too little, you sometimes lose everything, because the thing you bought was incapable of doing the thing it was bought to do. The common law of business balance prohibits paying a little and getting a lot - it canât be done. If you deal with the lowest bidder, it is well to add something for the risk you run, and if you do that you will have enough to pay for something better.â â John Ruskin
It was never my intention to sell single gateway solutions, I was merely trying to identify whether the Rx threshold on different gateways really did make a difference based on real world applications. It seems the feedback has been the the floor is generally around - 120 dbi. Which is part of my investigation into how to provide the best solution.
I agree 100% we are currently making use of RF coverage simulation software and testing the results on the ground to understand all these variables within our enviroment.
I think the idea is always to have the data to support the confidence in our solution.
