#LoRaWAN a Smarter Oldham


#1

Im in talks with Oldham Council and Dr Andrew Robinson of Code Bug and Piface to introduce #LoRaWAN to Oldham.

if you don’t know what this is, basically its a low power, low bandwidth long range wireless connection, the hope is we can cover the whole borough meaning any device can communicate with any other in the whole town.

So a motion detector in a park can let you know when people are fly tipping, or a flood can be detected on a river, or a smoke alarm can inform family of a fire all without a paid for internet connection.

We’ll keep you informed as it progresses, but time to start thinking of some interesting ideas.


#2

Moved is to general because we could do with some input from outside the members (might be worth making an effort to make more posts publicly available)

@binkythebomb can you tell us how many nodes it would take to cover oldham if each node has a range of 6km. Assume within reason roughly line of sight so they can’t go over hills.


#3

Give me an hour, I’ll need to check the map.


#4

Are we talking Max Range or Effective Range here? Because that will have to factor into the calculation.
Min spec based on area of coverage is 24, but that does not take into consideration the landscape, which will add roughly 1/3rd to 1/2 times more nodes.


#5

effective, max is in the region of 15km!

So if we are conservative and overlap we build in redundancy


#6

Okay, so there is an effective overlap, that’s helps.

Just working on a placement overlay now, standard square grid leaves much to be desired. Probably have to look more at a triangular placement. If done right each node will have a min of 3 points of contact along the edges.


#7

Okay, to hit high ground and low ground given the ranges and to give full coverage with a generous overlap, between 35 and 54 nodes would be needed, based on a square grid reference. 54 Does give the most redundancy and will allow the signlal to completely cover the hills and a fair swath of landscape outside the borough.


#8

Sounds like a lot, The city of Amsterdam took 10, granted I have no idea how Amsterdam compares to Oldham size wise


#9

Amsterdam is 1/3 larger than Oldham, but also laregly flat and more densely populated. It’s neat and orderly, because the landscape allows it (as does infrastructure). Oldham is none of those things. We could maybe cut it down, but we are looking at the extremes of range here. I’ll have another look at the map.


#10

True, we’d do it in stages any way start with the town centre and move out, possibly looking for sponsors to buy nodes (about £500)


#11

Looking at it, I think I see why it only tok them 10. They only cover about 60% of the city. Cheating methinks.

I’m looking at border to border coverage, in which case, we’d be looking at maybe… 12, but with no overlap and a lot of missed space around the edges. If we just stick to towns that make up the borough, maybe, no more than 14. If we want a signal to pass over the hills though, we’d need a few more, just for overlap to ensure signal viability. But even at 10 we’d outnumber the amount of mobile phone masts in the town I think.


#12

10 still sounds generous! This is a roughly 6km circle from the civic, assuming we can get in there.
There would ideally need to be one to cover the uppermill valley, form one of the surrounding peaks ideally, but after that populated areas are pretty much covered, so it’s just about redundancy and covering a few awkward valleys?


#13

I suppose a lot of it is down to the height they are placed, Andrews is in Royton but doesn’t reach oldham town Center as oldham edge is in the way.


#14

Possible use : Monitored alarms on postboxes


#15

oo this sounds interesting. Only flicked through the convo, depends on the freq used for range/interference, but lower bands are good at going thru large obstacles.

I’ve had and interest + read into research about similar things in the past.


#16

Hello!
This is an RF simulation of a transmitter on the Civic centre

As far as I can see it we need somewhere high in Shaw and Failsworth to get reasonable coverage of the more densely populated areas, then Saddleworth might be patchy!

We’re stuck fairly stuck on 868Mhz – (no-one really uses the 433Mhz version)

On a side note, there’s various tools for RF simulation – splatrf seems to power most http://www.qsl.net/kd2bd/splat.html


#17

is that UK legal? Thought it was mainly used in America (assuming unlicensed)


#18

It’s legal
http://stakeholders.ofcom.org.uk/binaries/spectrum/spectrum-policy-area/spectrum-management/research-guidelines-tech-info/interface-requirements/IR_2030-june2014.pdf

(though my post wasn’t English) – what I meant to say is most people use LoraWan on 868 by default.

915Mhz is the USA only


#19

“But even at 10 we’d outnumber the amount of mobile phone masts in the town I think.”

Oldham town centre alone has 7. If you include just one location say 1/2 mile away in Clarksfield, you’re up to 12.

http://www.sitefinder.ofcom.org.uk/search


#20

Failsworth unfortunately isn’t known for hills! Would one in Limeside reach across, or on the cemetery building? It would be in Hollinwood but have the height advantage. Maybe something Crompton Moor way would provide good North coverage?