Do you seriously think I would get something that simple wrong?
There are only three non overlapping channels at 2.4GHz to be correct. There are more at 5GHz which you can use if you have 802.11a.
You are forgetting about space, you don't have all the access points in the same place so they won't all interfere with each other. As I stated it depends on the shape of the building and the propagation of the signals.
History repeating itself?
In article , D.M.Chapman scribeth thus
Any idea of which make?..
Are they using radiating cables at all?..
;!/...
Problem is that the 2.4 Ghz band is a bit too narrow for today's needs it wasn't foreseen years ago and IIRC it can't be added to at either end and of course there is a lot of subscriber equipment out there that will need modifying if it was added to.
Some have 5.8 Ghz but its performance is lass then spectacular indoors whereas the band C outdoor can work very well. We have a link on the go of 17.5 miles an its 10 Meg most all the time:)...
One of the reasons I got the 3G one!
Well, I helped unload a lorry with 1200 cisco access points... so that'll be the remote end :-)
Don't think so. All Cisco access points running in lightweight mode (I'd guess this means Cisco wireless controlelers as well, but I don't know that for sure).
IIRC, the wireless controllers go some way to helping with this. You end up with loads of access points, all low power and handling small numbers of users. Users move silently between them - not in the way you normally move from one AP to another with the same SSID AIUI, this is different somehow. Anyway, this is well beyond my understanding (it's network magic innit, I do servers and storage ;-)).
These are all dual band AFAIK.
Darren
I knew Golf Course Management was dodgy.
You normally do.
How many iphones etc have 802.11a? Most are b/g/n only. And even so the raw data rate is only 54Mbps giving a real world through put of somthing over 20Mbps and that under ideal conditions.
But the "noise floor" will be very high from access points too distant to get a useable signal from but still enough to interfere. There are 12 (or
13) non-overlapping channels so a real world through put *under ideal RF conditions* of around 325Mbps, still short of the required 400Mbps...It's not a simple problem to solve, see other posts about the complex and expensive equipment required to even try to do it. 802.11a isn't a solution though stands a better chance than b/g/n as very little consumer kit supports it. So you are stuck with the 3 channels of WiFi, I seriously doubt that you can squeeze a sustained 400Mbps out of that at the density required in student accomodation.
Even if you can by controling your own kit very carefully what happens if a couple of students come along with their MiFi... Your carefully balanced system gets clobbered.
Most recent android phones have a. As do a lot of laptops and higher end tablets. There aren't many public access points though.
There are *three* non-overlapping channels at 2.4GHz at 54Mbs.
So you admit that it can be done then. I never said how much it would cost, I doubt if anyone cares. Just how you could do it. Most of the expense is to allow seamless roaming and to control intrusive signals like MiFi nodes BTW.
You would be correct if you are talking about supplying 400Mbps to one student. However if you are talking about supplying 2 Mbps to 200 students spread through 200 flats it is easy. You don't even need the expensive kit if you assume they aren't going to roam while watching live feeds and have building construction that helps. You don't see many problems getting 2Mbs from WiFi even on densely populated housing estates and student accommodation is any worse. Doing WiFi in a large lecture theatre is far more difficult but still possible.
MiFi is very low power so the AP would automatically increase their power to cope. It would only effect people within a few feet of the MiFi. That is assuming you are using the Cisco kit and have it configured to do that.
Plenty of 'practical people' at universities. I think you're generalising from seeing berks like Boris. I work in an architecture and planning department, with strong links to trades/professions. Many degrees have mandatory and assessed placements.
But I do think you're missing the point of education if you put such emphasis on doing stuff. Perhaps you need a few people about who might ask why it's being done, and done that way, and the consequences of doing it? And that to me is what further and higher education is about.
Not saying 'on-the tools' and apprenticeships don't offer that opportunity - but a dedicated teaching environment provides a good place to think in the round without commercial pressures, for example.
I take your point about rich kids. It's going to be rich kids and kids in debt from here on in. I'm far less inclined to ask people to try it and see than I used to be.
Rob
And again.
But the above is refering to 802.11a but comprehension never has been your strong point.
I still have serious doubts that you can have 2Mbps sustained across 200 users in a fairly confined area using WiFi b/g/n. That 54Mbps is a the raw data rate not what is available as user through put. User through put is about half that say 25Mbps or around ten users with 2Mbps each. Only three channels so 30 users, under ideal RF conditions. One might be able to arrange the channels/APs to reduce interference but you won't eliminate it.
Well if you actually bother to include useful stuff in what you say it would be a lot more clear.
You are still ignoring space. for example take a square room, put two APs on channel 1 in opposite corners. put two APs on channel 6 in the other two corners. Put another AP on channel 11 in the centre. Now adjust the power so they cover the minimum area needed.
So now you have five APs covering the room with no interference. Of course you can replicate the pattern for larger rooms or higher AP density.
This approximates what the cisco kit can do more or less automatically.
Do such places exist anymore?
A lot of students live in Halls of Residence. I doubt many places would have sufficient private accomodation to cater for the large number of students. Nowadays Uni Halls seem to charge 'market-level' rents and private accomodation is not a lot cheaper.
Personally I wouldn't recommend first years living anywhere but in Hall.
Hall rents vary quite a bit depending on the location and the facilities. Some Hall rents are astronomically expensive.
My son has just got a place at a University and we could only consider using the very cheapest accomodation. I am surprised that so many can afford the more expensive places.
Many years ago it was normal for unis to reserve the halls for first year students and getting a place in them after that was difficult.
In London they would be the foreign students, they always used to have an inexhaustible supply of cash, I expect its still that way.
The bars were safer and cheaper on campus near the halls too.
A lot of the demand for high quality halls tends to come from overseas students who are paying higher fees all round.
At the Uni my son is going to I would describe the majority of halls to be "unaffordable"[1].
[1] Cheapest is £3,300pa and most expensive is £10,100 (self catered). Average is £5,759. *Maximum* maintenance loan is £5,500 (not including London).
To think in 1986, a sinle room in York Uni was £21 pw not including food.
Food could be had for around a quid a meal in the college for a proper dinner. Beer was 60-80p/pint.
Living grant was just over £3k pa (fees paid seperately direct to Uni).
Indeed - fat chance in most buildings. Whether you actually need 2Mb/sec sustained is a different question.
Tends to be a non starter in practice... what works in an empty room while walk testing, falls about in a big heap the moment you fill it with nice (RF) absorbent people. Either a single 180 degree FoV AP on the long side of the room, or a 360 degree ceiling mount in the centre usually works best.
Power levels need to be high enough to cope with worst case (even if you can manage them dynamically), and that means overspill into adjacent rooms/floors.
You will get best results with 802.11n 5GHz compliant kit used throughout that also supports MiMo SDN - but generally you don't get the choice of kit that users turn up with.
yeah right...
IIRC I paid about £25 per *month* in the mid 80s for a room in a shared house. It was a bit of a dump tho.
Food £5 per week. Beer extra.
I got about £2-2.5Kpa paid directly to me (plus a small bursary).
Very good location, not that much human "clutter" to contend with..
There is a report on the Ofcom website somewhere with a 2.4 and 5 G comparison but it seemed that whilst 5G was very good in the open it wasn't that good in buildings..
IMHO a bit more spectrum is needed for such services...
Look anything's possible in denland Dontcha know;?..
The Cisco kit can dynamically adjust its self to the actual conditions. It can detect other access points and change things as required. That's why its so damn expensive. It can also do MiMo so the actual area covered by each AP can be varied.
basically the more APs you have the better it is as you use less and less power.
Its exactly the same as the mobile network, you make the cells smaller and smaller. You then have more bandwidth to each user and you can have more users.
The limit in each case is user mobility, you have to be able to switch the user between cells fast enough so they don't notice.
I know, I have complained about having no 5GHz support in many places. The UK just doesn't get it. There are a whole raft of phones, laptops and tablets that are dual band these days but public WiFi just doesn't do dual band ATM.
I know, thanks. ;-)
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.