I just redesigned one of our dorms.  It is just a simple rectangle with 20
rooms per floor 10 on each side.  I put the APs in the rooms and I put 3
per floor.  I stagger them per floor.  On the first floor I had 2 on the
West side of the rectangle and one on the East side.  Second floor has 1 on
the East side and 2 on the West side etc.  I find considerable vertical
overlap so I have to think in 3D.  I am very carefult to avoid co-channel
interferrence, per my training.  I make sure the APs are not stepping on
each other and I manually tune buildings when I see two APs that are close
using the same channel.  I have responded to performance issues by removing
an AP with positive results.

I do this stagger thing everywhere on campus and it works pretty well.  In
the example above students who may find themselves far from an AP on their
floor will have one directly below and above them and will be fine.

I have never done formal testing to determine how much of a problem
cochannel interference.  I do know it is not the end of the world if a
student has their own equipment on channels 1 6 or 11.  Equipment on any
other channel is a huge deal as it is interference at that point.

 I have heard crazy stories about colleges just letting students bring
their own wifi routers and washing their hands of it and it actually
working.  I can't imagine that.  I would think most of the low end routers
would be on channel 6 and folks would be using 40 Mhz channels and the
whole thing would just fall down.

I am comfortable with 50 people on an AP.  Ideally there would be enough
coverage so that everyone would have no less of a signal than -70 on the 5
Ghz band.  For us though the cost is too high and I have not found that it
is needed.  At this point 5 Ghz is a lost dream.  Not enough devices are
using it and those that do also do 2.4.

For surveying I just use WiFi analyzer on a tablet or my phone in a pinch.
I leave 20' whips where I have drop ceilings so I can move APs later if
need be but I have found the "survey by gut" method pretty affective.  I
guess, throw up some APs, do a quick poor-mans survey to verify and move on.

Another way I test, after the fact, is to create a test SSID.  I call it
Test, (I've aways been clever like that) and just assign it to a single
AP.  Then connect to it and do a series of speedtests until it falls down.
That gives me a sence of how far an AP goes in terms of performance rather
than signal.  It's a good test.  It is also how I know that APs from the
floor below offer plenty of singnal to get the job done.

We have done AirMagnet surveys but I find them resource intensive and not
really any more affective than my poor-man's methods. They did teach me
that I get 10-15% better coverage by dropping APs below the ceiling grid.

Good topic.







On Fri, Jan 11, 2013 at 9:50 AM, Tom O'Donnell <[email protected]> wrote:

> I was wondering what other schools have for a ratio of students to
> AP's in the residence halls, either definitely or approximately?
>
> If you have such a number, how do you count dual-band AP's?  They're
> doing more than a 2.4GHz AP, but not quite as much as two AP's.
>
> Then one last related question... Would anyone know their relative mix
> of 2.4GHz vs. 5GHz connections in residence halls?
>
> Thanks.
>
> ----------------------------------------------------------
> Tom O'Donnell
> Senior Manager of Network and Server Systems
> Information Technology Services
> University of Maine at Farmington
> (207) 778-7336
>
> **********
> Participation and subscription information for this EDUCAUSE Constituent
> Group discussion list can be found at http://www.educause.edu/groups/.
>



-- 
John Kaftan
IT Infrastructure Manager
Utica College

**********
Participation and subscription information for this EDUCAUSE Constituent Group 
discussion list can be found at http://www.educause.edu/groups/.

Reply via email to