[MLB-WIRELESS] Fitzroy Tower update

[LongHAUL]

Howdy all,

I have finally got my little modem to load up the locfinder and have added
the fitzroy tower to the node map. (CJG - Fitzroy Omni Tower)

I picked up lots of info at the pub meet before xmas - thanks to everyone
who helped out with my little scribbled diagrams. I know very little about
wireless stuff - and what I do I have picked up mainly from this group - so
please excuse (and please correct!) any mistakes I make.

Here is a summary of how things are going so far.


Basic concept:

To build and run a wireless public access point from the unused am radio
transmitter tower on my roof.

Diagram, 30K:
http://www.ebase3.com.au/fitzroytower/fitzroy_tower.pdf
(To save sending this as an attachment it is a link, 30K)

There are a bunch of immediate reasons to go for the AP on the tower instead
of just an ex galaxy antenna pointed at my nearest neighbour from the roof.
-I probably can't see anyone from my roof. There is bugger all line of sight
from regular houses / apartments around here because of the commercial
buildings getting in the way.
-I have a whopping 15m + tower on my roof and so might as well do this s*it
properly ;)
-It appeals to the l337 hax0r nerd boy in me (kidding ;)

Anyway, lets continue.

AP/aerial/network design issues:

Problem 1:
The tower is very high (15m+) - a 15m pigtail from aerial down to roof would
be very expensive and very lossy.

Possible Solutions:
1. A small self contained commerical access point in a weatherproof box up
the tower, with a very short pigtail to the aerial. Cat5 network cable from
the AP box down to the building. (best solution so far, thanks to the lads
at Puggs for coming up with that one)
2. Same as above but with a single board pc solution instead of the
commerical ap box (I think this would end up more expensive)
3. (other suggestions welcome)


Problem 2:
What self contained AP box to use?
I think that these are most of the requirements/features (in no particular
importance order)

-Can be remotely administered / upgraded (once up there its probably not
coming down again for a while :)
-Has external aerial connector, or can easily be hacked to do so.
-Supports enough wireless clients without falling over / losing too much
performance. (no idea how many we need yet, but I would say just from the
node database + 2 extras I am pulling in there might be as many as 5 people
shooting at this shortly)
-Supports whatever wireless security the group decides on (wifi?, ipsec?
(!?))
-Can be powered through cat5 cable (not really neccessary, but nice)
-Small size / weight
(other suggestions welcome)

Solution:
No idea - I have just started looking at specs on the web - but with my
limited wireless knowledge I am very undecided at this stage - if anyone has
some suggestions / tips, it would be most appreciated. A quick google search
for 'wireless access point' gave quite a few results. The prices on these
boxes looks quite alright - but I have no idea which one to go for.


Problem 3:
What sort of aerial / how do I get / make / buy it?

Solution:
Again a little shady on this one - I have had lots of possible solutions to
this.
1. Buy a commerical omni antenna (estimated cost $700+)
2. Build one myself (estimated cost $50 parts + lots of labour)
3. Hope for a lucky break and get my hands on a second hand commercial /
broken commercial antenna and repair it. (estimated cost - somewhere
inbetween the two above)
4. (other suggestions welcome)


Problem 4:
Work out how to get the aerial / ap box up the tower (!)

Solution:
I put up some photos and there was some discussion about this a while back -
apart from climbing monkeys, it seems that the most likely solution to this
is going to be a cherry picker + rigger for a few hours. Cost - no idea but
an arm and a leg I am sure.


Problem 5:
How do I make sure it is all going to work before I spend all this money and
stick it up there?

Solution:
Erm - obviously I can't. But I think one of the most expensive bits of this
whole operation is going to be the cherry picker. I am happy to go ahead and
try and get this working on the ground - if it doesn't work then as Tony
said - I am going to have a damn great little office wireless network :)

How many actual working aerial + pc combos do we have working so far?
Has anyone tested omni + lots of clients yet?
If possible it would be great to get as many as we could to one of the
monthly meets to see how this goes.

Apart from these few small(!) obstacles - there is nothing stopping this
from going ahead -
I have access and full use of the tower.
I have all the bits and pieces up to the AP + aerial point.
I have a bunch of people who live across the road and down the street who
will put up antennas immediately if I get this running.
I have a little bit of a budget to work with. (little!)
I have a bit of spare time to do stuff.
I am keen as mustard!

What I need at this stage is for all you gurus and wunderkids out there to
keep the Fitroy Tower Problem(tm) in the back of your mind as you are
pottering around building and researching all your other projects - the
problems I am having are probably a little unique because of the tower thing
and my utter lack of wireless knowledge - so all input / suggestions /
advice are most welcome.

Well thats about it - its 3:30 in the morning and I time to go to bed.

"If you build it, they will come." (I hope so :)

Micheal Hall
ebase3 Interactive
www.ebase3.com.au



--
To unsubscribe, send mail to minordomo at melbwireless.dyndns.org with a subject of 'unsubscribe melbwireless'  
Archive at: http://melbwireless.dyndns.org/cgi-bin/minorweb.pl?A=LIST&L=melbwireless
IRC at: au.austnet.org #melb-wireless