[Cu-wireless] update; thoughts on radios, experiments, parts

[David Young]

I imagine the bulkhead connectors are $5 each. I will see.

I do not understand why signals on different channels will interfere
if the feed lines are not cut to a multiple of the half-wavelength.

I think that purpose-made muxes/demuxes will be totally out of the picture
because, as you say, they will be pricey. What I think Zach has in mind
is that we hack a muxer/demuxer using splitters and the tuners on the
NICs, whose channels I am guessing we can fix. In the end, we haven't
really muxed/demuxed anything: we've mIxed the signals at one end,
and then we've fed both signals to receivers that pick out the signal
on their channel, only. Note that we are not asking for the radios on
the NICs to do anything out of the ordinary.

Dave

On Thu, Mar 28, 2002 at 05:54:30AM +0100, niteshad at whopper.de wrote:
> Dave,
> 
> 	First of all, how much are the bulkhead connectors.  Let me know and put
> me down for two.  If they're cheap, I may want to stock up on more of them,
> for experimental purposes.  
> 	Secondly, I agree that it's theoretically possible to mix two _different_
> channels of data in the same co-ax cable.  However, it may ultimately be
> cheaper just to run that second length of co-ax.  If we do try to mix signals,
> the feed lines from both antennas should each be an integer multiple of
> half-wavelength, to minimize interference between the two signals in the line. 
> Also, we'll need not just splitters, but multiplexers and de-multiplexers. 
> If we merely use splitters, we run the risk of both Lucent cards in the
> router listening to only one of the antenas (unless we can lock the driver on a
> certain channel, semi-permanently).  I have no real cost estimate, but my
> hunch is that the MUX/DEMUXes that we need will be pricey.  If it were a
> completely digital signal, which I'm not sure that it is (carrier wave), we
> might be able to build our own, but the tolerances that we'd have to hold to,
> given a 30 mW signal are pretty tight.
> 	If we want to push the envelope regarding what 802.11b is physically
> capable of, we need more theoretically oriented sources than the amateur radio
> publications that I currently have access to.  I'll call the UIUC EE Dept.
> tomorrow to see if there are any allies to be found there. 
> 
> -- 
> Sent through Global Message Exchange - http://www.gmx.net

-- 
David Young                   On the Job Consulting
dyoung at onthejob.net     Urbana, IL * (217) 278-3933