Towertalk
[Top] [All Lists]

[TowerTalk] arrays was Re: tower resonances

To: W2RU - Bud Hippisley <W2RU@frontiernet.net>
Subject: [TowerTalk] arrays was Re: tower resonances
From: jimlux <jimlux@earthlink.net>
Date: Wed, 11 Feb 2009 06:40:50 -0800
List-post: <towertalk@contesting.com">mailto:towertalk@contesting.com>
W2RU - Bud Hippisley wrote:
> On Feb 11, 2009, at 12:54 AM, jimlux wrote:
> 
>> I think if you're serious about phased arrays, then the few hundred
>> bucks for the VNA probably is the least of your problems.... in the
>> HF world, most users of arrays have large budgets...so they can  
>> spend the time and
>> money on careful measurements of whatever is needed, including running
>> far field patterns using slick stuff like RELEDOP.
> 
> 
> So if the choice is one of buying either a VNA or an airplane capable  
> of dragging a "sense" antenna around, you make the decision a lot  
> easier for me, Jim....:-)
> 
> 

Laugh if you must, but I bought a 1/3 scale R/C powered parachute for 
exactly this purpose.

http://www.luxfamily.com/jimlux/robot/ppc/firstflight.htm

The folks in the UK at DERA use a tethered blimp for this kind of 
application.   SRI uses a helicopter for RELEDOP.


The real challenge, by the way, is transmit, not receive.  For receive, 
you don't have to handle lots of power, so all digital approaches are a 
fairly decent way to do the job AND there's huge literature on adaptive 
algorithms to find and separate signals from an array. (e.g. ESPRIT and 
MUSIC), so you don't really have to rely on a-priori steering of the 
array. Heck, some are even available as canned toolboxes in Matlab. 
(the fact that you can also turn the receive antennas into decoupled 
voltage probes fairly easily also helps)

The problem is when you flip around to transmit.  Knowing the receive 
info doesn't necessarily give you enough to synthesize the transmit. 
For one thing, the goals are different for receive (reduce the 
interferers) and transmit (squirt as much power toward desired direction 
as possible), and the issues with moving reactive power around the array 
are a lot more important. (you could put all the elements far apart, so 
there's no mutual coupling, but then you don't get any superdirectivity, 
and I think you want superdirectivity in this application.. otherwise 
you're limited to a gain of N)
_______________________________________________



_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk

<Prev in Thread] Current Thread [Next in Thread>