Re: NEC-LIST: NEC-WIN PRO & Buildings

I've done some limited modelling using NEC for buildings by creating
the building as a huge array of wires with lots of resistance. Other
than burning CPU cycles, however, I don't know if it actually matches
the real thing. Part of the problem is finding a suitable number(s)
for the properties of the walls. This is something that isn't
published in the usual handbooks, and I haven't figured out a good way
to measure, for instance, the large scale dielectric properties of
stucco over chickenwire, tarpaper, and wooden framing.

I am sure someone has done this (lots of modelling of urban
environments in the wireless industry), but I haven't seen any obvious
papers in a casual look through the IEEE search engine. Maybe it's
all considered proprietary (good propagation models are worth money
when choosing cell sites)? Pointers or suggestions, anyone?

What I've done so far is run the model a bunch of times varying the
properties of the wires (resistance, mostly) to see how sensitive the
numbers I am looking at (usually pattern... sidelobe levels, etc.) are
to variations in the parameters. If the results don't change much (a
very subjective decision), then I figure it's gotta be fairly good.
If the results change a lot, then I fall back on empiricism... build
it and try it, so at least then, I'm worried about measurement errors
and not modeling errors.

----- Original Message -----
From: "George Lappas" <el96077_at_central.ntua.gr>
To: <nec-list_at_ee.ubc.ca>
Sent: Tuesday, March 27, 2001 3:54 AM
Subject: NEC-LIST: NEC-WIN PRO & Buildings

> I am using NEC-Win at my university in Athens.
>
> I am trying to find a way to model a building (a house or a block of
> apartments).
>
> Is that possible with NEC-PRO? (Nittany Scientific)
>
> I am a beginner with the program and I am stuck. I need this
> information to see how the far field of an antenna is modified because
> of the existence of houses near it. Without this info I cannot
> conclude my work. I would appreciate it if someone could help me.
>
> Thanks in advance.
> George, Greece.
>
Received on Thu Mar 29 2001 - 18:43:06 EST