Re: NEC-LIST: Ansoft HFSS vs NEC etc.

A problem very similar to one my current research is concerned with...

This is a pretty straightforward problem which has received an fair amount
of treatment in the GPS literature. For example, there are a number of
reports and papers on the problems of multipath in geodetic surveying, and
for the development of antennas with very stable phase centers, particularly
when covered by the dielectric radome, surrounded by a choke ring, etc.
Trimble's web site even has some of them. If you "google" for "GPS Antenna
Phase Center" you'll turn up lots of references, and a literature search in
IEEE xplore, or equivalent, will also turn up a fair amount.

The vast majority of these papers use numerical EM models to generate their
data, although there is some confirming far field range, and near field
range, and in-situ observations over long times using GPS satellites as the
sources. You might look up GIPSY in the latter connection.

Now, as to determining phase centers...
The standard far-field pattern data from NEC gives you E(theta) and E(phi)
with both magnitude and phase. Since you're concerned about CP (if you're
doing GPS), then you'll need to appropriately convert the two linear
polarizations into CP, but that's simply a matter of algebra, and can be
done quickly in, e.g., Excel (having done it a couple of weeks ago for just
this purpose, albeit for an antenna where the phase center definitely
moves).

One can set up multiple feedpoints and drive the element currents in the
usual ways (NT cards) or use voltage excitation, etc. to do your phased
array stuff.

One can also set the problem up for plane wave excitation and look at the
feedpoint results, although I haven't done this recently.

I suspect that the real problem with NEC in this scenario is that it doesn't
accomodate the radomes, dielectric substrates and antenna supports very
well. HFSS will, in fact, do all this. There are also some FDTD codes
being used for this sort of thing. NEC has the nice advantage of being
inexpensive and fairly fast, and if you've got a BIG problem, there are
parallel processor versions around with the matrix math calls replaced by
appropriate parallel processor libraries.

There is also a fair amount of literature in the array radio telescope world
and in the phased array radar world dealing with the effects of
uncertainties in patterns, mutual interaction changing phase centers, etc.
IEEE Transactions on Antennas and Propagation is full of this. If you're a
member of A&PS, you can get a CDROM with all the back issues for something
like $100, which would be a pretty good deal on this.

Jim Lux
Spacecraft Telecommunications Engineering Section
Jet Propulsion Laboratory
Pasadena CA
(818)354-2075

"John Orr" <orr_at_ece.wpi.edu> writes:
> Hello,
> I would appreciate whatever insights anyone can provide on the best
> software choice for the following problem:
> For precise global positioning, knowledge of the phase center of an
antenna
> or antenna array (in the presence of adaptive beamforming) is critical.
To
> begin to understand this problem I need to know the magnitudes and phases
> at the feed points for each element in an array (obviously including
mutual
> coupling effects) for an incident plane wave from a given angle. This is
> the reverse of the typical antenna pattern calculation. I am a novice at
> computational E&M and the software that I have been using (Ansoft HFSS)
> does not seem to make it easy to do this calculation (but I may just be
> ignorant of some aspect of HFSS).
> Comments both on the problem and on the best choice of software would be
> welcome.
> Thanks very much!
>
> John A. Orr
> Professor of Electrical and Computer Engineering
> Worcester Polytechnic Institute
> Visiting Professor of Aero and Astro Engineering
> Stanford University
> Stanford tel: 650-724-2464
> WPI tel: 508-831-5273
> orr_at_wpi.edu, jorr_at_stanford.edu
>
> --
> The NEC-List mailing list <nec-list_at_gweep.ca>
> http://www.gweep.ca/mailman/listinfo.cgi/nec-list

-- 
The NEC-List mailing list <nec-list_at_gweep.ca>
http://www.gweep.ca/mailman/listinfo.cgi/nec-list