From dfs@smilodon.nosc.mil Wed Feb 22 18:27:14 1995
Received: from trout.nosc.mil by osiris.ucsd.edu.UCSD.EDU (4.1/UCSDGENERIC.3)
	id AA21511 to merino; Wed, 22 Feb 95 18:27:12 PST
Received: from smilodon.nosc.mil by trout.nosc.mil (4.1/SMI-4.1)
	id AA27580; Wed, 22 Feb 95 18:28:15 PST
Received: by smilodon.nosc.mil (931110.SGI.ANONFTP/931108.SGI.ANONFTP)
	for @trout.nosc.mil:merino@osiris.ucsd.edu id AA07812; Wed, 22 Feb 95 18:27:25 -0800
Date: Wed, 22 Feb 95 18:27:25 -0800
From: dfs@smilodon.nosc.mil (David F. Schwartz)
Message-Id: <9502230227.AA07812@smilodon.nosc.mil>
To: merino@osiris.ucsd.edu
Status: R

Greetings Orlando!

Congratulations, Bill tells me you are buying a house now.

Hope all is going well with you and your family.

I think I've got some applications for some of the H-infinity
methods and am making a proposal with Jeff Allen.  The group
I'm working with does computational electromagnetics and
applies it to antenna patterns for design and analysis.  My
first use will be to interpolate some complex impedance data
over a frequency band.  But there are more applications.
Will be trying NewtonFit on this, but 

I'm interested to know what code you are using to solve the
disk problem
               |k(z) - h(z)| <= r(z)
for both the scalar and vector cases.  I'm going to prepare
a Matlab version for use in these problems.  I vaguely
remember the fortran routines that I coded up, but wonder
if they have been supeseeded by better algorithms and better
coding.  What is being used now?  What should I base my new
code on?

Thanks,

Dave

email:  dfs@nosc.mil
phone:  619-553-5075


