Mathematics Department

______________________________________________________________________________
Mathematics Department
Professor Thomas J. Sanders
Chair
______________________________________________________________________________
Mathematics provides a logical framework and a language indispensable to understanding the technical
world in which we live. The following description summarizes the many contributions to this field of study made
during the past academic year by the faculty and midshipmen majors of the Mathematics Department of the U.S.
Naval Academy. The results cited reveal the great scope, diversity, and applicability of mathematics and offer
glimpses of its intellectual beauty and appeal.
Several midshipmen conducted research projects as Honors Mathematics Majors, or in specially created projects
under the guidance of faculty members. There were eight mathematics honors majors in the class of 2006. They are
listed below with their project titles and their faculty mentors:
Midshipman 1/C Peter Barkley, Justice in Liver Transplant Allocation, an Integer Programming
Approach (Assistant Professor Sommer Gentry)
Midshipman 1/C Timothy B. Brock, Linear Feedback Shift Registers and Cyclic Codes in Sage
(Professor W. David Joyner)
Midshipman 1/C Greg Coy, Long Quadratic Residue Codes (Professor W. David Joyner)
Midshipman 1/C Gregory Dietzen, Singular Analysis of an N-dimensional Cosmic String (Professor
Deborah A. Konkowski, Professor Mitch Baker and Associate Professor Alexis A. Alveras)
Midshipman 1/C John H. Doherty, Dynamic Systems: Spring, Cycloid and Pendulum Systems (Professor
Mark Kidwell)
Midshipman 1/C Robert B. Irving, Higher Dimensional Linear Regression as a Mathematical Foundation
for Data Classification (Associate Professor Gary O. Fowler)
Midshipman 1/C Gordon R. McDonald, Mathematics Honors Project Summary: Low Density Parity
Check Codes (Professor W. David Joyner)
Midshipman 1/C Daniel C. Ryan, Gods Algorithm on the Edges and Corners of the Rubiks Cube
(Assistant Professor Amy E. Ksir)
A special note is that Midshipman Dietzen was the winner of the Naval Intelligence Foundation Award for the best
honors project at the Naval Academy.
Regularly meeting throughout the year are four colloquia series, in pure mathematics, applied mathematics,
operations research, and teaching methods. These both bring in outside speakers and provide a forum for
department members and mathematics majors to present their work. There are also weekly seminars in topics of
special interest, such as Riemann Surfaces, Algebra, Applied Math and the Mathematics of Fluid Flows.
Once again, the Mathematics Department produced a wide range of scholarly work that appeared as
technical reports or as publications in refereed journals throughout the world. Dozens of articles appeared as
applications of mathematics or as pure mathematical research. Topics that Mathematics Department faculty
researched include:
mathematics pedagogy, chaos and dynamical systems, mathematical physics and cosmology, algorithms for
computers, wavelets, cryptology, ocean acoustics, image recognition, fluid flows, beam propagation, and
basic mathematical research in areas such as algebra, analysis, combinatorics, computational geometry,
differential equations, differential geometry, matrices, number theory, operator theory, and statistics.
In addition to many independent research projects, several research projects were sponsored in whole or part by a
variety of sources, such as:
Defense Modeling and Simulation Office
Joint Technical Office, High Energy Laser
National Aeronautics and Space Administration
National Science Foundation
Naval Research Laboratory
Naval Surface Warfare Center
Office of Naval Research
The Johns Hopkins University School of Medicine, Department of Surgery
During the past year, members of the USNA Mathematics Department presented the results of their
scholarly activities on over forty occasions at professional mathematical meetings and colloquia throughout the
United States and abroad. This activity, along with publication, enhances the academic stature of the Naval
Academy and promotes the professional growth and reputation of those individuals involved. Through research
activity, the faculty expanded their intellectual horizons and stayed vital in their disciplines. They contribute to the
discovery of new mathematics. And they develop new materials and ideas that they can share with midshipmen
students in their mathematics courses and research projects.
Sponsored Research
Endomorphism-semigroups of B(H)
Researcher: Assistant Professor Alexis A. Alevras
Sponsor: Naval Academy Research Council (NARC)
In joint work with Professor R. T. Powers ,University of Pennsylvania and G. L. Price, US Naval
Academy, we initiated the study of non-conservative (i.e. non-unital) endomorphism semigroups on type I factors.
Cocycle conjugacy among such semigroups has many of the properties of cocycle conjugacy among (conservative)
E0-semigroups. We introduced and studied a natural relationship between endomorphism semigroups: if
is cocycle conjugate to a subordinate of . Using techniques developed in our previous work, we proved that
this relation is antisymmetric, at least among semigroups corresponding to one-dimensional boundary weights: if
and then and are cocycle conjugates. This is work in progress.
Geo-Spacial Data Analysis

Researcher: Professor James DArchangelo
Sponsor: Naval Research Laboratory (NRL)
The Counter Insurgency Pattern Assessment (CIPA) program seeks, among other objectives, to predict
adversarial events as a function of geo-spatial variables correlated with historical events. Large numbers - in the
order of hundreds or more - of variables may be under consideration as corollaries of these events.
The current task is two-fold:
(1) To collaboratively explore methods of multivariate data reduction (e.g. principal factors analysis) for
finding collinear relationships among the provisional independent variables. The objective is to map the large set of
independent variables into a reduced space of low dimensionality.
(2) To present the results of this analysis geospatially using the Matlab Mapping Toolbox. It should be
possible to represent the predicted events geospatially, as a function of either the full or reduced set of independent
variables.
The author investigated the feasibility of applying various spatial statistical and computational methods
including Bayesian analysis with Gibbs sampling as methods for predicting adversarial events.
On the Visualization of Large Point Cloud Data

Researchers: Associate Professor Sonia M. F. Garcia and David Harding
Sponsor: Naval Research Laboratory (NRL)
Visualization is the act of making a visible presentation of numerical data, particularly a graphical one. The
goal is to help the viewer to form a mental image. Our work was to produce a scientific visualization, i.e. using
analysis and graphing tools to classify, summarize, explore, and present large data sets.
Visualization tools are needed for many reasons for example; reactions to future management plans will be
influenced by experiences of current and past activities. Visualization can help with understanding these issues:
Habitat Disturbance, Visual impacts, Operational considerations and Highlighting specific features.
Our particular study used a three-dimensional mathematical morphology, a machine vision technique, to
extract measurements of individual seismic activity presented at the Mount St. Helens, Washington State. Mount St.
Helens caused a dreadful eruption in 1980 with many dead, innumerous houses destroyed, and hundred of beautiful
forest completely ruined. Mount St. Helens became active again in autumn 2004, indicated initially by hundreds and
then thousands of localized earthquakes, and followed by several significant emissions of steam and ash. Since then
small earthquakes (maximum about magnitude 1) continue to occur at a rate of about 1 per 5 to 10 minutes. Visual
observations and thermal imaging of the crater, the 1980-86 lava dome, and the intensely deforming and uplifting
area on the south side of the dome are occurring constantly. These observations were consistent with new lava
having reached the surface of the uplift. Additional visual and thermal observations are being used to further
evaluate this interpretation.
While geologists warn that an eruption similar to the May 1980 eruption is still possible, the chances are
low. More studies are needed to monitor the behavior of this nature beast.
Microaltimeter Flight Data Post Processing

Researchers: Jan McGarry, Phil Dabney and Associate Professor Sonia M. F. Garcia
Sponsor: Laboratory for Terrestrial Physics NASA / Goddard Space Flight Center
The Microaltimeter System is a prototype for a new concept in laser altimeter systems, using a low energy
per pulse (<10 MicroJoule), high repetition rate (~10khz) laser and a small diameter telescope (15 cm). This system
requires less mass and power than traditional altimeter designs, but has significant noise content in the data stream.
Signal processing is required to distinguish the signal from the noise.
Navigational (differential GPS and INS) and image data is collected along with the altimeter ranging data
to allow transformation of the range time of flight information into terrain heights at the geolocated laser footprints.
Calibration files (ranging to ground targets) to determine system delays, and flight data over bay water to determine
offsets in the INS attitudes are also needed to remove biases from the data. We refer to this transformation of range
times into latitudes, longitudes and heights, along with extracting the signal from noise, as the post processing.
To validate the Microaltimeter performance, we proposed to verify our geolocated terrain heights derived
from the ranges, against an independent Digital Elevation Model (DEM) of the Ocean City and Assateague Island
areas. The DEM covers roughly the upper half of Assateague Island and the lower half of Ocean City (from the
Route 90 bridge south).
We are, unfortunately, not finished with the post processing effort - it has proven a more challenging effort
than originally anticipated. A paper presenting the details of the Microaltimeter post processing will be written when
the effort reaches maturity.
Optimizing the Utilization of Live Donor Kidneys through Kidney Paired Donation
Researchers: Dorry Segeg and Assistant Professor Sommer Gentry
Sponsor: The Johns Hopkins University School of Medicine, Department of Surgery
Members of an end-stage-renal-disease patients family are often willing to be live kidney donors, but at
least a third of such offers must be rejected because of blood-type or tissue-matching incompatibility. In kidney
paired donation, two such patients and their donors exchange kidneys in simultaneous operations so that both
patients receive a compatible kidney: the donor of the first family gives to the patient of the second, and vice versa.
In a large pool of patients and their incompatible donors, deciding which pairs should exchange with which other
pairs is equivalent to matching on a graph. We have used computational trials of proposed interventions on
simulated patient databases to design and test these novel organ allocation systems in advance of implementation.
Because we provided some of the first numerical estimates of the extent to which this could alleviate the
organ shortage, this work kicked off an effort to create a national kidney paired donation registry. A bill explicitly
making kidney paired donation legal (its legality is now unclear) was introduced in February and is expected to pass
in the U.S. Senate. Our research has also generated excitement in the popular press, which can only help encourage
organ donation in general. Articles and programs on this research have appeared in TIME magazine, Readers
Digest, MIT Technology Review, the Baltimore Sun, CBS News, and National Public Radios Diane Rehm Show.
The TV show NUMB3RS also formulated an episode around this idea.
This grant focused our research more directly on the many questions that policy-makers, clinicians, and
especially patients will have about paired donation as it becomes more widely available. The grant funded computer
programming / support services for a 20- node Linux cluster housed at Johns Hopkins University, and to date 17
abstracts have been accepted / presented at national transplant meetings, from four different investigators working
on this project.
Combinatorics of Updown Categories

Researcher: Professor Michael E. Hoffman
In earlier work the investigator developed the notion of an updown category, which provides a uniform
framework for numerous (graded) partially ordered sets in combinatorics, including integer partitions, integer
compositions, rooted trees, and planar rooted trees. The definition naturally provides the graded vector space
generated by the set of objects of an updown category with operators U and D, and the coefficients of iterates of
these operators applied to objects of the category have an enumerative interpretation.
The investigator has singled out a number of properties of updown categories, based on his extensive study
of examples. There is a hierarchy of commutation conditions, based on properties of the commutator DU-UD.
There are also even covering conditions, which can be defined in terms of the effect of the operators U and D on
objects in a given rank. There are some interesting relations between the two sets of conditions, which can be
expressed in terms of generating functions associated to the updown category.
Automorphisms of Curves and Codes

Researchers: Professor W. David Joyner and Assistant Professor Amy E. Ksir
Given a projective algebraic curve X and two divisors D and E on X, one can use the Riemann-Roch space
L(D) to construct an error-correcting code C(D,E), called the Goppa code or AG code. If G is a finite group of
automorphisms of X, and D is invariant under G, then there is a natural action of G on the code. Our main goal this
year was to find families of examples of curves with a finite group action where the representation of G on the
Riemann-Roch space could be explicitly calculated. We found two such families of examples. The first, found with
the help of Roger Vogeler from Ohio State University, is the family of Hurwitz curves with an action of the finite
group PSL(2,q). Building on our previous work with modular curves, which have an action of the same group, we
were able to compute the representation structure of both the ramification module and L(D), for any G-equivariant
divisor D. Our second family of examples is the Artin-Schreier curves y = x x , where p is prime and is the
characteristic of the field. We realized that the techniques we were developing to compute Riemann-Roch spaces
2
for these curves could be generalized to any hyperelliptic curve y = h( x) . We then turned our focus to finding
explicit bases for Riemann-Roch spaces of hyperelliptic curves. This will have applications in computational
algebraic geometry beyond our project. We have written up and submitted both results.
2
Twin Screw Extrusion of Energetic Materials

Researchers: Assistant Professor Anastasios Liakos
and Dr. H. Bruck (Mechanical Engineering Department, University of Maryland)
Sponsor: Naval Surface Warfare Center in Indian Head and Naval Academy Research Council (NARC)
This project is still in progress. A second-generation three dimensional program has been written in C to
try to predict the values of the flow variables (velocity, pressure, stress). In addition, research was done to mesh the
complex geometry of the flow domain of the extruder.
Modeling and Analysis of Laser Propagation IV

Researcher: Professor Peter McCoy
Sponsor: Joint Technical Office, High Energy Laser
This research considers the basic physics and foundations of mathematical analysis modeling beam
propagation followed by computational-numerical simulation via code to be developed and adopted at the Naval
Academy.
The Modeling, Mathematical and Numerical Analysis includes the Paraxial Wave Equation
2
t
2ik z = 0 where t2 , the transverse Laplacian, is the basic partial differential equation describing
laser propagation through a uniform medium. It arises by making the so-called paraxial approximation to
Helmholtzs equation. A cylindrical coordinate based Gauss-Laguerre normal mode analysis of solutions of the
PWE initiated by Reza Malek-Madani (RMM) and Peter McCoy (PMc) is to be completed. An elliptical coordinate
based analysis of the PWE will be conducted by RMM and PMc based on Gauss-Ince normal modes. These normal
modes are state of the art in modeling Laser Propagation as well as in the study of orthogonal polynomials and
special functions.
Modeling of MEMS Gyroscopes

Researcher: Assistant Professor Aurelia Minut
Sponsor: Institute for Scientific Research/Office of Naval Research (ONR)
The MEMS (microelectromechanical systems) gyroscopes can be used in navigation systems in cases
where the GPS is denied. The MEMS gyroscopes are small devices that can be mounted on a wide range of
platforms, such as UAVs, guided missiles. This research attempts to improve the performance of current gyroscopes
by weakly coupling an array of MEMS gyroscopes.
Most MEMS vibratory gyroscopes are modeled by the mass spring system. A single proof mass with the
same x and y-axis spring constants is suspeded above a plate and it is supported by anchored comb-drive flexible
structures. These structures make the mass oscillate freely in two orthogonal directions: the drive (y-axis) and the
sense (x-axis). If an external drive force (usually a sinusoidal force) is applied, the proof mass oscillates in the drive
direction. The drive force is an electrostatic force induced by the comb-drive structures. The driving force in the
sense direction is the Coriolis force which is induced by rotation and causes dynamic coupling between the
oscillation axes.
We developed a model for the equations of motion and solved the equations numerically. Then we tried
coupling two, four, nine and twelve gyroscopes in different ways and compared the results. The goal of this work is
to study the stability and equilibrium of the system of equations for forty gyroscopes and find the best way to couple
them.
United States Naval Academy Chesapeake Bay Research

Researcher: Professor George Nakos
Sponsors: Defense Modeling and Simulation Office (DMSO)
In summer of 2005 and partially continuing throughout the Academic year, the investigator was part of an
effort to study the water movement and residence time in the Chesapeake Bay. The specific research participation in
this project was the installation in a linux environment of several libraries and the QUODDY software, as well as the
use of the above. The results of the QUODDY runs were crucial in Midshipmen research. The ability to run these
programs locally and fast enhances the overall Chesapeake Bay research program at the Naval Academy.
A Multi-Scale Edge Transform for Image Coding

Researchers: Assistant Professor Irina Popovici and Professor Wm. Douglas Withers
Sponsors: Office of Naval Research (ONR)
This project concerns lossy compression of either color or black-and white images, using an approach
radically different from the popular JPEG and JPEG 2000 standards. Our objective is to exploit the edges and
shadows in an image such that a smaller number of bits can be used to represent it, while maintaining an
``acceptable'' visual quality for the decompressed image. The correlation between spatial locations of pixels has been
exploited by almost all existing algorithms, but the driving principle of these algorithms is elimination of data in a
simple (say linear or cubic) dependency from proximate pixels. We propose a recurrent coding technique that also
exploits the discontinuities in data near the contours of an image.
The popular wavelet approach to image compression describes an image in terms of a linear basis of
elements which are self-similar and which also reflect the smooth nature of many image regions. In this basis,
however, a sharp, straight edge must be described as a superposition of many wavelet elements at all scales. The
first step in this project is development of a linear basis which shares the wavelet property of self-similarity but is
reflective of sharp, straight edge structures rather than smoothness. Such a basis has potential applications in pattern
analysis and image enhancement as well as image coding. Following the development of this basis, the remainder of
the coding problem can be approached in a fashion similar to standard wavelet image coders.
Cocycles for One-parameter Flows of B(H)

Researcher: Professor Geoffrey L. Price
Sponsor: National Science Foundation (NSF)
In the 1930s, E. Wigner proved that one-parameter groups of automorphisms on B(H), the algebra of
bounded operators on a Hilbert space H, are implemented by one-parameter group of unitary operators. Many years
later R. T. Powers began a systematic study to find the analogue of Wigners result for the time-irreversible systems
on B(H) known as E0-semigroups. Price has collaborated with Powers and A. Alevras in this work. A recent paper
by the three authors includes a study of the multiplicative structure of the family of local cocycles of certain E0semigroups that are called spatial, which means that they are intertwined with the identity on B(H) by a oneparameter semigroup of isometries. The local cocycles provide an invariant under the equivalence of E0-semigroups
known as cocycle conjugacy. In current work Alevras, Powers and Price are turning this problem around in a sense
by trying to classify, for a given family of local cocycles, all spatial E0-semigroups sharing this family.
Marginal Tracial States on Tensor Products of Matrix Algebras

Researcher: Professor Geoffrey L. Price
Sponsor: National Science Foundation (NSF)
A tracial state on an operator algebra A is a positive linear functional on A which satisfies 1) ( I )
= 1,
where I is the identity on A, and 2) ( xy ) = ( yx) for all x, y in A. It is well-known that there is a unique tracial
state TrB on B, the algebra of n n matrices over the complex numbers. Suppose B1 = B = B2 .
tracial state on the tensor product
A marginal
B1 B2 is a state that satisfies ( x I ) = TrB1 ( x) and
( I y ) = TrB ( y )
2
for all x, y in B. In 2003 K. R. Parthasarathy showed that if B is the algebra of
2 2
matrices the extremal marginal states on B1 B2 are all pure states, i.e., they are extremal among the convex set
of all states on B1 B2 .
Recently Price and Shoichuro Sakai have made some progress in their attempt to generalize this result to
the case where B is the algebra of n n matrices over the complex numbers.
Using Off-the-shelf Software in the Classroom

Researcher: Professor Thomas J. Sanders
Sponsor: Defense Modeling and Simulation Office (DMSO)
The objective of this project was to further investigate the possibility of using off-the-shelf software in the
classroom. In particular, I was interested in the possibilities of using the computer game Empire Earth as a vehicle
to teach tactical analysis to Midshipmen. After becoming somewhat familiar with the play of the game and the tools
included to build scenarios, I concluded that the software was not suitable for this purpose. The fidelity of the game
and the amount of control over the units is not sufficient for the intended use.
Independent Research
Endomorphism Semigroups Parametrized by the Forward Light Cone
Researcher: Assistant Professor Alexis A. Alevras
This project is in progress. We are interested in the study of strongly continuous semigroups of
endomorphisms of B(H), ={: C }, where C is the forward light cone in Minkowski space. Such semigroups
occur naturally, for example when one is given a system of local observables which is acted upon by the Poincare
group in such a way that the Haag-Kastler axioms are satisfied, and a subalgebra which is invariant under the action
of translations by elements of the forward light cone. Specific examples of this situation have been given using the
construction of Free Hermitian Scalar Fields. The immediate goal is the classification of these examples up to
cocycle conjugacy through index theory.
Seabed Acoustics
Researcher: Professor James L. Buchanan
Chotiros and Isakson [J. Acoust. Soc. Am., 116 (4), 2011-2022 (2004)] recently proposed an extension of
the Biot-Stoll model for poroelastic sediments that makes predictions for compressional wave speed and attenuation
which are in much better accord with the experimental measurements of these quantities extant in the literature than
either those of the conventional Biot-Stoll model or the rival model of Buckingham [J. Acoust. Soc. Am., 108 (6),
2796-2815 (2000)]. Using a local minimizer, the Nelder-Mead simplex method, it is shown that there are generally
at least two choices of the Chotiros-Isakson parameters which produce good agreement with experimental
measurements. Since one postulate of the Chotiros-Isakson model is that, due to the presence of air bubbles in the
pore space, the pore fluid compressibility is greater than that of water, an alternative model based on a conjecture by
Biot [J. Acoust. Soc. Am., 34 (5), 1254-1264 (1962)], air bubble resonance, is considered. While this model does as
well or better than the Chotiros-Isakson model in predicting measured values of wave speed and attenuation, the
Rayleigh-Plesset theory of bubble oscillation casts doubt on its plausibility as a general explanation of large
dispersion of velocity with respect to frequency.
Non-Parametric Density Estimation of Streaming Data Using Orthogonal Series

Researcher: Lieutenant Commander Kyle Caudle, USNR
Computer technology in the 21st century has allowed us to gather and collect data at rates that would have
seemed impossible less than a decade ago. As such, typical data base management systems (DBMS) are having
great difficult storing and analyzing data in the traditional way. Systems that receive large amounts of data in a
transient data streams generally need to analyze the data immediately without storing it on a disk. These systems are
referred to as data stream management systems (DSMS). This emerging field has been pushed to the forefront by
technology that demands analyzing data in realtime. Babcock et. el. [2002] analyzed the issues involved in mining
rapid time-varying data streams. To date most of the work in the area of DSMS has primarily been concerned with
querying the data streams. These queries provide estimates of parameters, such as the mean, and then continuously
update them as more data arrives. Recently, Heinz and Seeger [2004] used data streams to provide an estimate of
the underlying probability density function by dividing the data up into bins or windows containing the most recent
data. An estimate of the density was created using the standard wavelet cascading algorithm on the binned data.
Designing Digital Tools for Communicating Mathematics in Higher Education

Researcher: Professor Carol G. Crawford
The author extended prior research into the design and development of online tools for communicating
mathematics at the university and college level. A paper was published in September, 2005: Proceedings of the Joint
Meeting of the 3rd International Conference on Education and Information Systems, Technologies and Applications
(EISTA 2005), Volume I, pages 223-228. This work reflects work with Professor Mark Meyerson and Professor
Michael Chamberlain (USNA Mathematics Department) and Commander Thomas Logue, USN (USNA Computer
Science Department) in cooperative Curriculum Development Projects at the Naval Academy.
Nontrivially Discriminating Groups

Researcher: Professor Anthony M. Gaglione
Discriminating groups were first introduced by Baumslag, Myasnikov and Remeslennikov with an eye
toward applications to the universal theory of various groups and solving equations in groups. We need to fix some
terminology. Let G be a group. We say that G is discriminating provided that for every finite nonempty subset S of
nontrivial elements of the Cartesian square, GG, there is a homomorphism S:GG G such that S (s) 1 for all s
in S. It is clear that if GG embeds in G then G is discriminating. If this is the case, we term G trivially
discriminating. The objectives of this project were to try to find examples of nontrivially discriminating groups and
to address many of the open questions which have recently been put forth concerning them. This research was
started in a paper by A.M. Gaglione (et al) entitled Discriminating Groups (Journal of Group Theory, 2001, 463474.(Abbreviated FGMS here) At the time of the writing of FGMS, the only examples of discriminating groups
known were of two types: abelian groups and groups which embed their direct squares (i.e., trivially discriminating
groups). It is easy to show that any torsion free abelian group is discriminating. Thus nontrivially discriminating
groups do exist. The real open question was did they have to be abelian. Together with my collaborators, we have
discovered several classes of nonabelian nontrivially discriminating groups, i.e., nonabelian groups which are
discriminating but do not embed their direct squares. The methods used here go back to classical work of B.H.
Neumann on certain two-generator subgroups in an unrestricted direct product of an infinite family of finite
alternating groups. Here we were able to produce an uncountable family of two-generator nonabelian nontrivially
discriminating groups. This work will appear in a recently accepted paper in the Journal of Group Theory. Another
main question put forth in FGMS was whether finitely generated (f.g.) nilpotent groups are discriminating. We
proved that f.g. nilpotent groups are discriminating if and only if they are torsion free abelian. This was done using
the method of Malcev completions. As a matter of fact we proved more, i.e., the Malcev completion of a f.g. torsion
free nilpotent group is discriminating if and only if it is abelian. This work will appear in an article recently accepted
for publication in a volume of the American Mathematical Society (AMS), series Contemporary Mathematics. We
have answered several of the open questions which arose in FGMS, but these have given rise to new questions
concerning discriminating groups. Thus this research project is on going.
Maximum Matching Inequalities in Kidney Paired Donation

Researchers: Assistant Professor Sommer Gentry and Associate Professor T. S. Michael
Arranging kidney exchanges requires matching on a graph. Maximum cardinality matchings are desirable,
but the relative utility of kidney exchanges must be represented by edge weights. We have proved that setting edge
weights within a precisely described range guarantees that a maximum edge weight matching will also have
maximum cardinality. This gives a method for choosing edge weights in a national kidney paired donation
program, so as to guarantee that no opportunity for a kidney transplant will be lost but still allow for distinguishing
between preferred and non-preferred edges. We are extending these results to special cases in which some
exceptions to this maximum cardinality rule are allowed to give extraordinary preference to a small number of
patients.
Designing and Evaluating Weighting Functions for Matching on Graphs

Researcher: Assistant Professor Sommer Gentry
In medical resource allocation applications, objective functions do not necessarily simplify to a dollar
figure or even a life-years gained outcome. For instance, ethical considerations are paramount in deciding which of
two needy patients should receive a single available kidney transplant. In deceased donation the objective function
is essentially equivalent to rank-ordering, and allocation is done via an established points system. In kidney paired
donation, however, which is equivalent to maximum edge weight matching on graphs, designing an objective
function giving weights for different paired donation opportunities is presently more of an art than a science.
Whether increasing the weight on a certain type of edges by a given amount will change the optimal solution
depends greatly upon the structure of the graph. The goal of this work is to make the tradeoffs inherent in setting
edge weights comprehensible to the transplant professionals and ethicists who must set these weights in practice.
The gold standard would be proofs of simple statements about weights, but where these are impossible one may
provide guidance using simulations.
Hopf Algebras of Trees and Symmetric Functions

Several years ago, Dirk Kreimer introduced a Hopf algebra of rooted trees to describe renormalization in
quantum field theory. This has stimulated much algebraic work, both on Kreimers original Hopf algebra and on
similar objects, including a Hopf algebra of planar rooted trees described by Loic Foissy. In fact, the relation
between Kreimers Hopf algebra and Foissys Hopf algebra is analogous to that between the Hopf algebra of
symmetric functions and the Hopf algebra of quasi-symmetric functions; this analogy can be made precise in the
form of a commutative diagram, whose study may produce some interesting results.
The investigator has given two talks about his work on this topic, including one at the Mathematisches
Forschungsinstitut in Oberwolfach, Germany. A short account of his work will be published in the Oberwolfach
Reports, and he plans a longer publication on this topic later.
Combinatorics of Updown Categories

We define the notion of an updown category, which generalizes Stanleys notion of a sequentially
differential poset. The vector space generated by the set of objects of an updown category naturally admits two
operators U and D, and the coefficients of iterates of these operators applied to objects of the category have an
enumerative interpretation. We consider some conditions on the operators U and D, and establish their
combinatorial implications.
A Stability Criterion for Waves in a Coupled Optical System

Researchers: Assistant Professor Russell Jackson and Christopher K.R.T. Jones (University of North Carolina)
In my thesis, I identified a new bifurcation that led to many novel pulses within a coupled optical system,
including a family of N-pulses for all positive integer N. The geometry of this bifurcation provides a key
component (which we have now analyzed) in understanding the stability of these same waves, but this (geometric)
component needs to be incorporated into a more general (analytical) framework before the whole picture can be
understood.
Bifurcation and Stability of Standing Waves in a Gross-Pitaevskii Equation

Researcher: Assistant Professor Russell Jackson
The nonlinear Schrodinger Equation with potential provides a general model for Bose-Einstein
condensation. In most experimental setups, this potential is produced either by a magnetic trap (yielding a singlewelled potential) or an optical lattice (yielding a periodic or many-welled potential). Using phase-plane techniques,
a number of standing waves can be identified within this system, and once again, the geometric construction of these
pulses provides much immediate information concerning the stability properties of each pulse. We have completed
the analysis of the pulse construction in the N-well case, but of continuing interest is the relationship between the
stability properties between the system of N wells with N large (but finite) and the limiting periodic system with an
infinite number of wells.
A Numerical Package for Computation of the Evans Function

Numerical tools to help with the determination of stability in problems in nonlinear waves are being
prepared. Once a wave has been identified within a system, its stability can theoretically be determined by the
evaluation of the Evans Function. However, in practice this function is difficult to compute, except in special
limiting cases (like a singular limit in parameter space). I am working on developing robust numerical tools for the
evaluation of the Evans Function for pulse solutions within general nonlinear PDE. This involves linearizing the
PDE about the pulse solution and following the stable and unstable subspaces of the asymptotic states along the
pulse. These computations are most readily performed in the Space of Exterior Powers for the coupled system in
(Project 1) this becomes a 70 (8 choose 4) dimensional system; for the single system in (Project 2) this is a 6 (4
choose 2) dimensional system.
Pulses in Nonlinearly Coupled Schrodinger Equations: I. A Homoclinic Flip Bifurcation

In this work, we describe a new mechanism for the generation of multi-pulse solutions in a class of
nonlinearly coupled Schrodinger equations. Many novel pulses have been observed in such systems both
numerically and experimentally but, until now, an understanding of their origins had been lacking. The particular
bifurcation studied here is spurred by the passage through degeneracy of a one-component pulse in orbit-flip
configuration. We provide a straightforward geometric analysis, demonstrating the production not only of a multicomponent 1-pulse nearby the original one-component pulse, but also of an entire family of alternating N-pulses, for
all positive integers N.
Pulses in Nonlinearly Coupled Schrodinger Equations: II. An Instability Criterion

In this work, we derive a geometric criterion for the instability of pulses in a class of nonlinearly coupled
Schrodinger equations. Many novel pulses have been observed in such systems both numerically and
experimentally, and recent work has provided an understanding of the generation of such pulses. The criterion given
here allows an immediate transfer of the information now known about the mechanisms and geometry of pulse
creation into a new description of the stability of these same pulses. Additionally, we report the discovery of a
secondary bifurcation of (potentially stable) asymmetric pulses that occurs when the value of this criterion changes.
Bases for Riemann-Roch Spaces of Hyperelliptic Curves

Researchers: Professor W. D. Joyner and Assistant Professor Amy E. Ksir
Let X denote a hyperelliptic curve and let D be an effective divisor on X. We give an explicit constructive
method for computing a basis B of L(D). In particular, it is done in such a way that if E < D is another effective
divisor, the basis BE of L(E) so constructed will be a subset of B. We plan to use this in future work to compute
quotient representations, with applications to AG codes. We used SAGE to compute an example in some detail.
10
Group Actions on Riemann-Roch Spaces of Some Hurwitz Curves

Researchers: Professor W. D. Joyner, Assistant Professor Amy E. Ksir
and Dr. R. Ross Vogeler (The Ohio State University)
Let q > 1 denote an integer relatively prime to 2,3,7 and for which G = PSL(2,q) is a Hurwitz group for a
smooth projective curve X defined over the complex numbers. We compute the G-module structure of the RiemannRoch space L(D), where D is an invariant non-special divisor on X. This depends on a computation of the
ramification module, which we give explicitly. In particular, we obtain the decomposition of H1(X,C) as a Gmodule.
Ramified Artin-Schreier
Researcher: Professor W. D. Joyner and Assistant Professor Amy E. Ksir
We consider the curve y2 = xp x over the field GF(p), which is both an Artin-Schreier curve and a
hyperelliptic curve. This curve has SL(2, p) as its automorphism group. We compute the SL(2, p)-module structure
of the ramification module and Riemann-Roch spaces for this curve.
Classification of Singularities with SHEEP/CLASSI and/or MAPLE

Researcher: Professor Deborah A. Konkowski
This research is an investigation into ways to apply computer algebra programs (e.g. SHEEP/CLASSI and
MAPLE) to the problem of singularity classification. This computer algebra research was begun during a recent
sabbatical in London at Queen Mary and Westfield College with Professor Malcolm MacCallums group and has
been continued during periodic visits to London. Although it is impossible to classify topological singularities such
as quasiregular singularities in this manner, scalar curvature and non-scalar curvature singularities should yield to
0
analysis. Ways to study scalar and nonscalar curvature singularities using a complete listing of C - curvature
invariants and frames related to parallel propagated orthonormal ones are currently under consideration.
The use of MAPLE was just begun last summer whereas the use of SHEEP/CLASSI has been ongoing.
MAPLE is very promising for singularity classification because it can algebraically evaluate the invariants needed to
test for curvature singularities. I used it for special cases including power-law metrics. It would be nice if some
numerical evaluation could be obtained but this requires knowing where the singularity may be through geodesic
computation which is not automatic. This research together with the SHEEP/CLASSI work will be continued.
Classical and Quantum Singularities in General Relativistic Spacetimes

Researcher: Professor Deborah A. Konkowski
This researchers main research area is the analytical study of spacetimes with singularities. She is
particularly interested in studying spacetimes with mild singularities and Cauchy horizons. Mild singularities
include quasiregular and nonscalar curvature singularities. In the case of quasiregular singularities, particle paths
end suddenly with no warning from infinitely-increasing tidal forces, while in the case of nonscalar curvature
singularities some, but not all, particles moving near the singularity feel infinite tidal forces. A Cauchy horizon is
the boundary of the causal development of spacetime. Cauchy horizons and singularities are intimately connected,
including through the cosmic censorship conjecture, which continues to be debated.
D. A. Konkowski is studying (1) a quantum-mechanically singular spacetimes, (2) two previouslydeveloped stability conjectures, and (3) the global structure of various singular spacetimes. Some of this work is
done in collaboration with T.M. Helliwell (Harvey Mudd College) and students at both Harvey Mudd and at the
Naval Academy.
In the first case, following Horowitz and Marolf and work begun by T.M. Helliwell and D.A. Konkowski,
classical test particles are replaced with quantum test wave packets to test the singularity structure of spacetimes
which classically possess all types of classical singularities. The research began with mildly classically singular
spacetimes but has branched out in the past few years. The most recent research has been in answering the question:
Can quantum mechanics heal classical singularities? Broad classes of spacetimes with power-law metrics near the
origin are under study. These classes can be nonsingular or singular classically. The classically singular ones have
11
then been studied for quantum singularities and further analyzed. Whole parameter spaces of classically singular but
quantum mechanically nonsingular spacetimes have been discovered. This research is almost complete and a paper
will soon be in preparation. Whereas completed results for early research on primarily quasiregular spacetimes were
presented at summer 2005 conferences in Bern (Switzerland) and Paris (France), preliminary results for the latter
work was presented more recently in Dallas (Texas).
In the second case, conjectures proposed by T. M. Helliwell and D. A. Konkowski to predict whether
various mild singularities and Cauchy horizons are stable are being investigated further. Thus far, the singularity
conjecture always holds true but the Cauchy horizon conjecture misses purely Weyl singularities that occur in the
plane wave and exact mass-inflation spacetimes.
In the third case, spacetimes with nonscalar and quasiregular singularities are under analytical
investigatigation. This includes work on: (1) spacetimes with nonscalar singularities due to Siklos, (2) quasiregular
singularities in the context of dislocations and disclinations and in the context of Columbeaus extended theory of
distributions, and (3) the structure of some spacetimes with directional singularities.
Rationality of G-Modules
Researchers: Dr. Ted Chinburg (University of Pennsylvania) and Assistant Professor Amy E. Ksir
Let X be a projective algebraic curve and let G be a finite group of automorphisms of X. The ramification
module is a G-module induced from the action of the stabilizer subgroup on the cotangent space at the ramification
points of X over X/G. A recent paper of Joyner and Ksir established a simple formula for the structure of the
ramification module in the case that it is rational. This has led to the natural question: when will such a module will
be rational? More generally, when will a G-module induced from a cyclic subgroup be rational? This project has
two goals. The first is to determine, in general, when induced modules from cyclic subgroups will be defined over
the rational numbers. The second is to determine when such a G-module is defined over the p-adic numbers. The
answers to these questions are of interest both in the theory of error-correcting codes and in number theory.
Research on this project was started in April 2006; one theorem giving a criterion in the case that G is PSL(2,q) has
been proven and is being written up.
Expository Notes on Quantum Field Theory

Researcher: Assistant Professor Amy E. Ksir
In the past ten years, algebraic geometry has been deeply affected by results in theoretical physics,
especially the physics surrounding string theory. Quantum mechanics and quantum field theory are the basis for
many of these results, but these fields are unknown to many algebraic geometers. These notes aim to explain the
ideas behind quantum mechanics, quantum field theory, supersymmetry, and string theory to the audience of
algebraic geometers.
A Posteriori Error Estimators for a Two-Level Finite Element Discretization

of Viscoelastic Fluid Flow
Researchers: Assistant Professor A. Liakos and H.K. Lee (Clemson University)
We derive locally calculable a posteriori error estimators for a two-level method of discretizing the
equations of steady-state flow of a viscoelastic fluid obeying an Oldroyd-type constitutive equation with no-slip
boundary condition.
The two-level algorithm consists of solving a small non-linear system of equations on the coarse mesh and
then using that solution to solve a larger linear system on the fine mesh. Specifically, following Najib and Sandri,
we linearize the Oldroyd-type constitutive equation about the coarse mesh solution thus nullifying the difficulties
brought by the advection term. Our theoretical error estimates show that it has optimal order accuracy provided the
true solution is smooth and its norm is sufficiently small. In addition, our computational error estimates exhibit the
validity of our analysis.
12
Modeling Gas Diffusion in Nanocomposites

Researchers: Assistant Professor A. Liakos, C.S. Swannack (Massachusetts Institute of Technology)
and C.L. Cox (Clemson University)
Polymer-silicate layered nanocomposites (PSLNs) have received the attention of the food packaging
industry since they have exhibited enhanced barrier properties such as decreased oxygen permeability. In this
report, we present a model for the estimation of the average diffusion coefficient of a nanocomposite film.
No-Slip Boundary Conditions for the Navier-Stokes Equations by Penalty-penalty Method

Researchers: Assistant Professor A. Liakos and A. Caglar
We prove convergence of the finite element method for the Navier-Stokes equations in which the no-slip
condition and the no-penetration condition on flow boundary are both imposed by penalty methods. This approach
has been studied for the Stokes problem but it has not progressed beyond the linear Stokes problem. Since the
inertial effects dominate the motivating application, it is crucial to extend the validity of the method to the non-linear
Navier-Stokes case. We have started this extension, analyzed the method, and given numerical results. We have
shown that optimal order of convergence can be achieved if the computational boundary follows the real flow
boundary exactly.
Atiyah-Singer Index Theorem

Researcher: Professor Robert Lockhart
I am currently studying the Atiyah-Singer Index Theorem. I hope to eventually use it in nonlinear analysis
in particular in dealing with computability of various problems in the calculus of variations.
Seasonality Methods with Statistical Measure

Researcher: Lieutenant Commander Marc D. Lucas, USN
Current methods used by the US Government and standard actuarial practice for smoothing and
interpreting seasonal data lack statistical interpretation. Use of a linear regression on transformed data yields
identical factors with statistical measure.
Controls for Complete Controllability of Linear, Autonomous Systems of ODE

Researcher: Associate Professor T. J. Mahar
There is a well known theorem in control theory which gives a necessary and sufficient condition for a
system of ODE to be completely controllable. The proof is not constructive. An approach has been devised which
generates controls which steer the system to any desired target. The controls do not appear to be optimal in any
sense, though this has not been proven. The approach extends to systems of PDE in a natural way.
Classical Poincar Metric Planted off Singularities

using a Chow-Type Theorem and Desingularization
Researchers: Associate Professor Caroline Grant Melles and Pierre Milman (University of Toronto)
We construct complete Khler metrics on the nonsingular set of a subvariety X of a compact Khler
manifold. To that end, we develop (i) a constructive method for replacing a sequence of blow-ups along smooth
centers, with a single blow-up along a product of coherent ideals corresponding to the centers and (ii) an explicit
local formula for a Chern form associated to this singular blow-up. Our metrics have a particularly simple local
formula of a sum of the original metric and of the pull back of the classical Poincar metric on the punctured disc by
a size-function SI of a coherent ideal I used to resolve the singularities of X by a singular blow-up, where
(S I )2 := rj =1 f j
and the fj's are the local generators of the ideal I. Our proof of (i) makes use of our
generalization of Chow's theorem for coherent ideals. We prove Saper type growth for our metric near the singular
set and local boundedness of the gradient of a local generating function for our metric, motivated by results of
13
Donnelly-Fefferman, Ohsawa, and Gromov on vanishing of certain L2-cohomology groups. In an appendix we give
a simple constructive proof of a valuation criterion due to M. Lejeune and B. Teissier.
This 68-page joint paper has been accepted for publication in the Annales de la Facult des Sciences de
Toulouse.
Investigations into Sudoku

Researchers: Associate Professor T. S. Michael and Professor Mark D. Meyerson
This research in progress investigates extensions of a su doku problem that recently appeared in the
American Mathematical Monthly. We have constructed a set of four su doku squares, each of which is orthogonal to
the other three - and to its mirror-image about the main diagonal. The notion of orthogonality considered here is in
the combinatorial sense of Latin squares.
Maple Animation of Space Filling Curves

Researcher: Professor Mark D. Meyerson
Dr. Meyerson is writing Maple programs to help with the visualization of classical Space Filling Curves.
By varying the usual definition, one can produce families of curves with fractional (between 1 and 2) dimensions
(fractals).
Viewing Mbius Transformations with Maple Animations

Researcher: Professor Mark D. Meyerson
We use Maple's ability to easily draw and animate to examine all possible Mbius transformations. A grid
on the complex sphere stereographically embedded in 3 is distorted conformally as it continuously varies to the
image under any Mbius transformation. There are two different cases: one or two fixed points (parabolic and
loxodromic).
Wrestling Tournaments: How to Select First-Round Byes

Researcher: Associate Professor T. S. Michael
The Eastern Intercollegiate Wrestling Association (EIWA) encounters a scheduling problem at the seasonending conference tournament: How should the first-round byes be distributed among the wrestlers in each weight
class? For many years a random selection was used, but this caused controversies when some teams received many
more byes than others. I discovered a new mathematical method that is fair to individuals and teams. The new
method was implemented at the 2006 EIWA tournament, which is the regional qualifier for the NCAA
championship meet. The method performed well, and will be adopted at future EIWA tournaments. Because Navy is
a member of the athletic conference, the new method has a direct impact on midshipmen in athletic competition.
Tiling with Three Squares

Researchers: Associate Professor T.S. Michael and Associate Professor Courtney Moen
This research in progress investigates a particular tiling problem in the spirit of the Frobenius coinexchange problem. We are given three squares with pairwise relatively prime edge lengths. We seek the largest
integer square that cannot be tiled with the three given squares.
Nonlinear Interaction of Light with Liquid Crystals

Researcher: Assistant Professor Aurelia Minut
In this research, we propose a new model of Partial Differential Equations (PDE's) in the field of nonlinear
optics of liquid crystals to model the evolution of nematic liquid crystals in the presence of an electromagnetic field.
The current research in the field of liquid crystals uses the Oseen-Frank model, the Ericksen model or the EricksenLeslie theory and no electromagnetic fields are present. There are indications from this research that the Oseen-
14
Frank theory needs modifications in order to fit real physical phenomena. We provide modifications by coupling this
equation with Maxwell's Equations. One particular application we have in mind is to use this theory to model the
propagation of light through liquid crystals and compare it with the propagation of light through optical fibers. We
are currently working on the model for the system of differential equations and will study the existence of solutions.
Competition in Two-Dimensional Heterogeneous Environments

Researchers: Dr. Daniel Kern (University of Nevada - Las Vegas)
and Assistant Professor Aurelia Minut
Competition in a heterogeneous environment is studied in two spatial dimensions. We examine the case of
two competing populations (one native, one invasive) expanding in a random manner through a domain alternating
between two patch types. We focus on a small population of the invasive species being introduced in a limited
portion of the environment. We developed the model of differential equations and among the main goals for these
equations are:
Examining boundaries, especially between patch types,
Condition for successful invasion -- stability analysis, and
Traveling waves and wave propagation speed.
On a New Notion of Ellipticity

Researchers: Professor George Nakos
and Associate Professor Christopher W. Brown (USNA Computer Science Department)
In this work we generalize the notion of the Douglis-Nirenberg ellipticity (DN-ellipticity) for square
systems of partial differential equations. DN-ellipticity was introduced in 1955 for the following reason: under the
new notion ellipticity can be preserved when a higher order elliptic equation or system reduces to a lower order one.
Unfortunately, DN-ellipticity involves the choice of a weight system that makes the definition noninvariant under
singular transformations. This lack of good functoriality properties of DN-ellipticity is a serious obstacle in
developing properties of solutions for such systems. In this research we propose a new notion of ellipticity that has
all the desired properties and includes all DN-elliptic systems both square and rectangular and does not involve any
choice of weights.
Mathematical Foundations for Statistical Learning

Researchers: Associate Professor John F. Pierce and Associate Professor Gary O. Fowler
We are investigating how to develop measures to assess the reliability of empirical estimators for the
currents, salinity and temperature gradients in the Chesapeake Bay, and to assess bounds on the level of confidence
of decisions based upon them about accepting or rejecting mathematical models for the dynamics in the bay based
upon the numerically approximated systems of partial differential equations that arise from hydrodynamic and
thermodynamic principles.
The works of S. Smale, R. Devore and others that link results from the theory of functional analysis to
problems of statistical learning provide the rational foundation for this form of empirical reasoning. Their tools
specify more subtle bounds on rates of convergence for statistical estimators that, in principle, can be applied to the
empirical datasets of finite size being collected in the Chesapeake Bay and to the simulations developed from the
mathematical models of the bay. In principle, these estimators can be used to ascertain: (1) Are the empirical data
being collected even relevant for discerning the validity of the mathematical models? (2) If not, does the source of
the discordance lie in the nature of the mathematical models, in the assumptions that lead to the numerical
approximations and simulations, or in the experimental or sampling design which produces the empirical data?
A prototypical problem that demonstrates how to engage these functional analytical tools with empirical
data was fashioned at Cornell University by T. Joachims when he formulated a rational model for assessing designs
for creating learning machines capable of distinguishing documents according to subject. The endeavor for the fall
semester involves understanding how Joachim's design of the estimators for the elements in the a-priori bounds in
the functional analytical tools alter the design of the sampling of the texts and the design of the vector of features
erected from them. The endeavor for the spring semester will be to adapt this paradigm to the context of the
Chesapeake Bay.
15
A Dynamical Systems Approach to Cardiac Arrhythmias

Researchers: Assistant Professor Irina Popovici, Professor B. Mitchell Baker and Professor Mark Kidwell
This project concerns the properties of an iterated system proposed by B. M. Baker and R. Kline to model
the behavior of the heart, based on the connection between membrane current kinetics, action potential duration and
cardiac rhythm. The starting point is a dynamical system incorporating one or two kinetic parameters which control
repolarization, and hence action potential duration. A simply formulated family of even one-dimensional dynamical
systems is mathematically capable of producing restitution curves, bifurcation schemes, rhythmic response and
chaotic behavior mimicked by cardiac preparations. The family of one- and two-dimensional maps in which the
kinetic parameters are modeled follow three basic assumptions: 1) exponential increase of the parameter until a
threshold for repolarization is reached, 2) exponential decay of this parameter immediately after threshold is
reached, 3) a periodic constraint. The parameter sets we studied consisted of points where the dynamical system
transitions from nearly-periodic behavior to chaos for the models that include a refractory period and for the twodimensional (two loader) model.
Locating Edges and Removing Ringing Artifacts in JPEG Images

by Frequency-Domain Analysis
Researchers: Assistant Professor Irina Popovici and Professor W. Douglas Withers
We present a method of locating edges in JPEG-coded images which operates in frequency space on the
DCT coefficients. Applied to the quantized DCT coefficients of a block containing a sharp, straight edge, the
method an equation for the edge in a fraction of the operations needed to dequantize and transform the coefficients
to pixel values. As a sample application of this method, we present a technique for alleviating ringing artifacts in
JPEG-coded images.
A Dynamical Systems Approach to Membrane Phenomena

Underlying Cardiac Arrhythmias
Researchers: Assistant Professor Irina Popovici, Professor B. Mitchell Baker and Professor Mark Kidwell
We study a model of the human heart cells as a family of dynamical systems parameterized by the
concentrations of potassium and magnesium. The two-dimensional maps follow three basic assumptions: 1) increase
of the parameters until a threshold for re-polarization is reached, 2) exponential decay of these parameters
immediately after threshold is reached, 3) a periodic constraint. The principal results give the existence and stability
properties of orbits analogous to the so called escalator orbits of the one-dimensional family, the co-existence of
stable orbits for fixed values of the period parameter, existence of bunny-ears orbits. We consider the latter results
our most important, owing to conjectures that such period parameter values could produce arrhythmic behavior in a
potentially spatially disorganized fashion.
Stochastic Generalized Assignment Problem

Researcher: Commander David Spoerl, USN
Solving a stochastic version of the Elastic Generalized Assignment Problem (EGAP), which incorporates
independent, identically distributed resource-consumption coefficients. The Stochastic Elastic GAP (SEGAP) is a
two-stage stochastic integer program with simple recourse. We construct a deterministic equivalent and solve using
the branch-and-price algorithm. Currently I am completing the model and the math program to solve the SEGAP.
Differential Operators on Orbifolds

Researcher: Assistant Professor William N. Traves
We study differential operators on orbifolds, the quotient of a vector space by a finite group action.
Generators and relations for the ring of invariant differential operators are presented. These are also used to compute
the ring of differential operators on the orbifold. The techniques used involve Grobner bases and a careful study of
the map from the ring of invariant differential operators to the differential operators on the orbifold.
16
The Fundamental Theorem of Invariant Theory and Differential Operators

on Grassmannians
Researcher: Assistant Professor William N. Traves
A presentation for the ring of invariant differential operators on an affine cone over a Grassmannian variety
is presented. Generators can be obtained directly from the fundamental theorem of invariant theory, applied to an
associated graded ring; more care is needed to lift the relations to the ring of operators. A valid lifting of the
relations is presented and it is shown that these, together with the commutator relations, generate the entire ideal of
relations. Finally, the map g from the invariant operators to the ring of invariants is used to obtain a presentation of
the ring of differential operators on a Grassmannian variety. This is done explicitly for G(2,4), the variety
parameterizing two-planes in four-space (projective lines in projective 3-space). Finally a conjecture is presented
that describes the kernel of the map g.
Enumerative Algebraic Geometry: Counting Conics

Researchers: Ensign Andrew C. Bashelor, USNR (Class of 2005 Trident Scholar)
Assistant Professor Amy E. Ksir and Associate Professor William N. Traves
This project began as Midshipman Bashelors Trident Scholar project in 2004-2005. The problems studied
were generalizations of Steiners Problem, which asks, How many conics are tangent to five given plane conics?
Bashelor solved all of the twenty-one relevant generalizations of this problem, and wrote up these results in his 2005
Trident Report.
Algebraic geometry has a reputation for being very difficult to learn; most introductory books and articles
on the subject are aimed at second year graduate students and start by introducing a lot of heavy mathematical
machinery. We realized that Steiners problem and Bashelors solution of it would make a very nice expository
article, introducing some of the key ideas and tools of algebraic geometry to an undergraduate audience. Over the
course of this year, we wrote this expository article and submitted it to a leading mathematics journal aimed at a
broad audience. While the mathematical content of this article is largely the same as that of Bashelors Trident
report, the focus and explanation are entirely new.
Quadratic Forms and Cohomology

Researcher: Associate Professor JoAnn S. Turisco
The relatively new methods of motivic homotopy, developed by V. Voevodsky, (for which he received
the Fields Medal in 2002) have lead to many advances in the algebraic theory of quadratic forms. This theory in
algebraic geometry is similar to homotopy theory in algebraic topology. The motivic homotopy category provides a
motivic cohomology, K-theory, and algebraic cobordism.
My work involves the application of motivic methods, along with the more established methods using
Galois cohomolgy, to compute some of the several invariants associated to quadratic forms. Many new results in
quadratic form theory have been obtained using motivic methods. In particular, a Hopf condition for the
composition of quadratic forms over fields of characteristic greater than 2 has been proved fairly recently using
motivic cohomology. This problem has interested me for years and I am currently working to extend the previous
results.
Menu Program for using Stats/List Editor on Voyage 200

Researcher: Professor John C. Turner
This work is a direct outcome of my work for SM230 (and SM219 previously). The Voyage 200 and TI92
include a powerful statistical package. However, the default interface has several serious deficiencies. The biggest is
that after a problem has been worked, it is not possible to return to previous calculations. This lack of a paper trail
inhibits the student from comparing results from previous problems. It also prevents a student from learning from
mistakes, say, on an exam. When the exam is returned, it is not possible to verify what the student actually entered
into the calculator. Thus, it is not possible for the student to see how to correct this work in the future.
17
I used the Custom Menu feature of the calculator to create an interface to the Stats/List programs. This
approach leaves the inputs and outputs from previous problems on the home screen. This makes it easier for the
student to see what was done wrong. It also allows the instructor to see exactly what was typed in for a given
problem, rather than what the student thought or meant to type in. Also, the ability to save the home screen (and all
the student inputs) provides a way to verify the work on an exam.
A Fast Method for Determining Taylor Series Weights for Finite Difference
Approximations of Derivatives on Uniform Grids
Researcher: Commander Vincent Van Joolen, USN
Finite difference approximations for derivatives provide an important means for numerically solving
systems modeled with ordinary and partial differential equations. Taylor series polynomials are a major conveyance
by which finite difference expressions are generated. Though basic expressions for the lower derivatives are well
known, easily derived, or readily available in mathematical tables, expressions involving higher order derivatives or
that sample a greater number of grid points are more difficult to come by. Here a scheme is proposed that will allow
the reader to theoretically calculate finite difference weights to approximate any derivative using any combination of
grid points.
Matrices over Commutative Rings

Researcher: Professor William P. Wardlaw
For years I have been interested in matrices over commutative rings. I have continued my research into
such matrices with special attention given to (1) the various definitions of the rank of a matrix over a commutative
ring and (2) the characterization of the classical adjoint (adjugate) of a matrix over a commutative ring. This has led
to a short proof that row rank equals column rank which was published in Mathematics Magazine, as well as a paper
with Associate Professor Jody Lockhart on determinants of matrices over the integers modulo m, which has been
accepted by Mathematics Magazine.
Various Mathematical Problems

Three problems with solutions have been prepared for submission to mathematical journals. These were
done in collaboration with Associate Professor Courtney Moen, Assistant Professor Amy Ksir, and Lieutenant
Joshua Wood, USN.
A Modified Hill Cipher

Several modifications to the cipher invented in 1931 by Lester S. Hill are suggested to make the cipher
more secure. This paper has been accepted for publication in Cryptologia upon revision suggested by the referees.
Paving Small Matrices

Researchers: Assistant Professor Vrej A. Zarikian and Gary Weiss (University of Cincinnati)
Andersons Paving Conjecture states that for every > 0, there exists a positive integer k such that every
norm-one matrix A with zero diagonal admits a k-paving of norm less than or equal to 1 . The truth of the Paving
Conjecture is equivalent to an affirmative answer to the Kadison-Singer Problem, a famous question in operator
algebra theory, open since 1959. Owing to recent work of Casazza et al. showing that the Kadison-Singer Problem is
equivalent to a number of important questions in math and engineering, there is renewed interest in the KadisonSinger Problem, and therefore the Paving Conjecture. Together with Gary Weiss, we have investigated whether
every matrix is 3-paveable (it is known that not every matrix is 2-paveable). Using a combination of operator theory,
graph theory, and computer experimentation, we have produced examples of 7-by-7 circulant matrices whose 3paving norm is 0.8231. Before our work, the highest known 3-paving norm was 0.6667, and it was only attained for
very large matrices. We are hopeful that we can produce a small matrix whose 3-paving norm is 1.000, thereby
disproving the Paving Conjecture for the case k = 3 (we believe that the Paving Conjecture is false).
18
Alternating-Projection Algorithms for Operator-Theoretic Calculations

Researcher: Assistant Professor Vrej A. Zarikian
We show how alternating-projection algorithms can be used to solve a variety of operator-theoretic
problems, including deciding complete positivity, computing completely bounded norms, computing norms of Schur
multipliers, and matrix completion/approximation problems.
Alternating-Projection Algorithms for Operator-Theoretic Calculations

In the recently developed theory of operator spaces, the important linear transformations are not the
bounded ones, but rather the completely bounded ones. Often, problems in operator space theory come down to
computing the norm (or size) of such a linear transformation. Unfortunately, such calculations are usually
impossibly difficult, and one must be content with a reasonable estimate. In finite-dimensions, however, one would
hope that a numerical algorithm could be devised to compute completely bounded norms. I succeeded in doing just
that. The key realization was that computing completely bounded norms reduces to deciding whether certain convex
sets intersect, a problem which can be solved using von Neumanns alternating-projection algorithm from 1933.
One-Sided M-Ideals and Multipliers of Operator Spaces

The M-ideal theory of Alfsen and Effros, which dates to 1972, is an important tool in the study of Banach
spaces. In my dissertation under Effros, I generalized M-ideals to operator spaces, introducing the notion of a onesided M-ideal. One-sided M-ideals are best studied using Blechers theory of one-sided multipliers. Together with
Blecher, I have advanced the theory of one-sided M-ideals and multipliers considerably (see Attachment #4).
Consequences include streamlined proofs of characterization results for operator algebras and operator modules. The
current project attempts to resolve some of the many remaining questions. For example, the proximinality question:
Does every element of an operator space have a best approximation from the one-sided M-ideal?
Midshipman Research Course Projects

Justice in Liver Transplant Allocation, an Integer Programming Approach
Researcher: Midshipman 1/C Peter Barkley, USN
Adviser: Assistant Professor Sommer Gentry
The existing system for distributing donated livers is geographically inequitable. As livers are given to the
neediest person within the region, people in great need of a liver do not receive one in some regions, while people
with much less need receive a transplant in other regions. We show how to make the system more just, by
redistricting the regions so that the difference in the boundary need level at which people receive organs is
constrained to be small. Organs have a limited maximum shipping distance, and political constrains have frequently
hindered attempts to modify the regional system. Furthermore, the time between removal and transplation (Cold
Ischemia Time) is directly correlated with the subsequent success of the transplant. Our model, therefore, maintains
the current number of regions at 11, and minimizes the size of the regions while constraining the justice of the
distribution. Using methods similar to those used in school redistricting problems, we have designed an integer
program to minimize the distance between a seed geographical unit for each region and all the other geographical
units contained in that region. This minimization is constrained by a limit on the difference between the levels of
neediness at which each region runs out of donors to provide organs. The completed research constituted Mr.
Barkleys honors project, which won the math departments Military Applications Award for a thesis in operations
analysis.
19
Linear Feedback Shift Registers and Cyclic Codes in Sage

Researcher: Midshipman 1/C Timothy B. Brock, USN
Adviser: Professor W. David Joyner
Honors Project 2006: This paper discusses the history of linear feedback shift registers (LFSR) in
cryptographic applications and will attempt to implement an algorithm in SAGE and Python (www.python.org) to
create a linear feedback shift register sequence (LFSR sequence) in cryptography. Also, this paper will attempt to
implement the Berlekamp Iterative Algorithm in SAGE and Python. This algorithm will be able to use the Linear
Feedback Shift Register sequence generated by the first algorithm to find the sequences connection polynomial.
The research attempted to show that the connection polynomial of a given LFSR sequence is the reverse of
a generator polynomial of the cyclic code of length p, where p is also the period of the LFSR sequence. This will
provide a connection between cyclic error-correcting codes and LFSR sequences.
Long Quadratic Residue Codes

Researcher: Midshipman 1/C Greg Coy, USN
A long standing problem has been to develop good binary linear block codes to be used for error
correction. The goal of this project was to investigate a family of codes called Quasi-Quadratic Residue (QQR)
codes. The numerical and theoretical investigation of these codes yielded little in finding a good family to be used
in error-correction. However, an interesting family of codes was discovered called Long Quadratic Residue (LQR)
codes. These codes are based off the construction of the QQR codes. The LQR codes have good parameters,
essentially meaning that their information rate and relative minimum distance are bounded away from zero. Further,
I have shown that this family of curves is incompatible with the fake Goppa conjecture, since this family of codes
beats the Gilbert-Varshamov bound for binary codes. Most importantly, I have used LQR codes to develop the
small content conjecture. This essentially shows that the Goppa conjecture is incompatible with a conjecture on
hyperelliptic curves over finite fields.
Nearly-Involutive Matrices for the Keyspace of the Hill Cipher

Researcher: Midshipman 1/C Kindle Clarke, USN
Advisers: Professor George Nakos and Assistant Professor Alexis A. Alevras
The Hill Cipher was an important cryptosystem of classical cryptology in the 1930s (Hill 2; 3). The basic
idea of the Hill Cipher is you first pick a modulus m, where m is a positive integer m>1. A message is converted
into an n-vector x with components in Zm. Let A be in GL(n, Zm), so A is an invertible matrix modulo m with entries
in Zm. The encryption of x is y=Ax and therefore the decryption is x=A-1y. In 1931, Hill suggested that the use of
involutive matrices, i.e. matrices satisfying the identity: A = I
2
This project studied of the number of solutions of the closely related matrix equation: X = cI where
c Z p , p being a prime and I is the identity matrix. Results were obtained for 2x2 and 3x3 matrices.
2
Singular Analysis of an N-dimensional Cosmic String

Researcher: Midshipman 1/C Gregory Dietzen, USN
Advisers: Professor Deborah A. Konkowski, Professor Mitch Baker and Assistant Professor Alexis A. Alveras
Cosmic strings are theoretical vestiges of our early universe which may help us understand its formation.
Their study is also intriguing because of their possible link to M-Theory. In this research Midshipman Dietzen
strove to understand the singularities, both classical and quantum mechanical, of an n-dimensional cosmic string.
The study began with a study of the literature for 2- and 4-dimensional cosmic strings and then an analysis
of the next most difficult case, a 5-dimensional cosmic string. Classical singularities were studied using MAPLE for
a curvature singularity check and, finding no curvature singularities, a topological analysis was completed which
showed the existence of a quasiregular singularity. Classical scalar waves were then studied. This involved writing
and solving the general relativistic Klein-Gordon equation. With the classical analysis in hand quantum singularities
20
were next studied. This involved Midshipmen Dietzen learning some of the basics of operator theory in functional
analysis and the application of this knowledge to this particular case. Quantum singularities were found for some,
but not all, cases. Midshipman Dietzen then tackled the n-dimensional cosmic string and found similar classical and
quantum singularity structure.
In conclusion, Midshipman Dietzens analysis showed that a cosmic string has a classically singular
spacetime. It is cone-shaped causing a topological defect which made it impossible to analyze using the usual
curvature analysis of Einsteins relativity. However, basic quantum theory could be applied showing the ndimensional string may or may not be quantum mechanically singular. He successfully found the range of
parameters when it is, values that must be determined from the study of each wave mode and each individual string.
Dynamic Systems: Spring, Cycloid and Pendulum Systems

Researcher: Midshipman 1/C John H. Doherty, USN
Adviser: Professor Mark Kidwell
The simplest of the dynamic systems that was examined in this project is the motion of a mass attached to a
spring under the influence of an outside force such as an electromagnet. This produced a simple linear ordinary
differential equation where the position x is dependent upon a single variable t. The particular solution was found
with simplistic methods learned in an introductory course on differential equations.
The second system that was studied is the cycloid. This system can be physically represented by tracing
the path of a specific point of a bicycle wheel while the wheel moves along a surface. Although it is not intuitively
obvious, the upside down arc of the cycloid also plots the motion of the quickest possible path of a bead sliding
along a string acting solely under the influence of its own gravity traveling between two points as determined by
Johann Bernoulli in the Brachistochrone Problem. For the purposes of this study, the motion of a marble along a
frictionless and dragless Brachistochrone was considered. It was revealed that the motion of this marble had
numerous interesting properties under these circumstances including that it maintains constant period for any initial
displacement.
Another conceptually similar but more complicated dynamical system is a pendulum with an external
torque acting upon it. For this problem, the external torque acting upon the pendulum bob was either zero or a
constant value. However; the torque due to the weight of the bob changed as theta changes. The perpendicular
distance from the point of rotation served as the effective torque arm. This distance was equal to the product of the
length of the pendulum arm and the sine of the angle theta between the pendulum arm and a vector normal to the
ground. Therefore, the mass times the second derivative of theta with respect to time equaled the sum of the
moment due to the torque and the moment due to the weight. This can be expressed mathematically as follows:
m*L*
d 2 ( (t ))
= m * L * sin( (t )) + T0
d (t ) 2
It was revealed that these three unique systems have numerous interrelated properties and provide a
fascinating application of differential equations.
Higher Dimensional Linear Regression

as a Mathematical Foundation for Data Classification
Researcher: Midshipman 1/C Robert B. Irving, USN
Adviser: Associate Professor Gary O. Fowler
With the ultimate goal of constructing a classification machine which uses only the information contained
within a training sample in order to classify as-yet-unseen data points, this project built upon Shawe-Taylor and
Christianinis mathematical analysis and analyzed the geometrical features of their perspective (Shawe-Taylor, 25).
The author begin by creating a geometric model for the more familiar regression problem and then applied similar
techniques to the classification problem. More specifically, it was shown how higher-dimensional linear regression
and classification can be captured by embedding the sample data into a higher-dimensional space of features,
analyzing this embedded data with simple geometric tools, and then projecting the model back into our original data
space to obtain intricate fits (regression) or separations (classification).
21
Mathematics Honors Project Summary: Low Density Parity Check Codes

Researcher: Midshipman 1/C Gordon R. McDonald, USN
This honors project investigated the subject of Low Density Parity Check codes (LDPC codes). In it the
author attempted to define a general set of conditions that good' Low Density Parity Check codes each contained
two iterative methods were also investigated which theoretically decoded LDPC codes. Those algorithms are called
Bit-Flipping and Gallager Hard Decision decoding. The algorithms were illustrated by means of detailed examples
and implemented in GAP, an open source computer algebra system. Furthermore, the functions that were written in
GAP are now part of the newest GAP update. Overall, the author was highly unsuccessful in generating LDPC
codes which were of any use. This lead to some difficulty in implementing the few good' LDPC codes which
appeared in two decoding algorithms. In conclusion, it was found that the decoding algorithms depend upon very
specific constructions of LDPC codes which are oftentimes too idealistic to obtain using the method of construction
attempted. There is, however, an alternative method to construct LDPC codes which could possibly make the
decoding algorithms much more successful. This could easily be investigated with further research.
Dynamical Systems and Irrational Angle Construction by Paper-Folding

Researcher: Midshipman 1/C Cayanne McFarlane, USN
Adviser: Professor W. Douglas Withers
Hilton and Pedersen have previously shown how to construct any rational angle to arbitrary precision by a
periodic sequence of folding a strip of paper. We show how this method can be generalized to construct any angle,
rational or irrational, by using a connection between the method and a simple dynamical system to generate the
appropriate aperiodic folding sequence.
Gods Algorithm on the Edges and Corners of the Rubiks Cube

Researcher: Midshipman 1/C Daniel C. Ryan, USN
Adviser: Assistant Professor Amy E. Ksir
Rubiks cube is a puzzle that can be represented using group theory. Gods Algorithm is the unknown
answer to the following question: Given a Rubiks cube in its maximal state of disorder, what is the smallest
possible number of moves necessary to solve it? In this project, Midn Ryan tackled the two smaller, related
problems: finding Gods Algorithm for just the edges of the Rubiks cube, and finding Gods Algorithm for just the
corners of the Rubiks cube. Midn Ryan used group theory and his knowledge of the Rubiks cube to tackle the first
part of the problem, which was to understand the structure of the Rubiks cube group. For the second part of the
project, he was able to adapt some existing computer programs towards finding the solution. He was able to confirm
intermediate results, and only lack of computer power stopped him from obtaining a final result.
Profession A Probability City Patrol

Researcher: Midshipman 1/C William Sumpton, USN
Advisers: Lieutenant Commander Kyle Kliewer, USN, and Professor Emeritus Charles Mylander
The United States Marine Corps currently engages in city patrols in hostile areas, most noticeably cities
like Baghdad and Fallujah. A searchers ability to visually cover an area from its perimeter depends on how far the
lines of sight, propagating from the perimeter inward, can travel before becoming obstructed. The probability of
finding a target in the city block is then proportional to the area covered and time spent by the search. A battlefield
commander could use information as to what areas of a city can be covered by an observer on the ground to better
implement his assets.
The search of a block of structures in a city is modeled using a combination of matrices and line of sight
from a variety of observation positions based on searching around the parameter. From the data, two specific
observations are made: what area can the observer see and not see, as well as how many times a location is spotted
from a parameter position. The probabilities of finding a target for the entire block as well as specific locations in
the interior are computed. These probabilities will be affected by different parameters such as a diminishing line of
sight with distance, a randomly moving target, as well as a target that moves between predetermined points. Safer
ingress routes into the block can then be realized by examining the results of the computations.
22
Publications
Journal (Refereed) Manuscripts
ALEVRAS, Alexis A., Assistant Professor, and PRICE, Geoffrey, Professor, (co-authors), Cocycles for Oneparameter Flows of B(H), Journal of Functional Analysis, 230 (2006), pp. 1-64.
The set of local cocycles is a natural invariant for an E0-semigroup. It has a multiplicative structure, as well
as a partial order structure among its positive elements. In particular, the unitary local cocycles form a
topological group which may be appropriately viewed as the automorphism group of the E0-semigroup,
while the set of positive contractive local cocycles is order isomorphic to the set of flows of completely
positive maps dominated by the semigroup. The local cocycles have been computed for the standard, type I
examples of the CAR/CCR flows by W. Arveson and R. Bhat. In this paper we compute for the first time
the local cocycles for a class of type II E0-semigroups of B(H) with index zero, and describe the order
structure as well as the multiplication in terms of the boundary weight associated with such a semigroup.
ALEVRAS, Alexis A., Assistant Professor, The Gauge Group of an E0-semigroup, Contemporary Mathematics, in
press.
We discuss the role of the group of unitary local cocycles in the study of E0-semigroups of B(H) and
describe the computation of that group for a class of E0-semigroups of type II and index zero.
BUCHANAN, J.L., Professor, Gilbert, R.P., and Xu, Y., Ultrasound as a Diagnostic Tool to Determine
Osteoporosis, Advances in Analysis, pp. 345-354, World Scientific.
Cancellous bone is known to be poroelastic in structure. Ultrasonic wave propagation in cancellous bone
can be formulated by using Biots equations. In this paper we present some results in our ongoing research
on the reflection and transmission of ultrasonic waves in cancellous bone. We investigate the relations
among reflected waves, transmitted waves and the Biot coefficients. We present an algorithm for the
determination of the porosity of cancellous bone.
CRAWFORD, Carol G., Professor, 6 Steps to Successful, Interactive Math WEB Design - An Innovative Real
World Calculus Laboratory Designed with Java for the United States Naval Academy, Proceedings of the Joint
Meeting of the 3rd International Conference on Education and Information Systems, Technologies and Applications
(EISTA 2005) and the International Conference on Cybernetics and Information Technologies, Systems and
Applications (CITSA 2005). Volume I, pages 223-228.
The author presents 6 Steps to help in the design of an effective, interactive mathematics resource for
higher education. These steps reflect her work with Professor Mark Meyerson in the creation of an
innovative, online Real World Calculus Laboratory for the 3- semester Calculus sequence taken by all
midshipmen at The United States Naval Academy. The Curriculum Development Program at the Academy
funded the research for this two-year design project.
GAGLIONE, Anthony M., Professor (co-author), Notions of Discrimination, Communications in Algebra,
September 2005, in press.
As an outgrowth of the study of algebraic geometry over groups, discriminating groups were introduced.
Many important universal type groups such as Higmans universal group and Thompsons group F were
shown to be discriminating. Squarelike groups were then introduced to better capture axiomatic properties
of discrimination. In the present article squarelike groups are reinterpreted in terms of discrimination of
quasi-varieties, and the relationship with an older version of discrimination, termed varietal discrimination
here, is studied.
23
GAGLIONE, Anthony M., Professor (co-author), An Embedding Theorem for Groups Universally Equivalent to
Free Nilpotent Groups, Groups St. Andrews 2005, November 2005, in press.
Let F be a finitely generated nonabelian free group. Kharlampovich and Myasnikov proved that any
finitely generated group H containing a distinguished copy of F is universally equivalent to F in the
language of F if and only if there is an embedding of H into Lyndon's free exponential group FZ[t} which is
the identity on F. Myasnikov posed the question as to whether or not a similar result holds for finitely
generated free nilpotent groups with Lyndon's group replaced by Philip Hall's completion with respect to a
suitable binomial ring. We answer his question here in the affirmative.
GAGLIONE, Anthony M., Professor, (co-author), Reflections on Discriminating Groups, Journal of Group
Theory, February 2006, in press.
Here we continue the study of discriminating groups as introduced by Baumslag, Myasnikov and
Remeslennikov in [BMR2]. First we give examples of finitely generated groups which are discriminating
but not trivially discriminating in the sense they embed their direct squares and then show how to
generalize these examples. In the opposite direction we show that if F is a nonabelian free group and R is
normal in F such that F/R is torsion free, then F/R, where R=[R,R] is the commutator subgroup of R, is
nondiscriminating.
GAGLIONE, Anthony M., Professor (co-author), The Search for Origins of the Commutator Calculus,
Proceedings in Honor of Gehard Rosenberger, April 2006, in press.
The origins of the commutator calculus and the relationships between some of the key investigators are
discussed. Also their subsequent work in the Age of the Enigma,1930-1950,is presented. Several questions
about the history are proposed.
GAGLIONE, Anthony M., Professor (co-author), A Note on Nondiscrimination of Nilpotent Groups and Mal'cev
Completions, Proceedings in Honor of Gehard Rosenberger, April 2006, in press.
In this note we prove that if a finitely generated nilpotent group is discriminating then so is its Mal'cev
completion. From this, we recover that a finitely generated nilpotent group can be discriminating only if it
is torsion free abelian. This was proved in a very different way by Myasnikov and Shumyatsky but the
method of this paper should shed additional light on nondiscrimination in an alternate manner.
GAGLIONE, Anthony M., Professor, (co-authored), Finitely Presented Infinite Torsion Groups and a Question of
V.H. Dyson, Proceedings in Honor of Gehard Rosenberger, April 2006, in press.
Let L be a first-order language appropriate for group theory. The universal theory of a class of groups is the
set of all universal sentences of L true in every group in this class. In this note, it is shown that if the
universal theory of the torsion groups coincides with the universal theory of the finite groups, then there
cannot exist a finitlely presented infinite group of finite exponent.
GAGLIONE, Anthony M., Professor (co-author), Unions of Varieties and Quasivarieties, Proceedings in Honor
of Gehard Rosenberger, April 2006, in press.
In this paper we characterize unions and direct unions of varieties and quasivarieties of operator algebras in
terms of closure properties. In terms of groups these apply to many important classes such as nilpotent
groups, solvable groups and equationally Neotherian groups.
GENTRY, Sommer, Assistant Professor (co-author), Kidney Paired Donation: Optimizing the Use of Live Donor
Organs, Journal of the American Medical Association, vol. 293, pp.1883-1890, 2005. This paper was awarded the
2005 Vanguard Prize from the American Society for Transplant Surgeons.
Blood type and crossmatch incompatibility will exclude at least one third of patients in need from receiving
a live donor kidney transplant. Kidney paired donation (KPD) offers incompatible donor/recipient pairs the
24
opportunity to match for compatible transplants. We designed a mathematically verifiable optimized

matching algorithm and compared it with the scheme currently used in some centers and regions. We
simulated patients from the general community with characteristics drawn from distributions describing
end-stage renal disease patients eligible for renal transplantation and their willing and eligible live donors.
Even if only 7% of patients awaiting kidney transplantation participated in an optimized national KPD
program, the health care system could save as much as $750 million. Optimized matching affords patients
the flexibility of customizing their matching priorities and the security of knowing that the greatest number
of high-quality matches will be found and distributed equitably.
GENTRY, Sommer, Assistant Professor (co-author), A Comparison of Populations Served by Kidney Paired
Donation and List Donation, American Journal of Transplantation, 2005, August, 5(8), pp.1914-1921.
Options for utilizing live donor kidneys from those who are blood type incompatible or crossmatch positive
with their intended recipients include kidney paired donation (KPD), list paired donation (LPD) and
desensitization. We simulated patients and their potential donors to determine which recipients could
receive a kidney through KPD and LPD. At population sizes predicted to be achieved by a national paired
donation system, the role of LPD became minimal, with only 3.9% of pairs unmatched through KPD
eligible for LPD. Considerable overlap was seen between the pairs unmatchable by KPD and those
ineligible for LPD, namely less-demanded donors and hard-to-match recipients. For this population, the
best option may be desensitization.
GENTRY, Sommer, Assistant Professor (co-author), Characterization of Waiting Times in a Simulation of Kidney
Paired Donation, American Journal of Transplantation, 2005, October 5(10), pp.2448-2455.
A national kidney paired donation (KPD) program will substantially increase transplant opportunities for
recipients with blood type incompatible or cross-match positive donors. It seems likely that donor-recipient
pairs with certain blood types, races or restrictions will wait longer than others for a match, although no
data exist to confirm this assumption. We simulated patients and characterized the predicted waiting times
for different blood type sub-groups, as well as the effects of patient-imposed restrictions on waiting time.
We also compared waiting times of different racial sub-groups. These data provide the first waiting time
predictions that can aid patients with incompatible donors in choosing between KPD and desensitization,
and can facilitate planning for a national paired kidney donation program.
GENTRY, Sommer, Assistant Professor (co-author), Multiplying the Benefit of Live Non-directed Donors through
Domino Kidney Paired Donation, Lancet, in press.
The allocation of living nondirected donor (LNDD) organs has been at the discretion of individual
transplant programs, including the model that governs recipient selection. In all cases, the LNDDs gift has
resulted in a single transplant. We demonstrate using optimized allocation on a simulated patient databases,
that an LNDD donation can be multiplied by allocating the kidney to a recipient with an incompatible
donor who is willing to donate their kidney to a patient on the deceased donor waiting list, producing a
domino effect. We also present a case report of an LNDD kidney that was allocated to a domino paired
donation (DPD) and, as a result, two transplants were accomplished. This allocation strategy is then
discussed in the context of the ethical arguments that guide current models of allocation for LNDD.
GENTRY, Sommer, Assistant Professor (co-author), Relative Roles for List Paired Exchange, Live Donor Paired
Exchange, and Desensitization, American Journal of Transplantation, 2006, February, 6(2), p. 437.
This letter is an invited response to Morrissey P., in support of list paired exchange, American Journal of
Transplantation February, 6(2): 2006. We describe our data showing that list paired exchange will
become less important as living donor paired exchange becomes more widely available. Our simulated
matching results also suggest certain subgroups of patients are unlikely to be served well by either of these
two modalities and might benefit most from desensitization.
25
HOFFMAN, Michael E., Professor, Symmetric Functions, Quasi-symmetric Functions and Rooted Trees,
Oberwolfach Reports, in press.
The Hopf algebra of symmetric functions is a subalgebra of the larger Hopf algebra of quasi-symmetreic
functions. Similarly, there is an injective map of Hopf algebras from the Grossman-Larson Hopf algebra of
rooted trees to L. Foissys Hopf algebra of planar rooted trees. These four Hopf algebras fit into a
commutative diagram, and several results can be obtained by consideration of the images of objects under
the maps forming this diagram.
JACKSON, R.K., Assistant Professor, Pinto, D.J., and Wayne, C.E., Existence and Stability of Traveling Pulses in
a Continuous Neuronal Network, SIAM J. Appl.Dyn. Syst., 4 (2005), pp. 954-984.
In this work, we examined the existence and stability of traveling pulse solutions of a set of integrodifferential equations that describe activity in a spatially extended population of synaptically connected
neurons. Our results were guided by the local behavior of individual neurons. When neurons have a single
stable state, we demonstrated the existence of two traveling pulse solutions in a connected network. For
bistable neurons, we demonstrated the existence of a stationary pulse solution and, in some cases, a single
traveling pulse solution. Additionally, we performed a linear stability analysis of all these pulse solutions
by constructing an Evans function. Numerical evaluation of these traveling pulses suggests that for the
network of monostable neurons, fast pulses are stable and slow pulses are unstable. For the bistable system,
all pulses are unstable. The system simplifies considerably in the case of stationary pulses and we present
an alternate simpler analysis of their existence and instability.
JOYNER, W.D., Professor and KSIR, A.E. Assistant Professor, Decomposing Representations of Finite Groups on
Riemann-Roch Spaces, Proceedings of the American Mathematical Society, in press.
If G is a finite subgroup of the automorphism group of a projective curve X and D is a divisor on X
stabilized by G, then under the assumption that D is nonspecial, we compute a simplified formula for the
trace of the natural representation of G on the Riemann-Roch space L(D).
JOYNER, W.D. Professor and KSIR, A.E. Assistant Professor, Automorphism Groups of Some AG Codes, IEEE
Transactions on Information Theory, July 2006, in press.
We show that, in many cases the automorphism group of a curve and the permutation automorphism group
of a corresponding AG code are the same. This generalizes a result of Wesemeyer beyond the case of
planar curves.
JOYNER, W.D., Professor and KSIR, A.E. Assistant Professor, Modular Representations on Some Riemann-Roch
Spaces of Modular Curves X(N), Computational Aspects of Algebraic Curves, Lecture Notes Ser. Comput., 13,
World Sci. Publ., Hackensack, NJ, 2005, pp.163-205.
We compute the PSL(2,N)-module structure of the Riemann-Roch space L(D), where D is an invariant
non-special divisor on the modular curve X(N), with N > 5 prime. This depends on a computation of the
ramification module, which we give explicitly. These results hold for characteristic p if X(N) has good
reduction mod p and p does not divide the order of PSL(2,N). We give as examples the cases N=7, 11,
which were also computed using GAP. Applications to AG codes associated to this curve are considered,
and specific examples are computed using GAP and MAGMA.
KONKOWSKI, Deborah A., Professor, (co-author), Mining Metrics for Buried Treasure, Proceedings of
Malcolm@60 Festschrift Conference, General Relativity and Gravitation, to appear. It is currently available on the
Web
:On
Line
Link:
http://www.springerlink.com/(jneqzy45hfdwbz45gzuky355)/app/home/contribution.asp?referrer=parent&backto=iss
ue,9,25;journal,1,391;linkingpublicationresults,1:101151,1 ArXiv Link: http://arxiv.org/abs/gr-qc/0412137
The same but different: That might describe two metrics. On the surface CLASSI may show two metrics
are locally equivalent, but buried beneath one may be a wealth of further structure. This was beautifully
26
described in a paper by M.A.H. MacCallum in 1998. Here I will illustrate the effect with two flat metrics -one describing ordinary Minkowski spacetime and the other describing a three-parameter family of
Gal'tsov-Letelier-Tod spacetimes. I will dig out the beautiful hidden classical singularity structure of the
latter (a structure first noticed by Tod in 1994) and then show how quantum considerations can illuminate
the riches. I will then discuss how quantum structure can help us understand classical singularities and
metric parameters in a variety of exact solutions mined from the Exact Solutions book.
KONKOWSKI, Deborah A., Professor, (co-author), Classical and Quantum Singularity of Levi-Civita Spacetimes
with and without Cosmological Constant, appears in the Web Proceedings of Workshop on Dynamics and
Thermodynamics of Black Holes and Naked Singularities (Milan, Italy).(14 pages): Publication Link:
http://www.mate.polimi.it/bh/ ArXiv Link: http://arxiv.org/abs/gr-qc/0410114
Levi-Civita spacetimes have classical naked singularities. They also have quantum singularities. Quantum
singularities in general relativistic spacetimes are determined by the behavior of quantum test particles. A
static spacetime is said to be quantum mechanically singular if the spatial portion of the wave operator is
not essentially self-adjoint on a C0 domain in L2, a Hilbert space of square integrable functions. Here we
summarize how Weyl's limit point-limit circle criterion can be used to determine whether a wave operator
is essentially self-adjoint and how this test can then be applied to scalar wave packets in Levi-Civita
spacetimes with and without a cosmological constant to help elucidate the physical properties of these
spacetimes.
KSIR, A.E., Assistant Professor, (co-author), FoxTrot Brings Mathematics to the Comics Page, Math Horizons,
November 2005, pp. 18-20.
Fox Trot is a syndicated comic strip appearing in newspapers across the country and around the globe. Its
creator, Bill Amend, was a physics major in college and often includes mathematical references in the
strips. We examine some of these references and conduct an interview with Amend.
LIAKOS, A., Assistant Professor, (co-author), Time Dependent Flow Across a Step with Slip with Friction
Boundary Condition, International Journal for Numerical Methods in Fluids, to appear. (Manuscript number FLD05-0058.R1, July 2005.
This paper studies numerically the slip with friction boundary condition in the time-dependent
incompressible Navier-Stokes equations. Numerical tests on two-and three-dimensional channel flows
across a step using this boundary condition on the bottom wall are performed. The influence of the friction
parameter on the flow field is studied and the results are explained according to the physics of the flow.
Due to the stretching and tilting of vortices, the three-dimensional results differ in many respects from the
two-dimensional ones.
LIAKOS, A., Assistant Professor,
(co-author), A 3-dimensional Simulation of Barrier Properties of
Nanocomposite Films, Journal of Membrane Science, Volume 263, Issues 1-2, 15 October 2005, Pages 47-56.
Monte Carlo simulations were conducted in two and three dimensions to compute the diffusion coefficients
for membranes containing oriented platelets. The effects of platelet aspect ratio, relative separation, loading
and spatial dimension were studied. The model of Aris (1986) was found to accurately predict these effects
in two dimensions. However, in 3D this model over-predicts the effects of the filler by 50% when
compared to simulations. The results have particular value for nanocomposite film currently under
development for use in food packaging applications.
LOCKHART, Jody, Associate Professor, and WARDLAW, William P., Professor, Determinants of Matrices over
the Integers Modulo m, Mathematics Magazine, in press.
In this paper, the number of matrices over the integers modulo m with determinant k is computed. The
problem is first reduced to the case when m and k are both prime powers. This is done by showing that the
number of such matrices is multiplicative in m and that the number with determinant a constant times a
power of a prime p is equal to the number with determinant the power of p if the constant is not divisible by
27
p. Next, a recursive formula giving the number of matrices with determinant 0 over the integers modulo m
where m is a prime power is found. The proof of the formula involves a very careful counting argument.
Finally, the number of matrices over the integers modulo m with determinant k where m and k are both
prime powers is expressed in terms of the number of matrices with determinant 0.
LOCKHART, Jody, Associate Professor, and WARDLAW, William P., Professor, A Modified Hill Cipher,
Cryptologia, in press.
Several modifications of the cipher invented in 1931 by Lester S. Hill are suggested to make the cipher
more secure.
LOCKHART, Jody, Associate Professor, with WARDLAW, William P., Professor, Problem 1727, Mathematics
Magazine, October 2005.
LUCAS, Marc D., Lieutenant Commander, USN, (co-author), Drilling for $4,100 a Day, U.S. Naval Reserve
Reservist Magazine 2005.
This entry provides an actuarial perspective of the present value of a drill or active duty day as a portion of
the remaining commitment to obtain a reserve retirement benefit.
LUCAS, Marc D., Lieutenant Commander, USN, Drilling for $4,200 a Day, The Reserve Officer Association
Officer Magazine 2006.
This entry provides an actuarial perspective of the present value of a drill or active duty day as a portion of
the remaining commitment to obtain a reserve retirement benefit updated for 2006.
MARUSZEWSKI, Richard, Professor, and CAUDLE, Kyle, Lieutenant Commander USNR, Approximating
Integrals Using Probability, MACE Journal, volume 39, #2, pp 143-147, 2005.
Monte Carlo techniques and expectations from probability are used to approximate the value of a definite
integral. The concepts are implemented via algorithms written in Visual Basic.
MARUSZEWSKI, Richard, Professor, "Do Differential Equations Swing?, MACE Journal, in press.
A look at forcing functions and resonance using both analytic techniques and MAPLE approximations. The
ideas are illustrated by the example of the playground swing.
MELLES, Caroline, Associate Professor, (co-author), Classical Poincar Metric Planted off Singularities using a
Chow-Type Theorem and Desingularization Annales de la Facult des Sciences de Toulous, in press.
We construct complete Khler metrics on the nonsingular set of a subvariety X of a compact Khler
manifold. To that end, we develop (i) a constructive method for replacing a sequence of blow-ups along
smooth centers, with a single blow-up along a product of coherent ideals corresponding to the centers and
(ii) an explicit local formula for a Chern form associated to this singular blow-up. Our metrics have a
particularly simple local formula of a sum of the original metric and of the pull back of the classical
Poincar metric on the punctured disc by a size-function SI of a coherent ideal I used to resolve the
singularities of X by a singular blow-up, where (S I ) :=
2
r
j =1
fj
and the fj's are the local generators
of the ideal I. Our proof of (i) makes use of our generalization of Chow's theorem for coherent ideals. We
prove Saper type growth for our metric near the singular set and local boundedness of the gradient of a
local generating function for our metric, motivated by results of Donnelly-Fefferman, Ohsawa, and
Gromov on vanishing of certain L2-cohomology groups. In an appendix we give a simple constructive
proof of a valuation criterion due to M. Lejeune and B. Teissier.
28
MICHAEL, T. S., Associate Professor (co-author), Packing Boxes with Bricks, Mathematics Magazine, 79
(2006), pp.14-30.
We provide an expository account of arithmetic and geometric conditions that are necessary and sufficient
for a d-dimensional rectangular box to be tiled (packed) by translates of two given rectangular bricks. Our
work unifies a number of results in the tiling literature.
MICHAEL, T. S., Associate Professor (co-author), Optimal Strategies for Node Selection Games: Skew Matrices
and Symmetric Games, Linear Algebra and Its Applications, 412 (2006), pp.77-92.
Each of two players simultaneously selects a node in a directed graph with the payoff determined by the
direction of any arc between the selected nodes. We investigate the optimal strategies for the two players in
this node selection game. We show that when the graph is a non-trivial bipartite graph, the optimal strategy
is never unique.
MICHAEL, T. S., Associate Professor, Rysers Embedding Problem for Hadamard Matrices, Journal of
Combinatorial Designs, 14 (2006), pp.41-51.
We obtain better bounds for the minimum order of a Hadamard matrix that contains a rectangular all 1s
submatrix of a given size.
MICHAEL, T. S., Associate Professor (co-author), Guarding the Guards in Art Galleries, Math Horizons, 14
(2006), 22-23, 25.
We give an expository account of some recent research in computational geometry. Let P be a polygon (the
art gallery) with n sides. We wish to select a set G of points (the guards) so that (a) every point in the
polygon is visible to some guard; and (b) every guard is visible to at least one other guard. What is the
minimum number of guards g(n) that is guaranteed to suffice for all polygons with n sides? We prove that
g(n) = floor[ (3n-1)/7] for n at least 5. Also, if we restrict to rectilinear galleries the function is floor(n/3)
for n at least 6.
POPOVICI, Irina, Assistant Professor, and WITHERS, W. Douglas, Professor, Custom-Built Moments for Edge
Location, IEEE Transactions on Pattern Analysis and Machine Intelligence, 28 (2006), no. 4, pp.637-642.
We present a general construction of functions whose moments serve to locate and parameterize step edges
within an image. A characteristic of the moment approach to edge location is that the result is specified
parametrically rather than as a set of pixels; thus our method performs best on sharp, straight edges.
Previous use of moments to locate edges was limited to functions supported on a circular region, but our
method allows the use of ``custom-designed'' functions supported on circles, rectangles, or any desired
shape, and with graphs whose shape may be chosen with great freedom. We present analyses of the
sensitivity of our method to pixelization errors or discrepancy between the image and an idealized edge
model. The parametric edge description provided by our method makes it especially suitable as a
component of wedgelet image coding.
POPOVICI, Irina, Assistant Professor, and WITHERS, W. Douglas, Professor, Locating Thin Lines and Roof
Edges by Custom-Built Moments, Proceedings of the 2006 International Conference on Image Processing, in
press.
The method of custom-built moments allows location of step edges in an image in parametric form, as an
equation ax+by=c, using a customizable mask of almost any desired shape, or moment functions adapted
to calculation in a particular basis. We adapt this method to the problem of locating both thin lines and roof
edges in parametric form.
WARDLAW, William P., Professor, Row Rank Equals Column Rank, Mathematics Magazine, Vol. 78, No. 4,
(October 2005) 316-318.
This note gives a short (perhaps the shortest) proof that row rank is equal to column rank and continues to
derive some properties of matrix rank.
29
WARDLAW, William P., Professor, problem 811, College Mathematics Journal, Vol.36, No. 5, (November 2005),
413.
This problem is to find the characteristic polynomial of the classical adjoint (adjugate) of a matrix, given
the characteristic polynomial of the matrix.
ZARIKIAN, Vrej A., Assistant Professor (co-author), The Calculus of One-Sided M-Ideals and Multipliers in
Operator Spaces, Memoirs of the American Mathematical Society, Vol. 179, No. 842, viii+85 pp., January 2006.
The theory of one-sided M-ideals and multipliers of operator spaces is simultaneously a generalization of
classical M-ideals, ideals in operator algebras, and aspects of the theory of Hilbert C*-modules and their
maps. Here we give a systematic exposition of this theory. The main part of this memoir consists of a
calculus for one-sided M-ideals and multipliers, i.e. a collection of the properties of one-sided M-ideals
and multipliers with respect to the basic constructions met in functional analysis. This is intended to be a
reference tool for noncommutative functional analysts who may encounter a one-sided M-ideal or
multiplier in their work.
Technical Reports
BUCHANAN, J.L., Professor, An Assessment of the Biot-Stoll Model of a Poroelastic Seabed, Naval Research
Laboratory Memorandum Report NRL/MR/7140-05-8885, 83 pages, August 5, 2005, Washington DC.
Discussed in this report are:
The derivation of the equations of the Biot-Stoll model, including later extensions of Biots original
formulation.
The determination of the Biot-Stoll parameters.
The predictions of the Biot-Stoll model for wave speed, attenuation, reflection and transmission.
Difficulties with and controversies about the Biot-Stoll model.
The incorporation of frequency-dependent viscoelastic effects into the model.
GENTRY, Sommer E. Assistant Professor (co-author), Consensus Statement of the Mathematics Subgroup of the
United Network for Organ Sharing Kidney Paired Donation Working Group, 2 pages, April 2006.
This is an extremely short consensus statement on the mathematics of optimizing kidney paired donation
addressed from the primary academic researchers in the field to the federally-funded agency which is
planning to create and run a national kidney paired donation registry in the United States. The length of the
statement is inversely proportional to the difficulty we experienced in coming to the consensus that it
presents.
MINUT, Aurelia, Assistant Professor, Modeling of MEMS Gyroscopes, Institute for Scientific Research,
Technical Report.
The MEMS (microelectromechanical systems) gyroscopes can be used in navigation systems in cases
where the GPS is denied. The MEMS gyroscopes are small devices that can be mounted on a wide range of
platforms, such as UAVs, guided missiles.
We developed a model for the equations of motion and solved the equations numerically. Then we tried
coupling two, four, nine and twelve gyroscopes in different ways and compared the results. This research
attempts to improve the performance of current gyroscopes by weakly coupling an array of MEMS
gyroscopes.
30
Presentations at Professional Meetings and Conferences

ALEVRAS, Alexis A., Assistant Professor, The Gauge Group of an E0-semigroup, Third East Coast Operator
Algebras Symposium, University of Pennsylvania, 2 October 2006. (invited talk)
GAGLIONE, Anthony M. Professor, An Embedding Theorem for Groups Universally Equivalent to Free Nilpotent
Groups, Groups St. Andrews, St. Andrews, Scotland, 30 July - 6 August 2005.
GAGLIONE, Anthony M. Professor, Notions of Discrimination, Special Session of AMS, Bard College, New
York, 7-9 October 2005.
GAGLIONE, Anthony M. Professor, On a Question of Verena Huber Dyson, Conference in Honor of Olshanski,
Vanderbilt University, 6-10 May 2006.
GENTRY, Sommer E., Assistant Professor, Maximum Matching for Kidney Paired Donation, Computational
Research in Boston seminar, Massachusetts Institute of Technology, 2 September 2005.
GENTRY, Sommer E. Assistant Professor, Optimizing Kidney Paired Donation, Economics Department seminar,
U.S. Naval Academy, 13 September 13, 2005.
GENTRY, Sommer E., Assistant Professor, Mathematical Optimization for Transplant Professionals, University
of Cincinnati Institute for the Study of Health, 26 September 2005. (invited talk)
GENTRY, Sommer E., Assistant Professor, Matching Inequalities in Kidney Paired Donation, Institute for
Operations Research and Management Science (INFORMS) Conference on Optimization and Healthcare, San
Antonio, TX, 4 February 2006.
GENTRY, Sommer E., Assistant Professor, Maximum Matchings on Graphs for Kidney Paired Donation, The
Johns Hopkins University Applied Mathematics and Statistics department seminar, 6 April 2006.
GENTRY, Sommer E., Assistant Professor, Neat Teaching Ideas: Index Card Essays. MAA Section NeXT
meeting, 8 April 2006.
GENTRY, Sommer E., Assistant Professor. Graph Theory and Abstractions, extended interview taped 5 May
2006, as part of the Discovery Productions series entitled Discovering Math. Discovering Math is an educational
video series being developed for use in high school mathematics classrooms.
HOFFMAN, Michael E., Professor, Updown Categories and Algebraic Structures, AMS Special Section on
Algebraic and Enumerative Combinatorics, Joint Mathematics Meetings, San Antonio, TX, 13 January 2006.
HOFFMAN, Michael E., Professor, Quasi-symmetric Functions and Applications, Combinatorics seminar,
George Washington University, Washington, D.C., 15 February 2006.
HOFFMAN, Michael E., Professor, Some Hopf Algebras of Physical Interest, Renormalization and Galois
Theories, Centre International de Rencontres Mathmatiques, Marseille, France, 13 March 2006. (invited talk)
HOFFMAN, Michael E., Professor Some Generating Functions for Permutations, U.S. Naval Academy Pure
Mathematics Colloquium, 29 March 2006.
HOFFMAN, Michael E., Professor The Algebra of Multiple Zeta Values and Euler Sums, Mini-Conference on
Zeta Functions, Index and Twisted K-Theory: Interactions with Physics, Mathematisches Forschungsinstitut,
Oberwolfach, Germany, 2 May 2006.
31
HOFFMAN, Michael E., Professor, Rooted Trees and Symmetric Functions, Mini-Conference on Zeta Functions,
Index and Twisted K-Theory: Interactions with Physics, Mathematisches Forschungsinstitut, Oberwolfach,
Germany, 4 May 2006.
KONKOWSKI, Deborah A., Professor, Quantum Singularities in Generalized Spacetimes, 13th Meeting of the
European Physical Society, Bern, Switzerland, 11-15 July 2005.
KONKOWSKI, Deborah A., Professor, Quantum Singularities in General Relativistic Spacetimes, Albert Einstein
International Centenary Conference, 17-23 July 2005.
KONKOWSKI, Deborah A., Professor, Black Holes, Big Bangs and Cosmic Strings, World Year of Physics
lecture, Hood College, Fredrick, MD, 3 November 2005.
KONKOWSKI, Deborah A., Professor, Einstein and Quantum Mechanics, World Year of Physics lecture,
University of Maryland, Baltimore County, MD, 1 March 2006.
KONKOWSKI, Deborah A., Professor, Quantum Mechanical Healing of Classical Singularities, American
Physical Society April Meeting, Dallas, TX, 22-25 April 2006.
KSIR, Amy E., Assistant Professor, Finite Group Actions on Riemann Roch Spaces and Automorphisms of
Algebraic Geometry Codes. American Mathematical Society Summer Research Institute for Algebraic Geometry,
University of Washington, Seattle, WA, 28 July 2005.
KSIR, Amy E., Assistant Professor. Error-Correcting Codes and Algebraic Curves, Gettysburg College
Mathematics Department Colloquium, Gettysburg, PA, 17 November 2004.
KSIR, Amy E., Assistant Professor, Group Representations on Some Riemann-Roch Spaces of Hurwitz Curves,
AMS Eastern Section meeting, 22 April 2006.
LIAKOS, A., Assistant Professor, Slip with Friction Boundary Condition for the Navier-Stokes Equations
Numerical Studies for Time Dependent Laminar Flows, 27 October 2005. (Sigma Xi invited talk)
LOCKHART, Jody, Associate Professor, Determinants of Matrices over the Integers Modulo m, MD-VA-DC
Section Spring MAA meeting, Loyola College, Baltimore, MD, 8 April 2006.
LUCAS, Marc D., Lieutenant Commander, USN, Midshipman to Millionaire - Practical Applications of
Mathematics to Personal Financial Security. Math Open House U.S. Naval Academy, Annapolis, MD, 21 February
2006.
LUCAS, Marc D., Lieutenant Commander, USN, An Overview of Actuary Science, SM280, Topics in
Mathematics, U.S. Naval Academy Annapolis, MD, January 2006.
MCCOY, Peter A., Professor, On the Propagation of Electromagnetic Waves, CP 22 Part III, SIAM Annual
Meeting, New Orleans, LA, 15 July 2005.
MCCOY, Peter A., Professor, An Overview of Shannons Sampling Theorem, U.S. Naval Academy Applied
Math Seminar (Chesapeake Bay Group), Annapolis, MD, 2 November 2005.
MCCOY, Peter A., Professor, On the Normal Mode Expansion of Solutions to the Paraxial Wave Equation,
Session on PDEs I, AMS-MAA annual joint meetings, San Antonio, TX, 13 January 2006.
MICHAEL, T. S., Associate Professor, Art Gallery Theorems, University of Delaware, Combinatorics Seminar,
Newark, DE, October 2005.
MICHAEL, T. S., Associate Professor, Bridge-Ratings, Wrestling Tournaments, and Combinatorics George
Washington University, Combinatorics Seminar, Washington, D.C., April 2006.
32
PENN, Howard L., Professor Which Ballparks are Homer Friendly?, Mathematical Association of America
Mathfest, Albuquerque, NM, 12 August 2005.
PENN, Howard L., Professor Which Ballparks are Homer Friendly, Part II? Mathematical Association of America
Sectional Meeting, Baltimore, MD, 8 April 2006.
POPOVICI, Irina, Assistant Professor, A Two-dimensional Dynamical System Underlying Cardiac Arrhythmias,
AMS Special Session on Dynamic Equations with Applications, 2005 Joint Meeting of the Mathematical Societies,
San Antonio, TX, January 2006.
POPOVICI, Irina, Assistant Professor, A Dynamical Systems Approach To Membrane Phenomena Underlying
Cardiac Arrhythmias, International Conference on Chaos and Nonlinear Dynamics, Bethesda MD, January 2006.
POPOVICI, Irina, Assistant Professor, A Two-dimensional Dynamical System Underlying Cardiac Arrhythmias,
MAA MD/DC/VA Regional Fall Meeting, Montgomery College, MD, November 2005.
SANDERS, Thomas J., Professor, Core Mathematics at the United States Naval Academy, Academic External
Review Group Meeting, Institute of Defense Analysis, Arlington, VA, 4 October 2005.
TURNER, John, Professor, The Flat Earth Guide to Probability and Statistics, Joint Statistical Meetings,
Minneapolis, MN, 10 August 2005.
WARDLAW, William P., Professor, A Modified Hill Cipher", U.S. Naval Academy Colloquium Talk, 8 February
2006.
WARDLAW, William P., Professor, Row Rank Equals Column Rank, Spring meeting of MD-VA-DC Section of
Mathematical Association of America(MAA), Loyola College, Baltimore, MD, 8 April 2006.
ZARIKIAN, Vrej A., Assistant Professor, Paving Small Matrices, Special Session on Recent Progress in Operator
Algebras, American Mathematical Society (AMS) Central Section Meeting, Lincoln, NE, 22-23 October 2005.
ZARIKIAN, Vrej A., Assistant Professor, Alternating-Projection Algorithms for Operator-Theoretic Calculations,
Special Session on Geometry of Banach Spaces and Connections with Other Areas, American Mathematical Society
(AMS) Southeastern Section Meeting, Miami, FL, 1-2 April 2006.
33

Mathematics Department

Uploaded by

Copyright:

Available Formats

Mathematics Department

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mathematics Department

Uploaded by

Copyright:

Available Formats

______________________________________________________________________________

Geo-Spacial Data Analysis

On the Visualization of Large Point Cloud Data

Microaltimeter Flight Data Post Processing

Combinatorics of Updown Categories

Automorphisms of Curves and Codes

Twin Screw Extrusion of Energetic Materials

Modeling and Analysis of Laser Propagation IV

Modeling of MEMS Gyroscopes

United States Naval Academy Chesapeake Bay Research

A Multi-Scale Edge Transform for Image Coding

Cocycles for One-parameter Flows of B(H)

Marginal Tracial States on Tensor Products of Matrix Algebras

B1 B2 is a state that satisfies ( x I ) = TrB1 ( x) and

for all x, y in B. In 2003 K. R. Parthasarathy showed that if B is the algebra of

Using Off-the-shelf Software in the Classroom

Non-Parametric Density Estimation of Streaming Data Using Orthogonal Series

Designing Digital Tools for Communicating Mathematics in Higher Education

Nontrivially Discriminating Groups

Maximum Matching Inequalities in Kidney Paired Donation

Designing and Evaluating Weighting Functions for Matching on Graphs

Hopf Algebras of Trees and Symmetric Functions

Combinatorics of Updown Categories

A Stability Criterion for Waves in a Coupled Optical System

Bifurcation and Stability of Standing Waves in a Gross-Pitaevskii Equation

A Numerical Package for Computation of the Evans Function

Pulses in Nonlinearly Coupled Schrodinger Equations: I. A Homoclinic Flip Bifurcation

Pulses in Nonlinearly Coupled Schrodinger Equations: II. An Instability Criterion

Bases for Riemann-Roch Spaces of Hyperelliptic Curves

Group Actions on Riemann-Roch Spaces of Some Hurwitz Curves

Classification of Singularities with SHEEP/CLASSI and/or MAPLE

Classical and Quantum Singularities in General Relativistic Spacetimes

Expository Notes on Quantum Field Theory

A Posteriori Error Estimators for a Two-Level Finite Element Discretization

Modeling Gas Diffusion in Nanocomposites

No-Slip Boundary Conditions for the Navier-Stokes Equations by Penalty-penalty Method

Atiyah-Singer Index Theorem

Seasonality Methods with Statistical Measure

Controls for Complete Controllability of Linear, Autonomous Systems of ODE

Classical Poincar Metric Planted off Singularities

Investigations into Sudoku

Maple Animation of Space Filling Curves

Viewing Mbius Transformations with Maple Animations

Wrestling Tournaments: How to Select First-Round Byes

Tiling with Three Squares

Nonlinear Interaction of Light with Liquid Crystals

Competition in Two-Dimensional Heterogeneous Environments

On a New Notion of Ellipticity

Mathematical Foundations for Statistical Learning

A Dynamical Systems Approach to Cardiac Arrhythmias

Locating Edges and Removing Ringing Artifacts in JPEG Images

A Dynamical Systems Approach to Membrane Phenomena

Stochastic Generalized Assignment Problem

Differential Operators on Orbifolds

The Fundamental Theorem of Invariant Theory and Differential Operators

Enumerative Algebraic Geometry: Counting Conics

Quadratic Forms and Cohomology

Menu Program for using Stats/List Editor on Voyage 200

Matrices over Commutative Rings