Mathematics, Computer Science and Statistics.

Chair: Walter Brehm, Keesler Air Force Base 

Vicechair: Ravinder Kumar, Alcorn State University 

THURSDAY MORNING

Conference Room

8:50 Introduction

9:00 A RAPIDLY RELOCATABLE PREDICTION SYSTEM

Germana Peggion * and Daniel N. Fox, University of Southern Mississippi, Stennis Space Center, MS 39529, and Naval Research Laboratory, Stennis Space Center, MS 39529

MODAS-NRLPOM is a scalable, portable, and rapidly relocatable system for nowcasting and short-term (2-day) forecasting in support of real-time naval operations. The Modular Ocean Data Assimilation System (MODAS) combines remote sensed data (altimetry and sea surface temperature) with in situ measurements to produce an analysis of the ocean that can be considerably more accurate than conventional climatology. Data from several datasets and internet sites are gathered at NRL in an almost fully-automated fashion. The MODAS nowcast field provides initial and boundary condition for NRLPOM, a version of the Princeton Ocean Model (POM), implemented at the Naval Research Laboratory (NRL). The principal attributes of NRLPOM are a user-friendly interface, the inclusion of tidal flow, and options for several initialization procedures and boundary condition algorithms. The system also has the capability of one-way coupling with other real-time operational models or 1-way nesting (NRLPOM to NRLPOM). The real-time simulati ons are forced by the operational winds available for a given area. The system has been designed and implemented so that no data are ingested and assimilated during the forecasting simulations. Forecast predictions are usually available within 6 hours of the initial MODAS nowcast.

9:20 QUANTUM CHEMICAL INVESTIGATIONS OF EXCITED STATE PROPERTIES OF HYPOXANTHINE 

M.K. Shukla * and Jerzy Leszczynski, Jackson State University, Jackson, MS 39217

Hypoxanthine is a purine metabolic intermediate in living systems. Structurally it is close to guanine and can be formed by the deamination of guanine. It is also found as a minor purine base in transfer RNA. It shows keto-enol and N9H-N7H tautomerism. However, under aqueous environment the keto-N9H form dominates. The CASSCF, TDDFT and CIS levels of theories were used to study electronically excited singlet state properties of hypoxanthine. The ground state geometries were optimized at the MP2, B3LYP, and HF levels using 6-311++G(d,p) basis set. Excited state geometries were optimized at the CIS/6-311++G(d,p) level. The nature of potential energy surface was ascertained using the harmonic vibrational frequency analysis. The accuracy of different methods in explaining the experimental transition energies and change in features of molecular electrostatic potential maps in going from the ground state to different excited states are discussed.

9:40 UNINTERRUPTED DATA PROCESSING ACROSS NETWORKED PLATFORMS UTILIZING AUTOMATIC AUTHENTICATION OF ADVANCE GENERATED KERBEROS TICKETS

Deborah S. Franklin, Planning Systems Incorporated at the Naval Oceanographic Office, Stennis Space Center, MS 39529

Many sites incorporate the Massachusetts Institute of Technology (MIT) Kerberos network authentication protocol verification software to validate identity across unsecured network connections. While the software provides necessary security using encrypted ticket information, it requires intervention of manual ticket generation if the tickets are not renewed within a specific timeframe set by the users authentication Key Distribution Center (KDC). Therefore maintenance or network interrupts between the primary processing system and the KDC can cause remote processing to fail until manual intervention occurs to replace the obsolete ticket. However, a wrapper code around specific options within the basic Kerberos commands can create tickets with future timestamps for utilization during off hours creating seamless resumption of processing without human intervention. For our group, advance Kerberos ticket generation increased efficiency of operational processing, continuing a supply of timely products to customers immediately after lengthy downtime without any further delays. With renewed focus regarding security issues at computer sites regardless of orientation, implementation of procedures to augment autonomy without compromising security is an increasing concern for all individuals within the computing industry. This procedure works within the rules of the security while promoting automation.

10:00 Break

10:20 STUDY OF TWO DIFFERENT APPROACHES TO PARALLELIZING A SIMPLE NUMERICAL CODE

Moinuddin K. Shalam (1) *, Chandra Narayanan (2), and Matthew Bettencourt (2), (1) Mississippi State University, Stennis Space Center, MS 39529, and Center for Higher Learning (2), Stennis Space Center, MS 39529

Parallel computing has made a tremendous impact on many areas of computer application, especially in the areas of scientific computing like weather prediction, ocean modeling etc. Different approaches for parallelizing numerical models are available. We look at two of the available choices using a simple 2-D advection code as a test case. The first approach is to use MPI, a message-passing library specification widely used on distributed memory parallel machines. The other is to use the parallel computing framework provided by the Weather Research and Forecasting (WRF) model. The two approaches are compared for performance metrics like speedup and efficiency, and for characteristics like scalability and portability across different computer architectures. One limitation of this study is the use of a small code as a test case. Useful inferences can be derived when many test cases of varying sizes and complexities are analyzed.

10:40 VISUAL ANALYSIS OF PHASE SEPARATION AND FLOW IN A MULTI-COMPONENT DRIVEN SYSTEM

Ray Seyfarth (1) *, R.B. Pandey (1,2), and Joe F. Gettrust (2), (1) University of Southern Mississippi, Hattiesburg, MS 39406, and (2) Naval Research Laboratory, Stennis Space Center, MS 39529

Computer simulations are performed to study flow and phase separation in an interacting lattice gas of a multicomponent system. We consider a three dimensional lattice with as much as four components: two fluid constituents, sediments, and pores characterizing the effective host medium. Bottom of the lattice is connected to a source of fluid mixture consisting of constituents A and B with different masses; the top end is open. Nearest neighbor interactions among the particles are considered. Driving fields are concentration gradient, gravity, hydrostatic pressure, etc. Using an OpenGL visualization package, animp, we analyze the motion of individual particles, their collective (center of mass) motion, density profiles, phase separation etc.

11:00 AUTOMATED IMAGE REGISTRATION BY MATCHING EDGE-ENHANCED IMAGES

Hong Zhou * and Ray Seyfarth, University of Southern Mississippi, Hattiesburg, MS 39406

We present a method for performing automatic geographical registration of aircraft-acquired images based on matching the aircraft images with previously georeferenced images from the USGS (DOQQs). The process uses edge-enhancement based on areal standard deviation values to produce images which are less sensitive to temporal variation. The base data is repeatedly mapped to the warp data with differing mapping coefficients. For each such mapping a measure of fit is computed (dot product) and the best mapping coefficients are kept. This process continues until the fit is good enough and then the warp image is mapped to the base and processing continues to a new file. We consider affine and quadratic mappings and a variety of optimization strategies.

11:20 Break

11:40 VISUAL ANALYSIS OF GROWTH AND ROUGHNESS TN ELECTROPHORETIC DEPOSITION OF POLYMER CHAINS

R.B. Pandey * and Ray Seyfarth, University of Southern Mississippi, Hattiesburg, MS 39406

Monte Carlo method is used to simulate the interface growth and roughness in an electrophoretic deposition of polymer chains. We consider coarse-grained model of polymer chains on a discrete lattice. Polymer chains are driven by an electrophoretic field toward an impenetrable substrate. Short range interactions among polymers (repulsive) and between polymer and substrate are considered. Using an OpenGL visualization tool (animp), animations are prepared to show the dynamics of individual chains, and their collective densities and resulting interface. Visualization of a tracer chain, is presented to show the dynamics and conformation of chains from interface to bulk.

12:00 PICKING GEOREFERENCE CONTROL POINTS INTERACTIVELY USING IMAGE BLENDING

Arunkumar Rajendran * and Ray Seyfarth, University of Southern Mississippi, Hattiesburg, MS 39406

We present a method for picking control points for georeferencing using image blending to allow the image analyst to match areas visually. The typical control point picking process involves having an image analyst select a matching pixel from the base and warp images. We contend that this is process does not adequately utilize the immense visual processing capabilities of the human picking the point. Our process involves interactively translating, rotating and scaling one image against the other while presenting the analyst with a combination of the two images. In this fashion it is possible to match areas rather than individual pixels and it is possible to obtain fractional pixel coordinates. A variety of blending and user interface techniques are considered in an effort to make this process more effective and intuitive.

THURSDAY AFTERNOON

Conference Room

2:00 MISSISSIPPI CENTER FOR SUPERCOMPUTING RESEARCH (MCSR) USER ADVISORY GROUP MEETING

David G. Roach * and Germana Peggion, University of Mississippi, University, MS 38677 and University of Southern Mississippi, Stennis Space Center, MS 39522

The Mississippi Center for Supercomputing Research was established in 1987 by the Mississippi Legislature and the Institutions of Higher Learning (IHL) in order to provide high performance supercomputing (HPC) support for research and instruction at all state universities. The Mississippi Supercomputer User Advisory Committee (MSUAG) was established by the IHL Research Consortium to provide user input and advice to MCSR management and technical staff on policies and procedures for the Center's operations. It includes member representatives from all IHL institutions. The Advisory Group will meet at this MAS conference. Mr. David G. Roach, Director of the MCSR, and Dr. Germana Peggion, MSUAG Chair and Professor at USM Stennis, will conduct the meeting. The agenda includes an update on MCSR HPC facilities and services, introduction of new MCSR staff members, and site reports and ongoing research updates by MSUAG representatives. A Special Subsession of the Mathematics, Computer Science and Statistics Division, s ponsored by the MCSR, will also be held to serve as a forum on supercomputing in which faculty and graduate student researchers will have the opportunity to describe their research projects that involve HPC, Internet2, Grid Computing, Visualization, Network Security, Computer Systems Administration, and the use of MCSR resources. IHL faculty and graduate students, with an interest in HPC and/or MCSR facilities and services, are also invited to attend.

2:30 Divisional Poster Session

THE MOLECULAR STRUCTURES AND NATURE OF INTERACTIONS IN [N.sub.2][H.sup.+]-He, (n = 1 - 12) COMPLEXES

Yinghong Sheing (1) *, Szczepan Roszak (2), and Jerzy Leszczynski (1), (1) Jackson State University, Jackson, MS 39217, and (2) Wroclaw University of Technology, Wyb, Wyspianskiego 27, 50-370 Wroclaw, Poland

The [N.sub.2][H.sup.+]-H[e.sub.n] complexes were investigated for n up to 12. The molecular structures were optimized without any structural constraint using the second-order Moller-Plesset (MP2) perturbation level of theory. The harmonic vibrational frequencies were obtained at the same level of theory. The [N.sub.2][H.sup.+]-He dimer has a linear proton-bound structure, and further He ligands fill two equatorial solvation rings around the linear dimer core, each of them containing up to five He ligands. The first solvation shell is completed when the 12th helium atom is attached at the nitrogen end of [N.sub.2][H.sup.+]. The variation of the calculated dissociation energies, and vibrational frequencies is related to the increasing size of the studied complexes. The nature of interactions is discussed by analyzing the charge distribution and the energy decomposition scheme.

FRIDAY MORNING

Conference Room

8:20 Introduction

8:30 A PATTERN RECOGNITION SOFTWARE TOOL DEVELOPED USING MATLAB FOR THE CLASSIFICATION OF REMOTE SENSING SPECTRAL REFLECTANCE DATA OF STRESSED SOYBEAN LEAF

Abdullah Faruque *, Raj Bahadur, and Gregory A. Carter, Southern Polytechnic State University, Marietta, GA 30060; Mississippi Valley State University, Itta Bena, MS 38941; and Earth System Science Office, NASA, Stennis Space Center, MS 39529

This paper describes the implementation of LIP (Leaf Identification Program), a pattern recognition software tool intended to classify remote sensing spectral reflectance data of stressed soybean leaves by using neural network and other statistical pattern recognition techniques. Various data preprocessing techniques necessary to support the pattern recognition techniques are also provided. Data visualization tools are also provided to permit visual assessment of the spectral reflectance data patterns and their relationships. The development of this software system takes advantage of the high performance computational and visualization routines of the MATLAB programming environment. Data analysis component of LIP includes: principal component analysis, fisher and variance weight calculations and feature selection. Using MATLAB's graphics routines, 2-dimensional or 3-dimensional plots of the principal components can be displayed. Classification methods in LIP include both neural network and statistical pattern recognition techniques. Neural network methods include the back propagation neural network (BPN) and radial basis function (RBF) neural network. Statistical pattern recognition component of LIP includes linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), regularized discriminant analysis (RDA), soft independent modeling of class analogy (SIMCA) and discriminant analysis with shrunken covariance (DASCO). The objective of this study funded by National Aeronautics Space Administration (NASA) at Stennis Space Center was to record and classify the spectral reflectance differences of leaf stress caused by drought, fungal disease, and lead contamination of the soil. LIP software tool has been used successfully to classify the different classes of stressed leaves from their spectral signature.

8:50 INTEGRATING BORLAND C++ WITH MICROSOFT ACCESS

April Butler * and Joan Palmer, Tougaloo College, Tougaloo, MS 39174, and National Oceanic and Atmospheric Administration, Woods Hole, MA 02543

The purpose of my research was to open a printable Microsoft Access database report using Borland C++ Builder 5. Borland C++ Builder 5 is one of the most powerful and complete ANSI C++ integrated development environment that allows the user to develop, deploy, integrate, and access various software applications. With Borland, I was able to access Microsoft Office programs and interfaces, document and document formats, use task automation, and embed office applications into the C++ Builder form. Task automation allowed me to prepare specific operations to carry out, choose certain files, and execute the operation. In order to successfully complete my project, I had to access Borland newsgroups, and translate Visual Basic syntax into C++ in order to open a specific file in the database report. With the proper code, I was able to open Microsoft Access, the correct database, and report by clicking the button on the C++ Builder form. In addition, this research was added to the Database Management Systems departmen t library of tools software applications at the National Marine Fisheries Service.

9:10 EFFECT OF RANDOM VARIATION IN THE PARAMETER OF A DISCRETE LIFE TIME DISTRIBUTION ON VARIOUS RELIABILITY CHARACTERISICS

Harm Sharma * and K.K. Sharma, East Mississippi Community College, Mayhew, MS 39753, and C.C.S. University, Meerut, India

Prior distributions represent random variations in the parameters of lifetime distributions. This prior knowledge is updated by using experimental data in the Bayesian framework. However, updating the basic lifetime distribution in respect of these parametric variations is another important aspect to be considered in reliability analysis. The present study deals with the analysis of the reliability characteristics of a discrete lifetime distribution updated in respect of variations on its parameter. Reference of a k-out-of-in system is included as an example.

9:30 Break

9:45 THE AFRICAN AMERICAN WOMAN AND MATHEMATICS: A CRITICAL ANALYSIS

LaTrese Davis, Alcorn State University, Alcorn State, MS 39069

Why are minorities underrepresented in Professional Mathematics? Several researchers like Dr. Scott Williams, a professor of Mathematics at University of New York-Buffalo, Dr. William H. Tucker, an associate professor of psychology at Rutgers University, Dr. Susan Chipman, of the National Institute of Education, co-authored Women's Participation in Mathematics, and others have questioned the low involvement of minorities, especially African-American women. This research examines the African American female demographics of the Mathematics profession.

10:05 AN INTERACTIVE DERIVATIVE PRIMER WITH ONLINE CAPABILITIES

Ravinder Kumar, Alcorn State University, Alcorn State, MS 39069

Math Wright is a powerful mathematics authoring software. It creates microworlds that can be easily put on the web and downloaded from there or can be viewed in the browser with all dynamism and interactivity intact. We have developed a primer for derivatives that explains the theory as well as skills. The user can customize problems and examples or use thousands of randomly generated examples and exercises. Historical notes attempt to put the development of the concept in the right perspective. The primer is class-tested and is on the prestigious mathwright library. It provides an effective teaching/learning tool. A CD containing the primer with a document on reading it will be distributed free. If you are interested in viewing online, please go to either of the following URL's. www.math.metrostate.edu/welcome http://canufly.net/~kumar

10:25 GEOCODING: QUANTITATIVE RESEARCH ON THE MATCHING RATE AND ACCURACY OF STREET ADDRESS MATCHING STRATEGY

Lixin Yu, Alcorn State University, Alcorn State, MS 39069

Street address geocoding is the process of converting text version of street addresses into geographic coordinates. The process requires matching each street address with street addresses stored in an authority file that contains the geographic coordinates. One important issue in this process is how to deal with the imperfect data--various errors exist in both the address file to be processed and the authority file. Therefore, an approximate matching is required to raise the matching rate. In practice, a weighted matching process is widely used to compare the street name, prefix, street type, and zipcode separately. The system decides whether a match happens based on the weighted matching result. This research studies the impact of different weight settings to the matching rate and accuracy of the result. It uses third party data as an authority file to determine the correctness of the matching. Three hundred street addresses were geocoded 625 times to test the result of possible strategies. A program was mad e to analyze the result and generate the summary. The study further analyzed some interesting patterns of the data and suggested the best strategy to get the maximum matching rate or the maximum accuracy in the geocoding process.

10:45 STATISTICAL DESIGN FOR TWO-CHANNEL MICROARRAY EXPERIMENTS

Carolyn Boyle, Mississippi State University, Starkville, MS 39762

Microarray experiments are widely used to study gene expression profiles. For example, the pattern of genes expressed in tumors that respond to chemotherapy could be compared to the pattern in those that do not respond. A microarray is a grid of "spots" attached to a glass slide. Each spot consists of a known DNA sequence from a gene library and each slide can contain many thousands of sequences. Messenger RNA (mRNA) from each experimental sample is reverse-transcribed into a complementary DNA (cDNA) sample and labeled with a fluorescent dye. With a two-channel system, one cDNA sample is mixed with "red" dye and another with "green" dye. The two labeled cDNA samples are then combined, placed on the array, and allowed to hybridize to the spots. Finally, the array is scanned with a laser and fluorescence intensity measurements are made for each dye at each spot. The ratio of the intensities is related to the quantity of mRNA in the experimental samples. The main statistical design problems are determining which samples should be hybridized together on the same slide and the amount of replication. Those issues involve the research objectives, sources of variation between and within slides, the amount of mRNA available, and the cost of the procedure. This presentation will give examples of some designs that are appropriate for the most common research objectives.

FRIDAY AFTERNOON

Conference Room

1:40 CUBIC SPLINE AND GRAPHICS

Mikhail Korablin * and Oleg Bestseny, Alcorn State University, Alcorn State, MS 39069

The method of cubic splines is a very useful and efficient method of interpolating numerical data. In this paper, we have developed a Visual C++ program that will demonstrate how graphics can be obtained by using method of cubic splines. In the course of work we have developed the algorithms for solving the matrix set up for third degree polynomials representing the various splines.

2:00 APPLICATION OF MATH IN ENGINEERING--A CASE STUDY

David Loflin * and S. Kant Vajpayee, University of Southern Mississippi, Hattiesburg, MS 39406

Engineering and engineering technology programs heavily rely on the fundamental principles of mathematics. At the University of Southern Mississippi, those pursuing BS degrees in engineering technology take 15 semester hours of mathematics, comprising college algebra (MAT 101), trigonometry (MAT 103), applied calculus (MAT 136 & 137), and statistical methods (CSS 211). College algebra covers topics such as linear and quadratic equations, functions, and graphing. Trigonometry covers trigonometric functions, identities and equations, and common geometric shapes. Applied Calculus I (MAT 136) covers differentiation and integration techniques and their applications. Applied Calculus II (MAT 137) continues with advanced differentiation and integration techniques with applications to areas, centroids, and moments of inertia. CSS 211 covers measures of central tendency and dispersion, sampling and their distributions, and hypothesis testing. Several engineering technology courses at USM illustrate the application of mathematics. For example, Vector Statics (ENT 260) applies topics from college algebra, trigonometry, and calculus. These topics include solving equations and triangles, using trigonometric identities, and integration. Industrial Quality Control (IET 302 relies on statistical principles such as measures of dispersion, central tendency, and sampling distributions. Engineering Graphics (ENT 100) and Computer-Aided Drafting and Design (ENT 320) require a clear understanding of the Cartesian coordinate system, which is introduced in college algebra. A more detailed discussion based on the first author's experience, who is pursuing a BS degree in industrial engineering technology, will be presented.

2:20 CHARACTERIZATION OF VEHICLE TEST COURSES BY POWER SPECTRUMS

Andrew W. Harrell, Engineer Research and Development Center, Vicksburg, MS 39180

In the last several years smaller robotic vehicles and all-terrain vehicles have introduced a need to reevaluate some of the previous vehicle testing methodology that we use to test vehicles. The mobility branch of the Geotechnical and Structures Laboratory has several vehicle ride courses used to evaluate the dynamic vibrational effect of different types of terrain on vehicles. This talk will explain ways we characterize terrain for vehicle testing and generate performance analysis. Mathematical formulas have been derived for the characterizing the terrain in terms of the dimension of its power spectral density curve. Also, the slope of the line relating the elevation profile detrending length to some power of the detrended root mean squared (RMS) value is used. The talk will present the plots of some courses and their power spectral density curves and then consider several different detrending values in order to illustrate the application of the formulas. Then it will give a short discussion of how the math ematical theory of wavelets is related to the RMS averaging and detrending kernels.

2:40 AN INTERACTIVE MATHWRIGHT MICRO-WORLD FOR FUNCTIONAL SYMMETRIES

Kanchan Manaktala, Alcorn State University, Alcorn State, MS 39069

Symmetry of a function is a significant characteristic of a function which the students in college algebra and precalculus courses must learn. They should also be able to relate the symmetries to even and odd functions. We have used mathwright authoring software to develop an interactive microworld to explore symmetries of functions and graphs defined by parametric equations. This microworld can easily be read off the web also, and therefore used for distance learning.

3:00 Divisional Business Meeting