Physics and engineering.

Chair: Alexander B. Yakovlev, University of Mississippi 

Vice-chair: S. Kant Vajpayee, University of Southern Mississippi

THURSDAY MORNING

Meeting Room 4

Advances in Numerical Modeling Techniques

8:00 ELECTROMAGNETIC SCATTERING FROM CHIRAL MEDIA USING THE FINITE DIFFERENCE FREQUENCY DOMAIN TECHNIQUE

Lokman Kuzu (1), Veysel Demir (2*), Atef Z. Elsherbeni (2), and Ercument Arvas (1), (1) Syracuse University, Syracuse, NY 13244 and (2) University of Mississippi, University, MS 38677

The analysis of chiral materials has recently been an important topic in computational electromagnetics. Numerical analysis of chiral materials has been carried out using a variety of numerical methods, such as the method of moments (MoM), the finite-difference time-domain (FDTD) method, boundary value solutions (BVS) and so forth. In this paper, The finite difference frequency domain (FDFD) formulations have been developed for chiral materials analysis, and the validity of the formulations for one and two dimensional scattering problems has been proved by comparing the numerical results to exact solutions whenever possible. The influences of the chirality on the scattering are investigated. Numerical results for bistatic echo widths are presented and compared with reference solutions and it is found that the proposed FDFD method shows good agreement. It is realized that the presented method is relatively easy to program and can be applied to a wide variety of problems of complex and composite structures efficiently.

8:20 HYBRID FDFD AND MOM TECHNIQUES IN CONJUNCTION WITH THE ITERATIVE MULTIREGION ALGORITHM FOR THE SOLUTION OF LARGE ELECTROMAGNETIC PROBLEMS

Mohamed Al Sharkawy*, Veysel Demir, and Atef Z. Elsherbeni, University of Mississippi, University, MS, 38677

This work presents a hybrid technique, which combines the desirable features of two different numerical methods, finite difference frequency domain (FDFD) and the method of moments (MoM), to analyze large-scale electromagnetic problems. This is done by using the two techniques individually on separate regions of the scattering problem domain and then applying an iterative procedure between these regions, to calculate the total scattering from the entire domain. This iterative procedure is referred to as iterative multi-region (IMR) technique, which requires the solution of fields in the sub-regions a number of times instead of one solution of the complete computational domain. This technique effectively reduces the size of the required memory, especially for practical and three-dimensional problems. In this work the presented technique is applied on two-dimensional scatterers, and the bistatic echo widths are calculated. This hybrid FDFD/MoM approach takes advantage of the capability of the FDFD to analyze inhomogeneous bodies with arbitrary material properties and that of the MoM to model large metallic structures with less computational memory requirements. Both numerical methods provide a much stable solution relative to other available methods and a more convenient procedure for performing the interaction between the sub-regions based on well-known theorems

8:40 SOLUTION OF MAXWELL'S EQUATIONS ON BOUNDARY-FITTED GRIDS USING A CONVOLUTIONAL PERFECTLY MATCHED LAYER TECHNIQUE AS AN ENERGY ABSORBING BOUNDARY CONDITION

Terry Gerald* and Atef Z. Elsherbeni, University of Mississippi, University, MS 38677

The finite difference time domain (FDTD) is extended to boundary-fitted grids by transforming Maxwell's equations defined on a curvilinear physical grid to a modified form defined on a uniform Cartesian grid. The FDTD method is used to solve the transformed Maxwell's equations on the uniform Cartesian grid. The solution thus obtained is then mapped back to the original physical grid. Elimination of wave reflections at grid boundaries is accomplished by augmenting the uniform Cartesian grid with a convolutional perfectly-matched layer (CPML) absorbing boundary condition. The method is applied to several boundary-fitted grids representing irregular physical domains. Results obtained indicate the method is capable of generating an accurate estimate of the time-varying electromagnetic field distributions throughout a physical region.

9:00 FDTD ANALYSIS OF A PROBE-FED DIELECTRIC RESONATOR ANTENNA ARRAY WITH HARD HORN FOR SPATIAL POWER COMBINER 

Yizhe Zhang*, Ahmed A. Kishk, Alexander B. Yakovlev, and Allen W. Glisson, University of Mississippi, University, MS 38677

Probe-fed dielectric resonator antenna (DRA) arrays with hard horn excitation are investigated in an oversized dielectric loaded waveguide for their use in waveguide-based spatial power combining systems. Dielectric loaded oversized rectangular waveguide and hard horn are first utilized to achieve a uniform field distribution across the horn aperture at the design frequency. Then, 3 X 3 DRA array is studied in the whole environment. Metal plates inset into the oversized waveguide are used to decrease the E-plane mutual coupling between the array elements. Design of inter-element antenna spacing, length of inset plates, and the distance of the hard horn walls to the array elements are investigated to optimize the system for scattering and coupling characteristics. The coupling from the horn to each of the nine DRA elements is approximately -10 dB, which indicates that every element obtains 1/9 power. The Finite-Difference Time-Domain (FDTD) approach with region-by-region discretization scheme is proposed for the analysis of the entire system to reduce memory requirements and simulation time. Numerical Results for the scattering parameters are demonstrated for the example of 3 X 3 DRA array excited by a hard horn and compared with the results obtained using the commercial software HFSS.

9:20 OPTIMIZATION USING THE ORTHOGONAL ARRAYS FOR UNEQUALLY SPACED LINEAR ARRAY ANTENNAS AND GLOBAL NUMERICAL SOLUTIONS

Wei-Chung Weng*, Fan Yang, and Atef Z. Elsherbeni, University of Mississippi, University, MS 38677

The robustness of the use of orthogonal arrays (OAs) for optimization is described in this study. The OAs help provide high performance evolutionary optimization method. Using the systematic and efficient features of OAs, one can solve n-dimensional, linear, and nonlinear optimization problems. Compared to other optimization techniques, such as the genetic algorithm and particle swarm optimization, the OAs based optimization procedure is much easier to implement and requires less computational resources. Two test cases characterized as nonlinear and two-dimensional functions are analyzed using the procedure presented here to verify the performance for finding the global maximum. The results show that the global maximum can be found among many local maxima after few iterations. Linear antenna arrays have received great attention in the electromagnetic community. In this study, we also have utilized the OAs optimization procedure in order to suppress the undesired sidelobe levels (SLL) and at the same time to narrow down the beamwidth for a ten-element unequally spaced linear array of isotropic radiators. The results show that lower SLL and narrower beamwidth are achieved. The results also show that the algorithm presented here can be used not only for signal-objective global optimization but also for multi-objective global optimization.

9:40 Break

Antenna Research and Applications

10:00 DESIGN OF A WIDEBAND MICROSTRIP POWER DIVIDER

Veysel Demir, Dalia Elsherbeni*, Darko Kajfez, and Atef Z. Elsherbeni, University of Mississippi, Oxford, MS 38677

Power dividers are passive RF components that are commonly used to split and distribute power in various proportions to different components of networks. Usually, power dividers support narrow band. However, for wideband operations the limited band of power dividers prevents the use of network in its maximum capacity; e.g. a wideband antenna array will operate only in the frequency band of the supporting network. In this contribution, a power divider is proposed and designed that can support the wideband operation of the terminal network. In a power splitting circuit, in which the input line feeds two output lines, the input power on the 50 Ohms line faces a mismatch of 25 Ohms at the intersection of the input line and the two output lines. In order to eliminate this mismatch over a wideband a Klopfenstein taper is designed to match 50 Ohms to 25 Ohms for the frequencies above 2.3 GHz with a maximum reflection of -30 dB using simulation tools. Then, a junction from a 25 Ohms line to two 50 Ohms lines is designed. These two designs are cascaded to form a wideband high-pass power divider that effectively operates above 2.3 GHz. The designed circuit is built and measurements were performed. The measurement results agree very well with the simulations and verify that the proposed power divider design can be used in a bandwidth from 2.3 to 10 GHz.

10:20 DOUBLY-SIDED EXPONENTIALLY TAPERED SLOT ANTENNAS

Zach Hood and Erdem Topsakal*, Mississippi State University, Mississippi State, MS 39762

This paper compares a traditional single-sided double exponentially tapered slot antenna (DETSA) for use in ultra-wideband (UWB) applications to a new double-sided configuration. Applications for UWB (3.1-10.6 GHz) technology include broadband wireless communications, which benefit greatly from increased bit rates available to higher bandwidth. Tapered slot antennas have long shown great promise for achieving wide bandwidths. It has also been demonstrated that by exponentially tapering both the inner and outer edges of the traditional Vivaldi antenna, the bandwidth and other characteristics may be greatly improved. While traditional bowtie antennas lack the ability to operate over the full UWB range, a double-sided printed bow-tie antenna was introduced with bandwidth exceeding the required 3.1-10.6 GHz. By applying this concept and some scaling techniques to the DETSA, a compact UWB antenna is created. The antenna is capable of operation over the UWB band as well as higher frequencies, which opens it to the possibility of use in future applications. The antenna also exhibits strong directivity in along its central axis, as well as above and below the E-plane, as is expected from an end-fire design. While the double-sided DETSA does offer performance at least on par with single-sided configurations, the main advantage of the antenna is its extremely compact dimensions. Results for an optimized version of the double-sided DETSA will be presented.

10:40 FINITE ELEMENT-BOUNDARY INTEGRAL SIMULATION OF DOUBLE-SIDED ROUNDED BOW-TIE ANTENNAS FOR UWB COMMUNICATIONS

Tutku Karacolak and Erdem Topsakal*, Mississippi State University, Mississippi State, MS 39762

Finite Element-Boundary Integral Method (FE-BI), which is a very powerful numerical technique for radiation and scattering analysis of antennas, has been applied to a double sided rounded bow-tie antenna covering the spectrum from 3.1 GHz to 10.6 GHz for Ultra Wide Band (UWB) communications. Distorted hexahedral elements are widely used in analyzing complex 3-D electromagnetic problems. Because of the distorted shape, they offer higher flexibility, and are more suitable to automatic mesh generation of 3-D regions compared to other type of finite elements. For validation purposes, we also simulated the antenna with a commercial software HFSS, which uses Finite Element Method with tetrahedral elements. The antenna has omni-directional radiation characteristics and return loss below -10 dB for the whole frequency band. Rounded bowtie patches work better than the conventional ones for the UWB communications. Results regarding antenna parameters such as return loss, radiation pattern and gain will be presented.

11:00 DIELECTRIC RESONATOR ANTENNA ARRAY FOR MICROWAVE BREAST CANCER DETECTION

Wei Huang* and Ahmed A. Kishk, University of Mississppi, University, MS 38677

Breast cancer is the most leading cause of cancer death (other than lung cancer) of women according to the recent released statistics by the American Cancer Society. Early detection of breast cancer is key element for reducing the mortality. Currently X-ray mammography is the most widely used tool to diagnose and evaluate breast cancer. However, mammography and other ionizing X-ray based methods are not safe for the patients since the ionizing radiation properties of the X-ray may induce a new breast tumor. Also mammography is not precise since it is very sensitive to lesions (but not all lesions are cancerous). Moreover, mammography is too costly especially for low-income women and women live in developing countries. As a new promising non-ionizing and noninvasive radar-based breast detection method, microwave breast cancer detection based on the contracts in dielectric properties between healthy (& epsilon [.sub.r]=9 and & sigma=0.4S/M) and malignant (& epsilon [.sub.r]=50 and & sigma=4S/M) tissue, can achieve early detection and has the advantages of low health risk, noninvasive and comfortable, and cost effectively. One of the biggest challenges of microwave breast cancer detection is to design a right antenna for this application. In present paper, we design a compact microwave sensor based on compact wideband dielectric resonator antenna array element (14mmX14mmX6mm). The design is based on numerical simulations using commercial software. The compact antenna array with wide bandwidth is presented. The comparisons of tissue with tumor and without tumor cases are shown in both frequency domain and time domain.

11:20 A STUDY ON MICROSTRIP IMPLANTABLE ANTENNAS FOR MEDICAL TELEMETRY

Jose Pvillalta and Erdem Topsakal*, Mississippi State University, Mississippi State, MS 39762

Today, implantable antennas are increasingly used in medical fields. Some of these applications include cardiac pacemakers, radio frequency identification (RFID), ultra low power wireless devices for in-body monitoring, muscle stimulators, electromagnetic hyperthermia for cancer treatment etc. As designers develop new implanted medical devices taking advantage of RF technology to improve the quality of care for patients, antennas are key to these new systems. Numerical simulation tools that are fast and accurate are crucial in developing reliable and efficient implantable antenna systems. In this study, we present two common microstrip antennas; spiral and serpentine, with different design parameters. They are evaluated for the 402-405 MHz Medical Implant Communications Service band (MICS). The effects of the shape, length, size, location of feed point, substrate and superstrate materials, and their thicknesses are evaluated. The antenna is then embedded in the forearm and simulations are carried out using a robust Finite Element Boundary Integral Technique. Results regarding the antenna return loss, near and far fields will be presented for variety of antenna locations and diverse antenna parameters.

THURSDAY AFTERNOON

Meeting Room 4

Image Processing and Applications

1:10 THE AUTOMATED IMAGE REGISTRATION FOR MULTISENSOR CASE

Adnan Orduyilmaz* and Nareenart Raksuntorn, Mississippi State University, Mississippi State, MS 39762

A 3-band multi-spectral sensor system has been developed for acquiring remotely sensed aerial imagery using unmanned aerial vehicle (UAV). In this system 3 different cameras are recording images at different bands (Green, Red, and Near-Infrared (NIR)). Since the multi-spectral data are acquired using three different digital cameras, these three frames of imagery should be spectrally co-registered before the other products such as false color composites (FCC) or vegetation indices are generated. Advanced image processing techniques can be utilized to automate this process. The area-based method is used for the registration problem. This approach has four main steps. At the first step, the control points are selected using spectral variance. Second step is searching the corresponding windows using correlation coefficient similarity metric. At the third step, the parameters of affine transform are calculated. Last step is applying the transformation and filling the blank pixels by the nearest neighbor (NN) method. The control points are selected from the areas, which have the highest spectral variance. The centers of windows are defined as control pairs. The windows in the base image with different sizes will be selected. The optimum window size is computed after testing all windows size from 11 pixels to 61 pixels. The quality of the registered images are done by visually and using the distance difference between the distinct objects such as roads or houses in registered images and base image. The results of the registration process will be presented.

1:30 AN EMI LANDMINE DETECTION SYSTEM USING PULSE-RESPONSE MEASUREMENTS BASED ON MAXIMAL LENGTH SEQUENCE EXCITATION

Ying Chi* and Paul M. Goggans, University of Mississippi, University, MS 38677

The use of the maximal length sequence (MLS) method to measure the pulse response of linear systems is well established in areas such as room acoustics where it is commonly used to determine impulse response of concert halls. Here we present the use of the MLS method to determine impulse response (pulse response) in a new application area, electromagnetic induction (EMI) landmine detection. In this paper we introduce the theory of EMI landmine detection for low-metal-content landmines based on pulse-response measurement and describe the use of MLS sequences to measure the pulse response of an EMI landmine detection system. As part of this research, an experimental EMI landmine detection system was designed and built. In this paper, measured pulse-responses for the experimental system are presented and compared with theoretical results. The comparison shows that the MLS measurement method applied to EMI landmine detection is both accurate and repeatable.

1:50 IMAGE PROCESSING AND TARGET IDENTIFICATION USING GRAPHICAL PROCESSING UNITS

Matthew Inman*, Chye Hwa Loo, and Atef Z. Elsherbeni, University of Mississippi, University, MS 38677

Image processing and target identification are two related areas in which major research is currently being performed to increase the accuracy and utility of such systems. With new algorithms that are being developed for target tracking and identification the computational needs to perform them grow exponentially. It will be shown that graphical processing units (GPUs) that are commonly found in all modern computers can provide vastly increased power for processing images with these new techniques. With the proper programming and application of these graphical processing units, speed increases of orders of magnitude can be reached for some techniques. Both the application of common and custom image processing algorithms and increases in processing speed will be presented.

2:10 REAL-TIME DISTORTION INVARIANT OBJECT CLASSIFICATION WITH SDF BASED FJTC

Chye Hwa Loo* and Atef Z. Elsherbeni, University of Mississippi, University, MS 38677

The task of object classification is complicated by variations in the three-dimensional (3D) object, which translate into distortions in 2-D images. Distortions due to scale variations, illumination, background clutter, occlusion, and in-plane and out of plane rotation hinder successful recognition. The pattern matching for invariance classification requires a large amount of data and computation time. Proposed here is an efficient distortion invariant object classification algorithm for real-time fringe-adjusted joint transform correlator (FJTC) based automatic target recognition (ATR) system. The proposed classification technique employed the synthetic discriminant function (SDF) in the generation of distortion invariant correlation filter sets. The optoelectronic FJTC is then used to provide correlation of the filter sets and input under a proper arrangement. This classification method is simple and fast and hence is relevant to be in use by real-time ATR systems. A description of the optoelectronic system and the entire process is presented. In addition, simulation results are provided to prove the effectiveness of the proposed system in the classification of objects invariant to scale as well as in-plane and out-of-plane rotations.

2:30 Break

Wireless Propagation

2:40 PERFORMANCE OF DUAL-BRANCH MAXIMAL RATIO COMBINING DIVERSITY OVER NONIDENTICAL CORRELATED WEIBULL FADING CHANNELS USING PADE APPROXIMATION

Mahmoud Ismail* and Mustafa M. Matalgah, University of Mississippi, University, MS 38677

In this study, we evaluate the performance of the dual-branch maximal ratio combining (MRC) diversity scheme in non-identical correlated Weibull fading channels with arbitrary parameters. We first use Pade approximation (PA) to find closed-form rational expressions for the moment generating function (MGF) of the output signal-to-noise ratio (SNR) of the MRC receiver. Different performance measures such as the outage probability and the average symbol error rate (SER) for different linear modulations such as 8-phase shift keying (PSK) and 16-quadrature amplitude modulation (QAM) are then presented using the well-known MGF approach. Furthermore, the effect of input SNRs unbalancing, the severity of fading and the degree of correlation, between diversity branches, on the system performance are also studied. Our analytical results are validated by comparing them to computer simulations and we show that the PA technique is indeed a convenient tool for such performance evaluation studies.

3:00 OUTAGE PROBABILITY ANALYSIS IN A COOPERATIVE UAVS NETWORK OVER RAYLEIGH AND NAKAGAMI-M FADING CHANNELS

Ibrahim Abualhaol* and Mustafa M. Matalgah, University of Mississippi, University, MS 38677

In this work, the outage probability in a network of cooperative unmanned airborne vehicles (UAVs) is studied analytically. A master-slave topology with one ground control unit (GCU) is assumed where the GCU communicates directly with a master UAV. The master-slave and master-GCU links should satisfy a certain minimum data rate, which is related to the application to which the UAVs are assigned, to be considered in service. The outage probability of the system is defined as the probability that either the rate of transmission over any of the links goes below the predefined minimum value for this link, or the GCU-master link is not able to transmit the collected data from all UAVs besides the minimum rate related to the master UAV itself. The outage probability is considered over Rayleigh and Nakagami-m fading channels. The resultant expression for the outage probability can be used to provide guidelines for system designers to minimize the outage probability of the cooperative UAVs network. In the same time, it could be used to maximize the system throughput with the constraint of not exceeding a certain outage probability limit.

3:20 Divisional Poster Session

A NOVEL TECHNIQUE FOR STUDYING PHASE TRANSITIONS IN GRANULAR MATERIALS

Francis Tuluri, Jackson State University, Jackson, MS 39217

Granular materials composed of discrete particles exhibit a great variety of patterns and phases reflecting complexity of interactions between contiguous particles. A quick survey of the literature on the study of granular materials is mostly focused on symmetrical objects like spheres and cylinders. In the present work, a simple experimental model of vibrated granular materials is explored for the study of phase transitions occurrence by spontaneous symmetry breaking. The model considers two dimensional systems of asymmetric cylindrical objects. A simple physical mechanism of the phase transitions observed is accounted for in terms of symmetry breaking.

FRIDAY MORNING

Meeting Room 4

Future of Energy Resources

8:30 HOW IS THE U.S. INDUSTRY DOING QUALITY-WISE?

S. Kant Vajpayee, University of Southern Mississippi, Hattiesburg MS 39406

The American Society for Quality (ASQ) keeps track of the quality performance of the major sectors of U.S. industry. It has developed a yardstick, called Quality Index, to do this. Each quarter of the year it publishes its findings, based on a prominent economic indicator, called American Customer Satisfaction Index (ACSI). Established in 1994, ACSI is a measure of the quality of goods and services in the US market. The ACSI monitors trends in customer satisfaction. The Ross Business School at the University of Michigan, the ASQ, and the international consulting firm CFI Group work together as partners in conducting the study. More than 200 companies in 41 industries participate in the study. The results are reported on a scale of zero to 100. Eight economic sectors are covered; namely, Manufacturing Durables, Manufacturing/Non-durables, Transportation/Communications/Utilities, Retail, E-Commerce/E-Business, Finance/Insurance, Services, Public Administration/Government (local, state, and federal). More than 65,000 consumers are interviewed annually. Since the baseline study in 1994, a typical sample has amassed more than half-a-million respondents. Cell phone service providers are the lowest-rated industry, though their manufacturing counterparts--Nokia, Motorola, and Samsung--have a much higher perceived quality rating. Airlines are the second worst. Fast-food chains remain among the lowest-rated industries, though pizza chains do a great job in meeting customer expectations. Refer to www.asq.org for more information.

8:50 NUCLEAR FISSION ENERGY--THE ENERGY OF CHOICE

Amin Haque, Alcorn State University, Alcorn State, MS 39096

As the world becomes more industrialized, the demand for energy is increasing. Currently the energy sources available in the world are: fossil fuels, uranium (nuclear fuel), hydroelectric, wind, and solar. People are concerned about environmental quality, global warming, and acid rain caused by various air pollutants released from combustion of fossil fuels--the major source used in the world to produce electricity. Burning of fossil fuels requires expensive air pollution controls and extensive transportation systems. Nuclear fission produces virtually no air pollution, and the amount of waste produced is the least. Uranium is abundant and the overall cost of producing electricity is relatively small. One kilogram of natural uranium produces as much energy as 38.5 tons of coal. According to 2003 MIT published report, the generation cost for nuclear energy is 4.2 c/kWh, the same as coal without any carbon cost. According to 2004 report from the University of Chicago, the power costs of future nuclear, coal, and gas-fired power generation in the USA. range from 4.3 to 5.0 c/kWh, 3.5-4.1 c/kWh and 3.5-4.5 c/kWh, respectively. In the U.S. for 2001 the cost was 3.73 c/kWh for nuclear, 3.27 c/kWh for coal, and 5.87 c/kWh for gas. Nuclear energy plants are now safer because of the use of advanced technology and higher standards. New technologies and improved knowledge have made it possible to produce much safer light water nuclear plants.

9:10 NUCLEAR FUSION ENERGY--THE ENERGY OF THE FUTURE

Amin Haque, Alcorn State University, Alcorn State, MS 39096

A fusion reaction occurs at very high temperatures (10-15 million degrees Celsius) when the fuels are in a plasma state. The stars are natural fusion energy generators. In the labs, the fusion of deuteron and triton is more promising. About 17.6 million electron volt of energy per cycle are released. Deuterium is plentifully available from seawater, and tritium can be obtained by the breeding from lithium. One kilogram of fusion fuel would produce the same amount of energy as 10 million kg of fossil fuel. Since plasmas are very good electrical conductors, magnetic fields are used to confine and isolate them from the container walls. To produce self-sustaining fusion, part of the energy released is used to heat new reactant nuclei. Several methods are used to heat the plasma. The fusion energy is safer, and the likelihood of an accident is much smaller. Fusion reactions produce no atmospheric pollution, and low radiation levels. Commercially available fusion reactors are not expected for at least the next 35 years. In June 2005, it was announced that the first experimental fusion reactor designed to achieve sustained fusion reactions, the International Thermonuclear Experimental Reactor, will be built in Cadarache in Southern France at a cost of 10 billion dollars. This will be the first earthbound star. The technical requirements to achieve this goal are immense and a great challenge to the nuclear scientists. If they succeed, the rewards will be tremendous.

9:30 VIOLENT CELESTIAL EXPLOSIONS AND GAMMA RAY BURSTS

Amin Haque, Alcorn State University, Alcorn State, MS 39096

Gamma-Ray Bursts (GRBs) are the most powerful explosions and the brightest source of cosmic gamma-ray photons in the observable Universe, lasting anywhere from a few milliseconds to several minutes. They are brighter than a typical supernova and about a million trillion times as bright as the sun. GRBs are detected roughly once per day from random directions. Satellite measurements show that GRBs are isotropically distributed. Recent observations, and theoretical work link GRBs to supernovae in distant galaxies, billions of light-years (bly) away. This means that some GRBs actually originated while the universe was only a few billion years old. Although such stars died long ago, only now is the light from their explosive deaths reaching us. The relativistic fireball shock model show that these explosions appear very energetic because much of their energy is blasted outward in two narrow jets in opposite directions. This can result from either the merger of a double neutron stars which lead to short bursts (< 10 s) GRBs or by the collapse of the fast-rotating core of a massive star which leads to long GRB bursts (> 10 s) and could be associated with a supernova-like phenomenon. The Swift Gamma-Ray Burst obsevatory, recently launched by NASA, is expected to zero in on two GRBs a week as far away as 15 bly, representing the very first generation of stars, for a total of more than 200 during the planned two-year mission.

9:50 Break

Invited Speaker

10:00 WASTE WISE: COMPUTER WASTE

Pao-Chiang Yuan, Jackson State University, Jackson, MS 39217

Start from October 2001 until now, Jackson State University collaborates with Hinds County, Public Work Department received grant from State of Mississippi, and Department of Environmental Quality established computer recycling and training program. In the past years, 2388 Central Processing Units, 2864 monitors, 609 printers and more were collected. The program also refurbished 465 computer sets return back to the community; another 206 pallets of unwanted materials were shipped to the recyclers. The program received total 7 awards from Keep Jackson Beautiful, Mississippi Recycle Coalition and was chosen by United States Environmental Protection Agency (USEPA), Waste Wise program partner of the year. According to the institute for local Self-Reliance approximately 80 percent of obsolete electronics are being stored or warehoused somewhere before they will disposal of. Since the last three decades, computer industries are booming. Now, Personal Computer has become a necessity for our way of life. Everyone is focused on cyber technology. Cyber Technology is the theme for this new millennium: How many industries? Attention is given to the other end: when the Personal Computer (PC) has reached the end-of-its useful life. What do we do with them? In 2005, the National Safety Council projects more than 63 million personal computers to be retired according o a recent study. This paper discusses possible alternate methods, to solve the problems, and most importantly it makes the people aware of what kind of problems we are facing.

FRIDAY AFTERNOON

Meeting Room 4

Physical Phenomena and Experiments

1:00 THERMOACOUSTICS OF SMALL DEVICES

Carl Jensen, University of Mississippi, University, MS 38677

Small thermoacoustic devices are being built and modeled by our research group for applications such as microchip cooling and waste heat power generation. Due to their high frequency of operation, however, the optimal pore size for the stack material is very small making construction of conventional parallel plate stacks impossible. We are investigating the use of random packing of fibrous materials instead. The thermoacoustic properties of these random stack materials are being investigated using techniques common for sound absorbing materials. The extension of these techniques to thermoacoustics as well as the challenges in modeling a small thermoacoustic device will be discussed.

1:20 THREE DIFFERENT METHODS TO ACCOUNT FOR LOST VAPOR FROM LARGE USTS DURING THE SUMMER MONTHS IN THE SOUTHERN STATES

Leili Pirouzan and Sam Gordji*, University of Mississippi, University, MS 38677

The vapor loss from fuel tanks can be a significant environmental problem. The problem can impact air quality and bias leak detection compliance methods for the measurement of product into the ground water. The loss becomes larger for the larger tanks during the summer months in the southern states such as ours, "Mississippi." First to consider is the deterministic method using equations obtained through the laws of physics to account for the lost vapor gasoline. This method since it is deterministic is without the doubt the best approach and should yield the best estimate if one can capture the values of each parameter in the mathematical model. The second method uses the probabilistic approach and uses the method of regression analysis. We are now soliciting data from some companies that have large USTs. The third method is through the use of the applications of the Time Series Analysis. To apply this method several programs (more that ten) were written in SAS and Fortran using IMSL applying the applications of Time Series Analysis to tank inventory data. Two very important results were obtained from these analyses. 1. That the tank inventory data is not an "orphan child" and belongs to a large family of data sets called "random walk", and therefore, it has all the properties of a random walk data set and everything that is already been proved about random walk may be safely applied to tank inventory data. 2. The most important results obtained is that according to the previous research done by others in other fields such as air line travel, etc. the present trend of gas losses may be extended to forecast the future losses.

1:50 TEMPERATURE GRADIENT MEASUREMENT IN A SHALLOW WATER ENVIRONMENT

Patrick D. Coln*, John David Heffington, James Bell, and James P. Chambers, University of Mississippi, University, MS 38677

A temperature probe containing eight strategically placed precision integrated circuit temperature sensors has been developed and used to better understand the temperature gradient that forms in shallow ponds during the heat of the day. The motivation for this effort comes from the development of the Aquascanner Catfish SONAR to acoustically scan commercial sized catfish ponds. The unit is capable of providing detailed scans of catfish ponds. However, it is being run on large ponds in situ and as such the results are susceptible to variations from a variety of some what uncontrollable factors between experimental runs. In particular, sun loading acts to create vertical temperature gradients which cause refraction and act to bend sound rays toward the absorptive bottom. The temperature probe is designed to float to ensure constant relative depths with respect to the surface. Evaporation or precipitation may change the water depth during the measurement and either expose or submerge the upper probe. A total of eight sensors are spread out vertically at logarithmic intervals to measure the temperature gradient. Data is taken over multiple day cycles to observe diurnal heating with gradient formation during the day and a return to thermal equilibrium during the night. Measurements show that the temperature difference between the top and bottom of the water column, typically 36-48", can vary by 10 or more degrees during the strongest gradient.

2:10 WIND NOISE MEASUREMENTS OVER A FLAT PLATE

Jiao Yu*, Richard Raspet, and Jeremy Webster, University of Mississippi, University, MS 38677

Measurements by Kevin Dillion [MS thesis University of Mississippi, 2005] have shown that a thin layer of foam covering a flush mounted microphone attenuates wind noise as well as a spherical windscreen. This indicates that a better understanding of the interaction of turbulence with the shear boundary layer near a flat plane could lead to flush mounted systems with superior wind noise reduction. The experiments described here were performed to measure various properties of this interaction. Wind velocity profiles were measured in a wind tunnel close to hard and foam covered surfaces both with and without turbulence present. Measurements of velocity correlations with height were also taken, using two hot-wire anemometers placed near the surface. The implications of these measurements for predicting pressure fluctuations in flush mounted microphones will be briefly described.

2:30 Awards and Divisional Business Meeting