Internet, a connection of networks has unified the world with no boundaries and limitations. It fulfilled the dream of being virtually present in any part of the world within no time. Over the years internet spanned into every corner of the world and a tremendous growth in the use of internet had been noticed. Internet, with its numerous advantages and support of wide variety of applications has leaded the consumer demand for more bandwidth and high availability to meet his requirements. Addressing these factors great development has taken place in recent years in the field of network technologies. A new era of Gigabit and Terabit Ethernet technologies signify the developments that address the growing demands. With the development in Ethernet technologies the demand for new router architectures that could process the data at gigabit rates also has increased. Many different router architectures have been proposed and implemented to meet these requirements. In this thesis, different router architectures have been studied and explained in detail. Many factors that affect the performance of the router have been reviewed and solutions to overcome the limitations have been addressed in detail. Addressing the solutions for limitations seen in legacy router architectures, the author presents the design, operation of a proposed new routing architecture [16]. The proposed architecture introduces the operation of RDMA into router architecture to minimize the processing and forwarding delay, which improves the performance of routers. The operation of an RDMA enabled router and the operation between two RDMA enabled routers have been clearly explained. The author also presents the mathematical models for evaluating the delay within a router and between two routers, processor utilization and memory utilization for both the legacy and RDMA enabled router architectures. In order to understand the router operation in detail and to enable future researchers to work on router architecture, a simulator with router operation has been built for both the router architectures. The results show that the proposed architectural operation improves the performance of routers even in heavy traffic networks. / Wichita State University, College of Engineering, Dept. of Electrical and Computer Science Engineering / Includes bibliographic references (leaves 86-89)

The aviation industry is interested in advance technology that includes composites and friction stir welding. As there are many advantages to friction stir welding (FSW), there are always ways to improve a new process such as friction stir processing/welding. One such improvement is to produce a metal matrix/carbon nanotube (CNT) composite via FSW. It is theorized in this work that a weld can be tailored to desired mechanical and physical properties. The background for such research is needed, and that is the purpose of this thesis. Nickel and CNTs FSW composite material was produced by welding over a powder filled groove on the surface of 2024-T3 plate. Mechanically compacting the powder was found to result in no noticeable loss of powder. Further, the nickel powder welded with the parameters used in this experiment was found to create a well distributed composite when the groove was 0.05” deep. When the groove was deeper then the nickel particles tended to stay agglomerated just below the surface in a line. The line of nickel particles started from the advancing side and as the groove was made deeper the particles in a line move toward the retreating side. FSW/CNTs composite material was different with different depths of grooves. For a 0.05” depth grove the CNTs were seen to form lines in the metallography going down from the top advancing side to the upper mid-section of the advancing side of the weld. With deeper groves the CNTs would stay agglomerated starting from the original position just below 0.05” deep groove and at greater depths the CNTs would move up in circular motion until the CNTs reach the surface of the weld. The strength of a FSW/CNT composite under the welding parameters used was found to produce a weaker weld than the base aluminum weld. Cracks were seen to form along the CNTs lines going from the top advancing side to the upper middle advancing side. The tool used in this study was a high welding speed tool. Finally the flow of nickel and carbon nanotubes was analyzed, and the distribution of the particles after welding was characterized and summarized. To successfully produce FSW/CNT composite material, it is recommended that a tool designed for slow, high temperature friction stir processing, one that maximizes maelstrom flow, be tested in future work. / Wichita State University, College of Engineering, Dept. of Mechanical Engineering / Includes bibliographic references (leaves 117-121)

We present a general scheme for implementing fault-tolerant quantum computation. We discuss a general two-dimensional architecture of qubits involving only linear nearest neighbor interactions. Between the qubits by using ancillas, we show how to implement gate operations for encoding, error correction, fault-tolerant quantum computation and decoding procedures in our design. The architecture is designed with two different coupling parameters ξ1and ξ2 between the qubits. A universal set of gate operations (Controlled-NOT, Hadamard, Phase, T) are performed on the encoded logical qubits fault tolerantly, by pulsing the bias on the target qubit to a certain value for a chosen time duration. Initially, we designed an architecture (discussed in Appendix A) for fault-tolerant computation that was capable of correcting errors, non fault-tolerantly. Since it is possible that the error correction circuit itself propagate errors, we designed an architecture that corrects errors, fault tolerantly. Finally, we compare our architecture to an existing architecture for fault-tolerant computation employing a linear one-dimensional nearest neighbor array of qubits and show how ours is more efficient. Even though all through this work we use the specific instance of the 7-qubit Steane code in describing our gate implementations, our method can be extended to all systems employing transversal gates for fault-tolerant quantum computation. / Wichita State University, College of Engineering, Dept. of Electrical and Computer Science Engineering / Includes bibliographic references (leaves 21-22)

The roughness of a surface influences the surface mechanical properties, such as elastic modulus, hardness and so on. In a plain strain condition, a material under deformation does not undergo change in Elastic Modulus, but when it has voids it undergoes a change in Elastic Modulus due to damage caused by indentation or deformation. Past studies have shown how the behavior of hardness changes by surface roughness of the material using spherical indenter on the spherical surfaces. The effect of the elastic modulus changes with the roughness however has not been investigated. The main aim of this thesis is to develop the methodology to estimate the behavior of elastic modulus and hardness of the material with different surfaces and when the material has damage such as voids, cracks, etc. The models used in this thesis are developed using Msc Patran and simulated using finite element analysis code LS-DYNA. In this thesis four models are developed with different surfaces to estimate the behavior of elastic modulus and hardness by nanoindentation using a sharp tip conical indenter. The change in modulus and roughness are estimated by nano-indentation using Oliver and Pharr’s theory. The penetration depths normalized with respect to a roughness scale parameter, and the effective radius encountered by the indenter, a first order model of roughness dependency of hardness is of nano-indentation on a fractal surface. It is shown, as the roughness of a surface increases, the hardness measured at depths comparable with the roughness scale deviates increasingly from the actual hardness. Not only the roughness influences the material properties but also when the material has damage in, it also influence the material properties. For this problem, the models are developed by placing some voids in the material and the behavior of elastic modulus and hardness of material with and without damage are estimated. Finally, the penetration depth estimated from nano-indentation is compared using the classical Hertzian theory and interestingly found that this theory can also be used for nano-indentation of metals and solids. The results from this research can be utilized to estimate roughness and modulus degradation caused by nano-indentation. / Wichita State University, College of Engineering, Dept. of Mechanical Engineering / Includes bibliographic references (leaves 79-81)

This thesis examines Internet Small Computer System Interface (iSCSI) storage architecture and introduces a framework to furnish Quality of Service (QoS) for the qualified storage traffic. The framework incorporates IO scheduler that does priority scheduling for desired IO request and receives higher degree of service in its journey through the network by adopting Differentiated Services (DS) network architecture with each node exhibiting Expedited Forwarding (EF) behavior. In this thesis, to achieve end-to-end QoS for the application, a novel functional module is introduced in storage server to provision QoS for the requested data and the proposed framework is examined subsequently with an analytical model to estimate the maximum delay bound. The results derived in the analytical model are compared with the available experimental results in the literature. The distinct implication of the proposed framework is in achieving its objectives of guaranteed transmission of data with deterministic maximum delay, without modifying existing structural components and network protocol semantics. / Wichita State University, College of Engineering, Dept. of Electrical and Computer Science Engineering / Includes bibliographic references (leaves 60-65)

Decreasing noise and improving efficiency is an important trend for the future jetliner industry. The predominant sources of aerodynamic noise are created from attached turbulent boundary layers (TBL) as well as separated flows. The current numerical effort will focus on geometry of a fence to crudely model a windshield wiper blade for a large jetliner. This will produce an unsteady separated flow field. The recent literature lacks some of the characteristics that would make it most useful to aircraft applications. The numerical simulations performed were validated with a backward facing step experiment at M = 0.15. The backstep geometry was then altered one feature at a time until an angled fence system was produced. All simulations except the final were performed with conditions that matched the wind tunnel experiment. The final simulation was performed at cruise conditions at 35,000 feet, with rough flight deck conditions provided by the Boeing Company. The Mach number of the incoming boundary layer flow is 0.5. Results of mean velocity fields were in excellent agreement with experiments. Turbulence intensity as well as pressure quantities agreed. Spectra of pressure fluctuations followed trends of past experiments. Pressure fluctuation magnitudes were over predicted just as many other numerical investigations in literature concluded. / Wichita State University, College of Engineering, Dept. of Aerospace Engineering / Includes bibliographic references (leaves 82-86)

With current safety regulations, standards and safety equipment, commuters have never been safer while driving an automobile. An increasingly popular past time is to restore or repair antique and vintage automobiles that do not benefit from the current safety standards, regulations and equipment. The driver and occupants of the vehicles are at a much higher risk of injury or fatality because of a lack of current safety devices. Older vehicles also suffer from poor ride and handling characteristics that will affect how well the vehicle will behave in an accident avoidance event. Due to the lack of safety equipment and regulations for older vehicles, and poor ride and handling characteristics, a full vehicle finite element model and multi body dynamic model was created of a 1965 Nissan Patrol G60. The computer models were used to assess the crash and vehicle handling characteristics of the vehicle. The crashworthiness is evaluated by comparing the finite element models performance in a full frontal crash per NCAP criteria. The finite element model was validated by comparing acceleration pulses with an existing finite element model of a 1994 C1500 light pick up truck. Once validated, the acceleration pulses at different location in the finite element model were compared to physical test data of a 2007 jeep wrangler with similar construction and weight. The differences in the crash pulses were used to evaluate the crashworthiness of the pre safety standard vehicle. To evaluate the ride and handling of the vehicle, a multi body dynamic model was developed and simulated performing a single lane change event. The measure for improvement was to minimize the ride steer of the vehicle during the single lane change event. A full factorial design of experiments was created to minimize the ride steer by modifying the rear suspension mounting points within an allowable design space. The results from the finite element model correlated well with the validation model and thus can be used with reasonable assurance for evaluating safety improvements to the vehicle and comparison to physical test data. Comparison with physical test data showed that the current configuration of the vehicle performs well considering the age of the vehicle. With the data from the computer aided crash test and suspension analysis, improvements to the 1965 Nissan patrol can be identified and used to update the vehicle to meet current safety standards. A similar approach can be applied to any vehicle manufactured prior to safety standards and regulations. / Wichita State University, College of Engineering, Dept. of Mechanical Engineering / Includes bibliographic references (leaves 89-91)

Chronic obstructive pulmonary disease (COPD) is a major health problem that affects 16 million people in the US, and is currently the fourth most common cause of death. Although the pathogenesis of COPD is poorly understood, current studies indicate that COPD is a multi-factorial disorder characterized by a cigarette smoke-induced cycle of oxidative stress, alveolar septal cell apoptosis, a protease/antiprotease imbalance, and chronic inflammation. An array of serine (HNE, PR3), cysteine (cathepsin S), and metallo- (MMP-1, MMP-9, MMP-12) proteases released by neutrophils, macrophages, and T lymphocytes contribute to the degradation of lung connective tissue and mediate a multitude of signaling pathways associated with the pathophysiology of the disorder. Re-establishment of a protease/antiprotease balance by utilizing potent and selective protease inhibitors is a promising approach for the development of potential therapeutics for COPD. We describe herein the design, synthesis and biochemical evaluation of a novel class of mechanism-based inhibitors of HNE that exploit the catalytic machinery of the target enzyme to generate a Michael acceptor. Subsequent reaction with an active site nucleophilic residue leads to inactivation of the enzyme. A noteworthy feature of the inhibitors is their ability to interact with the S1-Sn’ subsites of the target enzyme. / Thesis (M.S.)--Wichita State University, College of Liberal Arts and Sciences, Dept. of Chemistry / Includes bibliographic references (leaves 42-46)

This report presents the fabrication of a micro fluidic device using the UV lithography method in order to separate nonmagnetic fluoresbrite carboxy microspheres from the mixed ferrofluids in microchannels. The microfluidic device is fabricated using a SU8-50 negative photoresist that is exposed to the UV lights with a mask (Y shape) on the top of the photoresist. By applying the external magnet from a side of the microchannel, the fluoresbrite carboxy microspheres and ferrofluids are separated into different channels because of the magnetic force acted on the nonmagnetic spheres. During the fabrication, a number of different parameters, such as UV exposure times, UV power, photoresist thickness, etc. were conducted and optimized for our needs. In addition, in the magnetic field testing, different pumping speeds, and particle concentrations associated with the different distances between the magnet and the microfluidic system were studied for an efficient separation. / Wichita State University, College of Engineering, Dept. of Mechanical Engineering / Includes bibliographic references (leaves 64-67)

The following research aims to determine whether or not the maxillomalar suture can determine group affiliation via quantitative assessment. Until now the standard method of determination involved a visual assessment of the shape of the suture, which can be biased. Quantification of this method would allow for more accuracy in the determination and greater repeatability of results. Twenty-three measurements were taken on two hundred and nine Black and White males at the Todd Collection in Cleveland and fifty Native American individuals of unknown age and sex from the Libben Site Collection housed at Kent State University. Three measurements were developed to measure the bimaxillary breaths from different points directly on the maxillomalar suture and three qualitative classifications were used to assess the visual shape. The three bimaxillary suture measurements appear to be the best indicators for group affiliation compared to the other metric measurements. Native Americans had the highest mean for those measurements, indicating the widest facial crania, while Blacks had the lowest means. The bimaxillary breadth measures were all statistically significant at the .05 alpha level. There was slight difference between groups in the maxillomalar suture length noted and this difference was statistically significant on the right side only. The visual qualitative assessment showed recognizable variation in the White and Black samples and very little in the Native American sample. / Wichita State University, College of Liberal Arts and Science, Dept. of Anthropology / Includes bibliographic references (leaves 53-57)