The purpose of this research has been the development of signaling protocol and associated architecture for Wave Division Multiplexing burst-switching network. The basic premise of this architecture is simple ? data, aggregated in bursts can be transferred from one end point to the other by setting up light path ahead of the arrival of the data. Optical Burst switched network is viewed as one pioneering effort to bring the most bandwidth available from the emerging dWDM technologies to end applications with minimum overhead and latency.

The classification of Internet traffic is an active research topic due to its applicability in the areas like differentiated services and network security. The introduction of voice, video and other real-time applications to the Internet has resulted in increasing demand for service differentiation and has triggered the need for change in traffic handling on the Internet. Traditionally, such a classification is done using the packet header field of ?port number,? which is a unique number associated with the application that generated the packet. In addition to adding complexity and extra computation in traffic handling, certain recent developments in networking techniques have rendered port numbers unreliable for this purpose. This motivates our scheme of classification that uses the distribution of packet sizes in a buffer or collected during a short time interval at a switch or a router. We demonstrate that the applications can be classified by these distributions. This ?implicit? classification builds a foundation for estimation and prediction of traffic mix, which is a long-term goal of this research project.

As feature sizes continue to decrease, fundamental properties of MOSFET devices begin to hinder the performance gains from one generation to another. The advent of the Tunneling Field Effect Transistor (TFET) provides hope for continued reduction in feature size whilst solving some of the scaling issues such as leakage current. The purpose of this work is to discuss key metrics that help to quantify the improvements among technology nodes, specifically a comparison between TFETs and traditional MOSFETs. Test structures that allow for the measurement of on and off current, device speed, variation as it relates to on current and threshold voltage, as well as SRAM yield and bitcell read and write noise margins are discussed. In addition, a slight modification to a rapid characterization test structure used to measure threshold variation is proven to help reduce leakage seen within the test structure. Lastly, the structures are actually fabricated in a 90nm bulk and a 45nm SOI process and measurements from the 90nm bulk process are presented.

The purpose of this research has been to develop technology necessary for Low-Cost Magnetic Resonance Imaging (LC-MRI) machines. This dissertation describes an overview of the tradeoffs necessary for LC-MRI machines and focuses on compensating for magnetic field inhomogeneity. The trade-offs for allowing inhomogeneous fields is examined. The equations describing inhomogeneity are derived for both one and two dimensions. Two methods of image compensation are compared, one from the literature and a new, proposed method. Both methods are demonstrated and analyzed on one-dimensional and two-dimensional images. The robustness of each method to uncompensated quantization errors and field mis-measurement is also examined. The computational complexity of each method is also investigated. The appendix of the dissertation covers background on MRI. The classical model of MRI is explained as well as the meaning of the time constants T1, T2, and T2*. Two fundamental pulse sequences (Field Echo and Spin Echo) are explained in detail, along with a description of k-space.

Recent advancements in gate stack engineering has led to the development of aggressively scaled, high mobility, high-k dielectric based NMOSFETs with metal gates. Most of the current literature on the subject also stressed on the need for a high temperature process step to attain the high mobility under minimal change of effective oxide thickness. However, the physical origin of high mobility is not well understood. In this work, fundamental insight into the necessity of the high temperature process step is provided. Novel experimental strategies are developed to understand the impact of interface states and bulk traps separately and exclusively on channel mobility. It is conjectured that the interface states at the SiO2/(100) bulk-Si interface are identical in nature (as far as coupling with the channel electrons is concerned) to those at the high-k/SiO2/(100) bulk-Si interface. Thus, the response of interface states on channel electrons in high-k insulator based NMOSFETs is properly calibrated by a novel thermal desorption of hydrogen experiment on SiO2/(100) bulk-Si NMOSFETs to yield a highly accurate parameterized equation. The value of interface state response parameter determined by the aforementioned experiment is compared with theoretical predictions, and independently determined projections from electrical stress measurements. The impact of transient charging on transport in the channel is investigated. It is conclusively shown that remote charge has minimal impact on mobility in the channel. The role of nitrogen induced fixed oxide charge is studied on a set of Hf-silicate samples. Role of soft optical phonon scattering and the beneficial impact of metal gates on soft optical phonon limited mobility are thoroughly investigated both theoretically and experimentally. Conclusions are drawn on the fundamental limit of mobility attainable in high-k dielectric based NMOSFETs.

The purpose of this research has been to develop an agent based protection and reconfiguration scheme for power electronic building block based (PEBB) DC distribution systems. One of the foremost applications would be in the new zonal DC distribution on naval ships. The research involves the design of an agent based protection scheme which uses the PEBBs for current limiting and circuit breaking purposes. Considerations are given to reduce the system downtime under fault conditions, allow proper coordination and provide backup protection. The research also involves the design of a reconfiguration management scheme based on collaborative agents. The collaboration ensures that the reconfiguration is achieved at a global level, enhancing the system survivability under the conditions of multiple faults and damages. The coordination ensures that only the faulted part of the system is isolated and the reconfiguration makes sure that the power to the healthy part of the system is supplied continuously. The reconfiguration management also performs load shedding if the generation does not meet the load demand of the reconfigured system due to a fault or damage in the generator.

In motor condition monitoring applications, traditional human expert approach for sensor exploitation is not cost-effective. The training requirements for human experts are extensive, and the overall training process is a very time-consuming task. In addition, the performance of human experts has limitations. For human experts, it is difficult to examine all the input-output data from the motor system under varying noise and motor load conditions. With a motor condition monitoring system that can automatically generate rules in the form of interpretable linguistic fuzzy "if-then" rules and membership functions, it would be easier for experts to understand and modify the rule base and also to track the motor condition for maintenance and replacement requirements. In this research, a methodology for fuzzy rule and membership function generation for broken rotor bar detection of squirrel-cage induction motors was developed. The methodology consists of a set of steps that an expert might do for fuzzy rule and membership function design. The methodology is named "H-ROC", since it utilizes histogram analysis with overlapping bins and a weighted cost function based on ROC (Receiver Operating Characteristics) curve analysis. As a second method, an existing fuzzy rule extraction method was extended to broken rotor bar detection problem. The performance and sensitivity analyses of the two methods were conducted.

This dissertation focuses on sample-data modeling for double edge current programmed mode control (DECPM) and its application to high frequency and wide range DC-DC converters. Steady state conditions and subharmonic oscillation issues for DECPM are addressed. By combining the conventional peak and valley current programmed mode control, a sample-data model for DECPM is proposed. A small signal model for DECPM is developed by deriving the modulation gains (Fm) and the sampling gains (He) for DECPM from the proposed sample-data model. The sampling frequency dependence on the duty ratio and a large current loop gain at high frequency for DECPM are emphasized. The analytical results are verified by the simulation. Finally, DECPM is proposed as a method to control the high frequency and wide range DC-DC converters. A 10MHz four switch buck boost converter is implemented with DECPM to verify the viability of its application to high frequency and wide range converters.

The efficiency of body biasing technique is evaluated in 90-nm process technology for regular and low threshold voltage devices. A new leakage monitor circuit for detecting an optimum reverse body bias voltage is designed. The simulation results shows that the monitor circuit accurately tracks the leakage currents within +/-5% of the actual leakage current values. The standby leakage reduction in static CMOS circuits using reverse body biasing is presented. The results indicate that the reverse body biasing is more beneficial for high speed circuits using low threshold voltage devices. For circuits using nominal threshold voltage devices, the efficiency of reverse body biasing decreases due to the presence of gate leakage. Speed improvement in ring oscillator and ripple carry adder using forward body bias is measured. The results show that the forward body biasing is less effective due to the lower body effect parameter. Supply voltage scaling technique for active power reduction is implemented using 180-nm technology. Power savings up to 50% is achieved by scaling the supply voltage as per the operating frequency requirements.

Conventional source/drain junction and contact formation processes can not meet the stringent requirements of future nanoscale complimentary metal oxide silicon (CMOS) technologies. The selective Si_1-xGe_x source/drain technology was proposed in this laboratory as an alternative to conventional junction and contact schemes. The technology is based on selective chemical vapor deposition of in-situ boron or phosphorus doped Si_1-xGe_x in source/drain areas. The fact that the dopant atoms occupy substitutional sites during growth make the high temperature activation anneals unnecessary virtually eliminating dopant diffusion to yield abrupt doping profiles. Furthermore, the smaller band gap of Si_1-xGe_x results in a smaller Schottky barrier height, which can translate into significant reductions in contact resistivity due to the exponential dependence of contact resistivity on barrier height. This study is focused on formation of self-aligned germanosilicide contacts to phosphorous-doped Si_1-xGe_x alloys. The experimental results obtained in this study indicate that self-aligned nickel germanosilicide (NiSi_1-xGe_x) contacts can be formed on Si_1-xGe_x layers at temperatures as low as 350¢XC. Contacts can yield a contact resistivity of 1E-8 ohm-cm² with no sign of germanosilicide induced leakage. However, above a threshold temperature determined by the Ge concentration in the alloy, the NiSi_1-xGe_x/Si_1-xGe_x interface begins to roughen, which affects the junction leakage. For phosphorus doped layers considered in this study, the threshold temperature was around 500¢XC, which is roughly 100¢XC higher than the threshold temperature for NiSi_1-xGe_x contacts formed on boron doped Si_1-xGe_x layers with a Ge percentage of ~ 50%. Nickel and zirconium germanosilicides were also considered as contact candidates but they were found to result in a contact resistivity near 1E-7 ohm-cm².