As silicon CMOS circuit technology is scaled above the GHz range, semiconductor industries face increasingly difficult challenges in implementing high speed metal interconnects. Metal trances are limited in density-speed performance due to the skin effect, electrical conductivity, and cross talk. Optical based interconnects have much higher available bandwidth by virtue of the extremely high carrier frequencies of optical signals (>100 THz). As more and more optical lightwave circuits and intra-chip optical interconnects are fabricated, a compact, low loss optical tapping technology is essential to incorporate optical interconnects into mainstream CMOS processes. In this thesis, a new optical tapping device, which is based on optical refractive index variation induced by ion-implantation, was studied for both applications of intra-chip optical interconnect and optical power monitoring for planar lightwave circuits. Beam propagation simulation was used to study the guided waveguide properties of ion-implanted optical taps. The length and tapping ratio of optical tap was optimized as functions of implanted ion energy and concentration for 1550nm guided light. The behavior of ion-implanted optical taps can be explained by waveguide coupling between ion-implanted waveguides and as-fabricated waveguides. Ion implantation experiments have also been performed on silica-on-silicon planar lightwave circuit devices using hydrogen ions. Ion concentration up to 5×1016 ion/cm3 were used in experiments with implantation energy from 1.2 MeV to 1.35 MeV. Transmission experimental results are qualitatively consistent with the simulation results. Both simulation and experimental results proves the feasibility of ion-implantation to produce effective and CMOS compatible optical tapping devices for wide applications in chip-level optical interconnect and planar lightwave circuits.

Collections of several hundred, thousands, or even millions of small devices scattered or placed throughout an area monitoring the environment called sensor networks have several useful applications. Until recently, the economic cost of development, manufacture, and deployment limited the use of sensor networks to military and government applications. Recent advances in technology provide a means for economical development, deployment, and manufacture of sensor networks. Current methodology designs, then implements and simulates the sensor network, then goes back and redesigns to better meet the specifications. The model developed in this dissertation provides an early indication of what types of solutions will meet the requirements and what types of solutions will not. With this ability, the time required for simulation and proof of concept is reduced, allowing more time and money for design and testing of the real world system. The model developed characterizes the energy consumption of a sensor or RFID network as a whole is extremely beneficial and is needed. The model provides a means to benchmark different types of sensor networks (i.e. different protocols, hardware, software) and to determine which type is the better solution. A model such as this removes the requirement to develop a simulation to compare different types. Using the model reduces the time (and save money) needed to verify the solution and helps with development as multiple designs can be quickly tested and compared possibly at a much earlier stage in the development cycle allowing a thorough investigation of different design alternatives.

Cyber intrusions have become a serious problem with growing frequency and complexity. Current Intrusion Detection/Prevention Systems (IDS/IPS) are deficient in speed and/or accuracy. Expert systems are one functionally effective IDS/IPS method. However, they are in general computationally intensive and too slow for real time requirements. This poor performance prohibits expert system's applications in gigabit networks. This dissertation describes a novel intrusion prevention expert system architecture that utilizes the parallel search capability of Content Addressable Memory (CAM) to perform intrusion detection at gigabit/second wire speed. A CAM is a parallel search memory that compares all of its entries against input data in parallel. This parallel search is much faster than the serial search operation in Random Access Memory (RAM). The major contribution of this thesis is to accelerate the expert system's performance bottleneck "match" processes using the parallel search power of a CAM, thereby enabling the expert systems for wire speed network IDS/IPS applications. To map an expert system's Match process into a CAM, this research introduces a novel "Contextual Rule" (C-Rule) method that fundamentally changes expert systems' computational structures without changing its functionality for the IDS/IPS problem domain. This "Contextual Rule" method combines expert system rules and current network states into a new type of dynamic rule that exists only under specific network state conditions. This method converts the conventional two-database match process into a one-database search process. Therefore it enables the core functionality of the expert system to be mapped into a CAM and take advantage of its search parallelism. This thesis also introduces the CAM-Assisted Intrusion Prevention Expert System (CAIPES) architecture and shows how it can support the vast majority of the rules in the 1999 Lincoln Lab's DARPA Intrusion Detection Evaluation data set, and rules in the open source IDS "Snort". Supported rules are able to detect single-packet attacks, abusive traffic and packet flooding attacks, sequences of packets attacks, and flooding of sequences attacks. Prototyping and simulation have been performed to demonstrate the detection capability of these four types of attacks. Hardware simulation of an existing CAM shows that the CAIPES architecture enables gigabit/s IDS/IPS.

As the need to support high speed data exchange in modern communication networks grows rapidly, effective and fair sharing of the network resources becomes very important. Todays communication networks typically involve a large number of users that share the same network resources but may have different, and often competing, objectives. Advanced network protocols that are implemented to optimize the performance of such networks typically assume that the users are passive and are willing to accept compromising their own performance for the sake of optimizing the performance of the overall network. However, considering the trend towards more decentralization in the future, it is natural to assume that the users in a large network may take a more active approach and become more interested in optimizing their own individual performances without giving much consideration to the overall performance of the network. A similar situation occurs when the users are members of teams that are sharing the network resources. A user may find itself cooperating with other members of its team which itself is competing with the other teams in the network. Game theory appears to provide the necessary framework and mathematical tools for formulating and analyzing the strategic interactions among users, or teams of users, of such networks. In this thesis, we investigate networks in which users, or teams of users, either compete or cooperate for the same network resources. We considered two important network topologies and used many examples to illustrate the various solution concepts that we have investigated.. First we consider two-node iii parallel link networks with non-cooperative users trying to optimally distribute their flows among the links. For these networks, we established a condition which guarantees the existence and uniqueness of a Nash equilibrium for the link flows. We derived an analytical expression for the Nash equilibrium and investigated its properties in terms of the network parameters and the users preferences. We showed that in a competitive environment users can achieve larger flow rates by properly emphasizing the corresponding term in their utility functions, but that this can only be done at the expense of an increase in the expected delay. Next, we considered a general network structure with multiple links, multiple nodes, and multiple competing users. We proved the existence of a unique Nash equilibrium. We also investigated many of its intuitive properties. We also extended the model to a network where multiple teams of users compete with each other while cooperating within the teams to optimize a team level performance. For this model, we studied the Noninferior Nash solution and compared its results with the standard Nash equilibrium solution.

<div>When a sound wave interacts with an object, the acoustic energy may excite resonances in the object corresponding to its natural modes of vibration. The backscattered wave will then contain information which can be used to distinguish among different objects. As the wave propagates, it can be changed by the propagation channel, which complicates automatic classification of the echo. For example, in a dispersive channel, the duration of the wave increases with propagation distance. Our goal is to identify features of propagating waves that may be used for automatic classification. </div><div><br /></div><div>In this work, we implement and test a class of moment-like features that are invariant to specific propagation effects, in particular dispersion. Our tests of the classification utility of the "dispersion-invariant moments" (DIMS) are performed on numerical models of dispersive propagation and acoustic scattering from steel shells. We consider the case of real dispersion relations and in the conclusion discuss the implementation of complex dispersion and a future direction for research.</div>

Physical phenomena (optical, electronic and optoelectronic) occurring in metallic nanostructures offer an interesting potential in that they may allow us to overcome the limits of diffractive optics and to develop new functional devices complementing the dielectric-based conventional optics. Optical waves incident to a metallic structure, for example, can excite a collective oscillation of electrons, so-called surface plasmons (SPs). The spatial extension of SP fields is governed by the size of the nanostructure and can be made much smaller than the wavelength of light. These features are potentially useful in developing ultracompact photonic chips, i.e., miniaturizing the optics into subwavelength dimensions. In this thesis, we have investigated the plasmonic phenomena occurring in metallic nanoaperture array structures. A novel fabrication process has been developed to form highly ordered nanoaperture (both slits and holes) arrays on metallic layers. Optical characterization of the fabricated nanostructures revealed many interesting properties (in transmission, reflection, filtering, confinement, etc.) involving plasmonic interactions. The plasmonic phenomena in nanoaperture arrays have been analyzed theoretically: analytical solutions of plasmonic waveguiding inside nanoslits were formulated; funneling of light into nanoslit was simulated; the in-plane surface plasmon band structures at the metal/dielectric interfaces were modeled; the dynamic evolution of polarization in metal islands was analyzed. The finite-difference time-domain (FDTD) analysis of optical field distribution and propagation has been performed, and the simulation results were compared with the analytic results and experimental data. Detailed mechanisms of the plasmonic interactions in nanoaperture arrays have been developed and proposed based on this experimental and theoretical study. We further studied optical transmission properties of bi-layer metallic nanoslit arrays. The structure is found to reveal Fabry-Perot-resonator-like characteristics and the transmittance, passband, and beam polarization properties are determined by structure, dimension, and configuration. Near-field interaction and coupling in the bi-layer slit array structures were also analyzed with FDTD simulation. We also studied surface plasmon effects in reflective metallic grating structures, which show strong reflection quenching under cross-metal SP coupling conditions. We have designed and analyzed metallic nano-optic lenses based on nanoslit array structures. The phase of optical radiation emanating from each aperture is controlled by the metal thickness and aperture size. FDTD simulation of the nano-optic lenses demonstrates refractive transmission of light and beam shaping (focusing and collimation). This study opens up the possibility of developing a new class of optics that can complement the conventional dielectric-based refractive/diffractive optics.

There are many possible uses for remotely powered environmental sensing devices. The University of Pittsburgh has obtained a subcontract to assist in the first development phase of one such device for NASA, a wireless temperature sensor that could ultimately be used to measure the temperature of panels on their spacecraft. This thesis describes the work that has been done to completely meet the project specifications set forth in the subcontract, with particular emphasis being given to the contributions made by the author. In addition to the remote sensor board hardware and software, an embedded protocol is developed that can allow hundreds of these devices to transmit their temperature readings over a single communications channel (amplitude shift keying at 418 MHz) without interference or the need for an on-board receiver. Laboratory testing results that verify the proper operation of the final prototype are included.

Parkinsons disease is a neurodegenerative disorder that causes tremor, stiffness, and slowness of movement. The first line of treatment for the disease is the administration of drugs. Over a period of time, these drugs slowly lose their affect to arrest the symptoms associated with Parkinsons disease. Once a patient becomes refractory to drug treatment, one alternative treatment option is Deep Brain Stimulation (DBS). In DBS, a probe is implanted in the basal ganglia area of the brain to administer electric pulses that curb the aforementioned symptoms. Although not fully understood, DBS is becoming a more widely accepted treatment, with various implantable devices currently on the market. These devices, however, require the implantation of a relatively large battery and control pack in the chest with subcutaneous wires threaded up through the neck to the top of the skull. The control pack and wires are a common source of irritation and infection, sometimes necessitating long periods of antibiotics or even removal of the device. Furthermore, the device is susceptible to magnetic interference and has a limited battery life. After the average 3- to 5-year lifespan of an implants battery, another surgery is required to replace the device. The aim of this research is to design a small remotely powered device capable of driving a DBS probe from directly under the scalp. Successful development and proof of viability will form a basis for the conceptual redesign of currently marketed devices in order to eliminate the intrusive battery pack and wires, as well as the health risks commonly associated with them and the implantation procedure.

Radio frequency (RF) energy harvesting is a promising technology that finds applications in such products as Radio Frequency Identification (RFID) and Active Remote Sensing (ARS). In order to reduce the overall size and the manufacturing cost of the device, it is highly desirable to integrate the energy-harvesting antenna, onto the same monolithic CMOS integrated circuit as the functional circuitry. The focus of this dissertation is on the extension of the more traditional approach to antenna design while overcoming the many barriers to the design and analysis of tiny antennas that are fabricated on a CMOS die resulting in an extremely unfriendly environment. Specifically, the major challenges for building antennas on CMOS ICs have been identified. The Finite Element Method (FEM) was found to be the most suitable numerical method for the full-wave analyses of antennas on CMOS ICs after a comparison of the major numerical methods available for electromagnetic simulations. A complete power measurement system that requires no cable connection to the antenna under test has been constructed. It offers accurate measurement of the available power from the on-chip antennas with the help of the annealing approach to impedance matching, which was also developed in this research. The various design factors for antennas on CMOS ICs have been evaluated through both simulation and experiments. It was concluded that the properly designed spiral antennas are good candidates for the on-chip energy-harvesting applications.

A novel complex ultra wideband RF pulse forming technique has been implemented in this research, using the coefficients derived from discrete Fourier transform of a virtual pulse train. Incorporated in this technique is a multiple frequency communication systems designed such that transmitter receiver proximity and the fading effect of the individual frequencies make part of a corresponding modulation technique. A code division multiple access (CDMA) application to RFID to greatly reduce read time, while at the same time eliminating inter tag interference, has been investigated with the analysis of a typical cart aisle scenario. With the current rate of growth of inventory world wide there is a tremendous need for more efficient method of data gathering, data storage, and data retrieval. In this dissertation, the application of the CDMA RFID technology has been analyzed to demonstrate the potentials of integrating the RFID technology to the EPC global numbering system.