From clock signals for digital circuits to frequency references and voltage controlled oscillators for communication systems, oscillators are a fundamental element in many electronic circuits. With the current trend toward smaller and more tightly integrated systems, many applications would benefit greatly from the ability to integrate an oscillator onto a chip. In this thesis, we describe the design and development of a new oscillator technology which makes use of a piezoelectric resonator for frequency control. By using a specially constructed piezoelectric structure, the proposed design exhibits several significant advantages over other currently available technologies including the ability to be manufactured on a silicon substrate. This document presents the design of the tuning and control circuitry for such a resonator as well as experimental results with a macro-scale prototype. Details of a truly single-chip oscillator solution, including the design of on-chip capacitors, switches and an amplifier, are also presented. Tuning of the prototype resonator structure is accomplished electrically and the design has shown frequency tuning of up to 1.5% of the nominal value with a theoretical limit of 40% given the appropriate resonator design.

When designing and implementing energy harvesting devices, additional unanticipated energy may be harvested, dissipated or lost due to interference involving proximate elements on the Printed Circuit Board (PCB) or Integrated Circuit (IC). These components can be PCB or IC traces, planes or discrete circuit components. This alters a designers expectations as to how much energy should be harvested. Empirical analysis can be used to quantify the energy added or subtracted by proximate elements. In this thesis, the Active Remote Sensor (ARS) PCB is modified a number of times to provide enough variations to compare each sections contribution to the overall energy harvested. The differences in harvested energy will provide future designers with a better sense of layout in order to reduce or harness, this effect. Additionally, the number and arrangement of antenna and voltage doublers are modified to see which configuration(s) work most effectively.

On-chip impedance matching has become a major focus as companies and institutions move closer to a complete System on a Chip (SoC). With limited design area, it is important to obtain maximum power transfer to the required load. This research presents commonly used impedance matching techniques and extends them to include on-chip networks. These networks have inherent problems caused by the common substrate. It will be shown that the resulting parasitics can be calculated to allow analysis and manipulation of the overall design. It will also be demonstrated that the use of on-chip inductors will cause severe mismatch and loss due to their low quality factors. Finally, test networks will be fabricated in a 1.5-micron process to show the validity of the concepts presented.

Traffic Congestion is one of the salient issues that affect overall network performance. Network traffic has become very dynamic due to a variety of factors, such as, the number of users varies with time of the day, multimedia applications, bursts in traffic due to a failure and so on. Recently, Multi-Protocol Label Switching (MPLS) networks have emerged as a technology with many promising features such as traffic engineering, QoS provisioning, and speeding up the traffic transmission. However, MPLS still suffers from the nonstationary/transient conditions that sometimes cause congestion. Actually, congestion does not always occur when the network is short capacity, but rather, when the network resources are not efficiently utilized. Thus, it is very important to develop an algorithm that efficiently and dynamically adjusts the available capacity. In this thesis, we propose an adaptive capacity allocation scheme. We have started our consideration with a single traffic class system that has dynamic traffic where traffic arrival is considered at the level of connection/call arrival. We assume that the virtual network for this traffic class operates as a loss system; i.e. if a connection does not find bandwidth, the connection is blocked and cleared from the system. Then, we extended our work to include the multiple traffic classes. Two cases have been studied and analyzed; when classes have no coupling and when they are coupled. The capacity allocation scheme is derived from a first-order, differential equation-based, fluid-flow model that captures the traffic dynamics. The scheme aims to maintain the connection blocking probability within a specified range by dynamically adjusting the allocated capacity. A fluid flow differential equation model is developed to model the changing traffic environment. Using the fluid flow model, Lyapunov Stability theory is used to derive a novel adaptive capacity adjustment scheme which guarantees overall system stability while maintaining the target QoS parameters. Numerical results are given which show that the Lyapunov control based scheme successfully provides the desired QoS requirements and performs better than existing schemes in the literature.

Despite the significant advantages of direct sequence spread spectrum communications, whenever the number of users increases or the received signal is corrupted by an intentional jammer signal, it is necessary to model and estimate the channel effects in order to equalize the received signal, as well as to excise the jamming signals from it. Due to multi-path and Doppler effects in the transmission channels, they are modeled as random, time-varying systems. Considering a wide sense stationary channel during the transmission of a number of bits, a linear time-varying model characterized by a random number of paths, each being characterized by a delay, an attenuation factor and a Doppler frequency shift, is shown to be an appropriate channel model. It is shown that the estimation of the parameters of such models is possible by means of the spreading function, related to the time-varying frequency response of the system and the associated evolutionary kernels. Applying the time-frequency or frequency-frequency discrete evolutionary transforms, we show that a blind estimation procedure is possible by computing the spreading function from the discrete evolutionary transform of the received signal. The estimation also requires the synchronized pseudo-noise sequence for either of the users we are interested in. The estimation procedure requires to adaptively implementing the discrete evolutionary transform to estimate the spreading function and determine the channel parameters. Once the number of paths, delays, Doppler frequencies and attenuations characterizing the channel are found, a decision parameter can be obtained to determine the transmitted bit. We will show also that our estimation approach supports multiuser communication applications such as uplink and downlink in wireless communication transmissions. In the case of an intentional jamming, common in military applications, we consider a receiver based on non-stationary Wiener masking that excises such jammer as well as interference from other users. Both the mask and the optimal estimator are obtained from the discrete evolutionary transformation. The estimated parameters from the computed spreading function, corresponding to the closest to the line of sight signal path, provide an efficient detection scheme. Our procedures are illustrated with simulations, that display the bit-error rate for different levels of channel noise and jammer signals.

In this thesis, DCT domain video compositing with embedded zerotree coding for multi-point video conferencing is considered. In a typical video compositing system, video sequences coming from different sources are composited into one video stream and sent using a single channel to the receiver points. There are mainly three stages of video compositing: decoding of incoming video streams, decimation of video frames, and encoding of the composited video. Conventional spatial domain video compositing requires transformations between the DCT and the spatial domains increasing the complexity of computations. The advantage of the DCT domain video compositing is that the decoding, decimation and encoding remain fully in the DCT domain resulting in faster processing time and better quality of the composited videos. The composited videos are encoded via a DCT based embedded zerotree coder which was originally developed for wavelet coding. An adaptive arithmetic coder is used to encode the symbols obtained from the DCT based zerotree coding resulting in embedded bit stream. By using the embedded zerotree coder the quality of the composited videos is improved when compared to a conventional encoder. An advanced version of zerotree coder is also used to increase the performance of the compositing system. Another improvement is due to the use of local cosine transform to decrease the blocking effect at low bit rates. We also apply the proposed DCT decimation/interpolation for single stream video coding achieving better quality than regular encoding process at low bit rates. The bit rate control problem is easily solved by taking the advantage the embedded property of zerotree coding since the coding control parameter is the bit rate itself. We also achieve the optimum bit rate allocation among the composited frames in a GOP without using subframe layer bit rate allocation, since zerotree coding uses successive approximation quantization allowing DCT coefficients to be encoded in descending significance order.

Recent progress in wireless communication is sustained through integrated circuit technologies that offer a low cost and low power devices that operate in the Radio Frequency (RF) range with relatively low noise figure. The submicrometer CMOS technology presents a serious alternative to the more expensive, high power GaAs and Si bipolar technologies that have been used for the design of high frequency ICs. Design testing and verification through circuit simulation is a critical step in the design cycle of RF integrated circuits (RFICs). Accurate device models are therefore required to reduce design cycles and to achieve success when the circuit is finally committed to silicon. This thesis addresses the Radio Frequency (RF) small-signal and large-signal models for the MOS transistor. The quasi-static (QS) and non-quasi-static (NQS) models are discussed and the assumptions used in their development are examined. The various charge components are briefly introduced and the source/drain charge partitioning is presented. The limitation of the QS approach at high frequency is investigated using the Bsim3v3.1 model. The development of a first order NQS small-signal model is briefly presented and its suitability for RF applications is indicated. The effect of the distributed gate, channel, and substrate resistances on the high frequency characteristics of the MOS transistor is examined. We propose a Radio Frequency small-signal equivalent circuit (EC) together with an efficient parameter extraction algorithm that is necessary for the device optimization and the development of accurate large-signal models. The validity of the proposed model and the accuracy of the extraction method are verified by comparing Pspice simulation results of the EC to experimental data and the Bsim3v3.1 model up to 10GHz.

The task of optimally assigning military ordinance to enemy targets, often termed the Weapon Target Assignment (WTA) problem, has become a major focus of modern military thought. Current formulations of this problem consider the enemy targets as either passive or entirely defensive. As a result, the assignment problem is solved purely as a one sided team optimization problem. In practice, however, especially in environments characterized by the presence of an intelligent adversary, this one sided optimization approach has very limited use. The presence of an adversary often necessitates incorporating its intended actions in the process of solving the weapons assignment problem. In this dissertation, we formulate the weapon target assignment problem in the presence of an intelligent adversary within the framework of game theory. We consider two teams of opposing units simultaneously targeting each other and examine several possible game theoretic solutions of this problem. An issue that arises when searching for any solution is the dimensionality of the search space which quickly becomes overwhelming even for simple problems with a small number of units on each side. To solve this scalability issue, we present a novel algorithm called Unit Level Team Resource Allocation (ULTRA), which is capable of generating approximate solutions by searching within appropriate subspaces of the search space. We evaluate the performance of this algorithm on several realistic simulation scenarios. We also show that this algorithm can be effectively implemented in real-time as an automatic target assigning controller in a dynamic multi-stage problem involving two teams with large number of units in conflict.

We believe that the auditory system, like the visual system, may be sensitive to abrupt stimulus changes and the transient component in speech may be particularly critical to speech perception. If this component can be identified and selectively amplified, improved speech perception in background noise may be possible. This project describes a method to decompose speech into tonal, transient, and residual components. The modified discrete cosine transform (MDCT) and the wavelet transform are transforms used to capture tonal and transient features in speech. The tonal and transient components were identified by using a small number of MDCT and wavelet coefficients, respectively. In previous studies, all of the MDCT and all of the wavelet coefficients were assumed to be independent, and identifications of the significant MDCT and the significant wavelet coefficients were achieved by thresholds. However, an appropriate threshold is not known and the MDCT and the wavelet coefficients show statistical dependencies, described by the clustering and persistence properties. In this work, the hidden Markov chain (HMC) model and the hidden Markov tree (HMT) model were applied to describe the clustering and persistence properties between the MDCT coefficients and between the wavelet coefficients. The MDCT coefficients in each frequency index were modeled as a two-state mixture of two univariate Gaussian distributions. The wavelet coefficients in each scale of each tree were modeled as a two-state mixture of two univariate Gaussian distributions. The initial parameters of Gaussian mixtures were estimated by the greedy EM algorithm. By utilizing the Viterbi and the MAP algorithms used to find the optimal state distribution, the significant MDCT and the significant wavelet coefficients were determined without relying on a threshold. The transient component isolated by our method was selectively amplified and recombined with the original speech to generate enhanced speech, with energy adjusted to equal to the energy of the original speech. The intelligibility of the original and enhanced speech was evaluated in eleven human subjects using the modified rhyme protocol. Word recognition rate results show that the enhanced speech can improve speech intelligibility at low SNR levels (8% at -15 dB, 14% at -20dB, and 18% at -25 dB).

Flexible AC Transmission System (FACTS) controllers, such as the Static Var Compensator (SVC), employ the latest technology of power electronic switching devices in electric power transmission systems to control voltage and powerflow, and improve short-term voltage stability. Given a profit-driven, deregulated electric power industry coupled with increased load growth, the power transmission infrastructure is being stressed to its upper operating limits to achieve maximum economic returns to both generator and transmission system owners. In such an environment, system stability problems such as inadequate voltage control and system instability must be resolved in the most cost-effective manner to improve overall grid security and reliability. Static Var Compensators are being increasingly applied in electric transmission systems to economically improve voltage control and post-disturbance recovery voltages that can lead to system instability. An SVC provides such system improvements and benefits by controlling shunt reactive power sources, both capacitive and inductive, with state-of-the-art power electronic switching devices. This thesis will discuss and demonstrate how SVC has successfully been applied to control transmission systems dynamic performance for system disturbances and effectively regulate system voltage. System and SVC modeling will also be discussed.