<p>The main objective of this thesis is to design a PCS system with high capacity. We find that the hybrid DS/FH CDMA method is a promising technology to achieve this goal. Designing such a system and analyzing its capacity are the major contributions of this thesis.</p> <p>CDMA technology potentially offers higher capacity than the other multiple access methods. However, there are serious problems in implementing DS-CDMA to PCS. These are the long acquisition and "near-far" problem. Both these two problems reduce the system performance, and require complicated hardware solutions.</p> <p>The idea of applying hybrid DS/FH CDMA technique to PCS system is initiated for trying to solve the long acquisition problem. We also find that by taking advantage of FH, we can solve the "near-far" problem without increasing system complexity.</p> <p>The capacity of CDMA systems is limited by mutual interference of users, and the interference level is determined by two major factors: the precision of synchronization and the code property. The most important property of the codes is the cross-correlation. The lower the cross correlation among the codes, the higher the capacity will be. Based on this criterion we explore the performance of different codes. In addition, we proposed to use codes generated from the same maximum-length sequence with a unique phase, which offer the lowest cross correlation in the family of non-orthogonal codes.</p> <p>Another detriment to the performance of the system is the imperfection of synchronization due to random access and propagation delays. We propose to implement time slots and deterministic coordinated hopping patterns. Such a measure not only helps to save acquisition time and power control, but also increases the capacity dramatically, by way of establishing synchronous access and avoiding frequency collisions. We show that quasi-synchronous access can be achieved for uplink and completely synchronous access can be achieved for downlink. Our theoretical analysis as well as simulations showed that, by ignoring the adjacent cell interference, system capacity in a single cell can approach its hard limit defined as the ratio of total bandwidth to the message bandwidth for a reasonable bit error rate.</p> <p>A major concern in applying CDMA to PCS is the interference from adjacent cells, since the whole spectrum is reused in all the adjacent cells. Mathematical models for calculating such interference power have been developed, in terms of propagation characteristics, code properties, etc. The results reveal significant impairment on the capacity due to this interference. Voice activity cycle plays an important role now and can triple the capacity in one cell to approach the hard limit N. Without it, the capacity can only be about N/3 in one cell, for an acceptable BER (10ˉ³) and reasonable Eb/N₀. The capacity of this hybrid system has the potential of being much larger than that of TDMA system.</p> / Doctor of Philosophy (PhD)

<p>This thesis contributes to computer oriented techniques for nominal and statistical design of analog circuits. Innovative software concepts aiming at shortening the design time are described. The problem of interfacing circuit CAD tools for design automation is addressed.</p> <p>The concept of response modeling is reviewed. Benefits and limitations of utilizing response modeling in nominal and statistical design are discussed. A multilevel approach to response modeling offering additional CPU time savings is introduced. Challenges associated with employing grid- based simulators with discrete parameters in the design process are addressed.</p> <p>A highly efficient quadratic interpolation technique is reviewed, extended and used for response modeling. Dedicated data base techniques and an effective parallel processing scheme are developed to complement the quadratic interpolation and further increase the CPU time savings.</p> <p>Flexible circuit optimization where a powerful optimization system can interact with various external simulators is addressed. An interface capable of integrating such an optimization system with external and independent simulators is developed. Approaches to automating the process of preparing the external simulator input and capturing its output are described.</p> <p>The novel Space Mapping approach to optimization with CPU intensive simulators is reviewed.</p> <p>The theoretical and software developments are implemented and tested on several advanced and industrially relevant circuit problems. Results of designs utilizing circuit theory-based and electromagnetic simulators are reported.</p> / Doctor of Philosophy (PhD)

<p>Stimulus waveform parameters and stimulation protocols are fundamental to the use of electrical stimulation in medical applications. This thesis presents new simulation and experimental procedures that for the first time can quantify the effects on nerve fiber recruitment patterns of variable stimulus waveform parameters, such as pulse width changes in the stimulation protocol with respect to electrode orientation. The study of the effect of variable electrical stimulus waveform parameters and stimulation protocols is important from the perspective of therapeutic and diagnostic medicine. Variations in the stimulus waveform such as stimulus pulse width have been shown to offer some promise in allowing for selective recruitment of nerve fibers and motor units based on nerve fiber diameter. The degree of selectivity achievable has not however been quantified under any stimulus electrode protocol. Standardization of electrodiagnostic techniques such as motor unit number estimation would also benefit from a quantified study of nerve fiber recruitment patterns under different stimulus pulse width conditions. Changes in the stimulus electrode orientation result in marked changes in the muscle response during conduction velocity tests. This phenomenon has not been investigated in any systematic fashion. In order to quantitatively study these effects, both theoretically and experimentally, a number of tools and techniques had to first be selected or developed that include: (i) a sufficiently realistic model of the myelinated nerve fiber and nerve trunk that, for the first time in electrical nerve excitation studies, incorporates information associated with anatomically consistent fiber diameter distributions; (ii) a more realistic model of the tissue surrounding this nerve trunk that includes electrical anisotropy; (iii) a field simulation technique used to determine the potential fields at the nerve fiber surfaces (with different diameters and electrode distances) resulting from stimulus pulses with different pulse widths; (iv) two novel sets of experiments, the first of which is used to stimulate a nerve trunk in vivo and from the resulting electrical response, determine the diameters of the nerve fibers that have "fired" under conditions of variable stimulus pulse width, the second of which is used to study the effect of electrode orientation on the stimulus response. The techniques outlined above facilitate a quantitative comparison between experimental and simulation results for the stimulus current pulse width studies as opposed to purely qualitative comparisons that have been reported in the literature. A novel instrument design prototype is presented based on the electrode orientation experiment, that can be used to standardize stimulus electrode orientation for multiple EMG (Electromyography) tests performed on a single subject at different times.</p> / Doctor of Philosophy (PhD)

<p>Irregularities in the magnetic circuits of machines may result in spurious voltages that lead to shaft currents through the shaft, bearings, bearing supports and closing through the machine framework. The IEEE Standard Test Procedure for Polyphase Induction Motors and Generators discusses the shaft current and presents a measurement method for recording either the voltage across the ends of the shaft or the current. This thesis discusses an alternative measurement approach and its application to the identification of shaft current in a large induction machine. Possible causes of shaft current are discussed. The magnetomotive force-permeance approach is used in the determination of the possible causes. An eight pole 2611 kW oil ring lubricated induction motor which exhibited signs of bearing damage due to shaft current was used in a series of experiments conducted to evaluate the cause(s) of the bearing damage. The new method of shaft current measurement, known as the Rogowski Coil Method, was applied to this machine. The experiments included oil analysis and microscopic analysis of the bearing. The experiments showed that the main cause of the shaft current is the presence of joints between stator lamination segments. Saturation accentuates the magnitude of shaft current. The Rogowski coil method is shown to be reliable; it does not disturb the shaft current conducting circuit. Furthermore, it requires little maintenance, unlike the ammeter/brush method. Oil ring delivers oil to the bearing. An oil film will develop between the shaft journal and the bearing. The oil film acts as an insulator against shaft current. It is thus important that sufficient quantity of oil is delivered to the bearing and an oil film is developed as quickly as possible to minimize the shaft current. Hence, the performance of the oil ring is studied. The Rogowski method is used to study the performance of the oil ring. The oil ring is determined to slip early with respect to the shaft. As a result, the amount of oil delivered to the bearing is determined to be less than calculated. The proposed method is simple, reliable and less labor intensive than other methods. Patent application for this method is being filed. A method of minimizing shaft current is described following the experimental work to determine the cause(s) of the shaft current. Patent application for this method is being filed. The present on-line monitoring system utilizes the spectral analysis of the stator current to determine stator and rotor faults in alternating current machines. The Rogowski coil method can be used as an alternative on-line monitoring system for such faults.</p> / Doctor of Philosophy (PhD)

<p>This thesis concerns tracking of planar shapes using a single camera view. Two dimensional silhouette projections of planar shapes randomly oriented and positioned and rotational orientation with respect to the camera. This information can then be provided to robots or automated guided vehicles for automated assembly, docking, shape identification and retrieval.</p> <p>Silhouettes are initially encoded in the system as a linked list of vertex points representing changes in the direction of the contour. This contour is employed to get an initial estimate of the shape's position. Corner feature points of the shape are then more accurately located using a fast corner location algorithm. A photogrammetric solution is then applied to the projection points of the corners to determine the three dimensional position of the shape. The photogrammetric technique employed does not require a least squares fit as does the traditional photogrammetric approach. This new approach is contrasted with the traditional method in terms of accuracy and computational complexity.</p> / Doctor of Philosophy (PhD)

<p>Over the past decade, lattices have been increasingly recognized as an important source of codes for the Gaussian channel. The 8-dimensional Gosset lattice has figured prominently in these new developments because it offers an asymptotic coding gain of 3 dB over conventional pulse amplitude modulation and can be soft decision demodulated/decoded with a reasonable amount of speed. In the present work, we revive a little used definition of the Gosset lattice and show that codes derived from this construction exhibit a null at dc in their baseband spectrum. Such codes are useful as line codes for baseband signalling on channels that do not support a dc spectral component or for bandpass transmission where spectral shaping is required to combat intersymbol interference.</p> <p>Previous applications of lattice codes have been aimed primarily at voiceband data communications. This thesis was motivated by the need to develop a decoder that would make applications in the multi-megabit range of data transmission possible. To achieve this goal, a two-stage approach to demodulation was developed. The first stage makes a fast estimate of the transmitted vector and has the ability to declare an erasure when it knows its estimate is unreliable. This initial erasure declaring stage controls the throughput of the demodulator. Because it is far simpler than a maximum likelihood demodulator, greater speed is achieved. The second stage is provided to correct the occasional occurrence of an erasure and maintain the error performance of the lattice. To complete the decoder structure, we outline a method of lexicographic ordering the signal set that leads to a compact set of decoder look-up tables used to obtain a binary message sequence from each demodulated vector. Finally, we evaluate the effects of quantization on the probabilities of erasure and error and give results from a Monte-Carlo simulation undertaken to verify the demodulators performance.</p> / Doctor of Philosophy (PhD)

<p>There are many different approaches to knowledge based control, in this thesis, three different approaches have been taken for study. These are: -Expert control -Fuzzy logic based control -Qualitative reasoning for control and diagnosis In the first approach, the knowledge based systems acts as a supervisor ro a traditional controller, with the ability to advise an operator about degradation in the performance or if needed tune/change the control algorithm used. This approach is most appropriate when the mathematical model of the systemis known. Several new theoretical insights to the problem of expert control is given in this thesis. An implementation architecture, along with the lessons learned from using such architecture are also discussed.</p> <p>Alternatively, the knowledge based system may be placed in the control loop; i.e. the control laws may be embedded in rules that are a part of the knowledge based system. Fuzzy logic controllers fall within this category. Fuzzy logic controllers circumvent the problem of designing a controller based on a detail dynamic model by employing the approach of a human operator to an ill-defined system. A step by step procedure for developing such controllers is given in this thesis. This procedure is applied to solve two problems of increasing complexity in the areas of robotic deburring and mobile robotics. Implementation and test results are also given.</p> <p>The first two knowledge based control techniques have some disadvantages. This is mainly due to the manner in which knowledge is represented, each fact or rule stands on its own as an underivable 'axiom' making it impossible to question the validity or basis of it. Qualitative control system, instead is based on deep knowledge extracted from the understanding of physical interaction between the different components of the plant. This knowledge is used to construct a qualitative model that is used in control operations. In this thesis, an architecture for qualitative control is given and tested on a simple system via simulations. Furthermore, a strategy for expanding qualitative control to provide process diagnostics is also proposed.</p> / Doctor of Philosophy (PhD)

<p>Optoelectronic devices are important devices in optical fiber communications, optical signal processing, and optoelectronic equipment. This thesis develops new models for optoelectronic devices using electromagnetic field simulation techniques. To demonstrate this new technique, two kinds of optoelectronic devices are investigated in this thesis. Vertical cavity surface emitting laseer (VCSEL) is taken as the first research example due to the promising features in the application of fiber communications. A LiTaO₃ electro-optic (E-O) high-speed probe for external electro-optic sampling measurement is the second research example in terms of the need to develop a three-port electrical and optical model for the probe with the coplanar waveguide (CPW) test structure. A new microcavity model for VCSEL is developed in this thesis. The model is based on optical wave equations and implemented with state-space techniques to predict the characteristics of the microcavities. The model can be used to analyze microcavities in both frequency domains in terms of the material parameters and physical parameters of lasers and provides a simple and fast way to optimize cavity length, reflective mirror and material parameters. A three-dimensional discrete time-domain electromagnetic modeling method for microcavities is developed in this thesis. Finite-difference transmission like matrix (FD-TLM) method is modified to involve the distributed optical gain region into the full wave simulation. A simplified microcavity model is developed with the effective mirror derived in this thesis to replace the quarter-wave stacks, and can be used in the full wave modeling of optoelectronic integrated systems having complex inhomogeneities. The near-field and far-field distribution of VCSELs is obtained for the design of interconnections in fiber communication systems. A new dynamic equivalent circuit model for VCSEL is developed in this thesis to add to the equivalent circuit model family of lasers. To describe the laser dynamics, the microcavity model is coupled to the rate equations. A nonlinear resistance is used to represent the optical gain in the cavity. A spontaneous emission noise source is added to the equivalent circuit model to stimulate the noise process throughout the laser operation, which is favored in the system modeling. Stimulation results are compared with the direct rate equation solutions and show that the model is accurate and effective for providing the carrier, photon, optical field, output power, and frequency chirping response of the VCSEL in one simulation. The model enables the laser and its electrical driving circuit to be directly connected and analyzed in a unified manner and can be a powerful computer CAD model to be incorporated into nonlinear circuit modeling software. The model is a useful design tool because it is such a close analogue to the laser device which may be easily modified and enhanced. This thesis addresses the FD-TLM modeling of the LiTaO₃ E-O probe with coplanar waveguides (CPW) test structure with various configurations. A new system transfer function for the probe is defined to sample the signal directly on the sampling point. The full wave modeling of the probe with CPW test structure lays the foundation of the three-port electrical and optical model of LiTaO₃ probe with CPW test structure. To diagnose integrated circuits, it is desirable to measure signal waveforms on internal circuits. LiTaO₃ probe is widely used in E-O sampling due to its high sensitivity and transverse sampling configuration matched to the CPW test structure. Three-port electrical and optical model of the LiTaO₃ probe with CPW test structure is extremely important to the calibration of E-O measurement. With neural network techniques, the field based three-port model is constructed to provide necessary data for the optimum E-O measurement. The three-port has the same accuracy as the full wave modeling, but its characteristics are directly obtained from the neural network weighting parameters rather than the complicated full wave simulation. With this model, it is possible to de-embed both invasiveness and wave-form distortion in the external E-O sampling by giving the sampling configurations, which moves E-O sampling further towards the quantitative measurement.</p> / Doctor of Philosophy (PhD)

<p>Wavelet Packet Division Multiplexing (WPDM) is a multiple signal transmission technique in which the message signals are waveform coded on wavelet packet basis functions for transmission. It has the advantage of a parallel transmission system, but the overlapping nature of such waveforms in time and frequency provides a bandwidth efficiency improvement over the Frequency Division Multiplexing (FDM), whilst their orthogonality properties ensure that the overlapping message signals can be separated by a signal correlation receiver. The interface caused by timing errors in the receiver is examined, and modeled as the information signal sequence filtered by interference filters. A design procedure which exploits the inherent degrees of freedom in the WPDM structure to mitigate the effects of timing error is introduced and a waveform which minimizes the energy of the timing error is introduced and a waveform which minimizes the energy of the timing error interference is designed. An expression for the probability of error due to the presence of Gaussian noise and timing error for the transmission of binary data is derived. The performance advantages of the designed waveform over standard wavelet packet basis functions are demonstrated by both analytical and simulation methods. The performance of WPDM in impulsive noise and Gaussian noise is analyzed, simulated, and compared with the performance of a commonly used serial transmission system, Time Division Multiplexing (TDM). The results support the instinct that WPDM as a parallel transmission system provides greater immunity to impulsive noise than TDM. The performance of WPDM in fading channels is analyzed, and the time domain equalization technique - Pilot Symbol Assisted Modulation (PSAM) - is applied to suppress the error floor which is common in digital communication with coherent reception in fading channels. The derivation and simulation show that in flat fading channels, the error floor in WPDM is suppressed as successfully as FDM with the PSAM, and the frequency of pilot symbols used in WPDM is less than that in FDM to acquire the sufficient channel estimation, thus the WPDM wastes less capacity and energy on pilot symbols than FDM. The analysis of WPDM in two-ray frequency selective fading channels with PSAM is done, and shows that error floor is also suppressed to some extent, but not as striking as for flat fading channels. The reason for this is revealed in the analysis. The bandwidth efficiency improvement of WPDM, its simple implementation, its greater immunity to impulsive noise, and its good performance in fading channels indicate that WPDM holds considerable promise as a multiple signal transmission technique.</p> / Doctor of Philosophy (PhD)

<p>As a result of the "unbundling" of traditional monopolistic electricity generation and transmission enterprises into a free-market economy, a power distribution utilities are faced with very difficult decisions pertaining to electricity supply options and quality of service to the customers. The management of distribution utilities has become increasingly complex, versatile, and dynamic to the extent that conventional, non-automated management tools are almost useless and obsolete. This thesis presents a novel and unified approach to managing electricity supply options and quality of service to customers. The technique formulates the problem in terms of variables, parameters, and constraints. An advanced Mixed Integer Programming (MIP) optimization formulation is developed together with novel, logical, decision-making algorithms. These tools enable the utility management to optimize various cost components and assess their time-trend impacts, taking into account the intangible issues such as customer perception, customer expectation, social pressures, and public response to service deterioration. The abovee concepts are further generalized and a Logical Proportion Analysis (LPA) methodology and associated software have been developed. Solutions using numbers are replaced with solutions using words (character strings) which more closely emulate the human decision-making process and advance the art of decision-making in the power utility environment. Using practical distribution utility operation data and customer surveys, the developments outlined in this thesis are successfully applied to several important utility management problems. These involve the evaluation of alternative electricity supply options, the impact of rate structures on utility business, and the decision of whether to continue to purchase from a main grid or generate locally (partially or totally) by building Non-Utility Generation (NUG).</p> / Doctor of Philosophy (PhD)