<p>Impulse radar possess some attractive features that make it particularly useful in probing objects that are buried, encased in other materials or structures. Testing has demonstrated its potential in the detection of deterioration in concrete bridge deck slabs that are covered with bituminous surfacing. In order to benefit fully from using impulse radar in the bridge deck inspection, however, it is necessary to take advantage of the progress in signal processing techniques. This thesis is an attempt to provide a comprehensive treatment of the optimal extraction of information from the reflected radar signals, as to determine the subsurface structure and condition of the bridge decks. Generally, it follows a statistical estimation approach to the problem.</p> <p>In this thesis, a parametric representation is derived ti approximate the radar-generated pulses for probing. The asphalt-covered bridge decks are regarded as a system stratified, lossless, and horizontally layered media with each layer being homogeneous and isotropic. The propagation of electromagnetic waves in such a multi-layered media system can be completely determined by a set of characteristic parameters of the media. Under such assumptions, the reflected radar signals may be well described by a delayed sum model which is specified by the characteristic parameters of the media.</p> <p>Based upon the parametric signal model, a maximum likelihood estimator is formulated to determine the parameters of reflected signals. Computer experiments show that the ML estimation is capable of resolving closely spaced returns in the received signal and producing very accurate parameter estimates. ML estimation of real radar signals reflected from a bridge deck is also performed with success. However, to carry out the ML estimation requires an explicit knowledge of the probability density function of received signal which may not always be available. Moreover, the search for the ML estimates usually involves a global, nonlinar optimization procedure which can be extremely costly in computation time.</p> <p>To overcome the difficulties with the ML method, a new eigenstructure-based (EB) method for parameter estimation is developed in this thesis. The implementation of the new estimation method requires only the autocovariance of the reflected signal, and it is more efficient in computation than the ML method. Computer simulation demonstrates that the EB method results in very satisfactory estimates at high SNR levels, but it becomes inaccurate when the SNR level is low or the radar returns in the recieved signal are very close spaced in time.</p> <p>The error performances of the parameter estimators under varous conditions are evaluated and compared via computer simulation. A detailed analysis of the Cramer-Rao Lower Bound on the estimation error is performed to gain an insight into how various factors affect the estimation performance.</p> <p>An alternative to the parameter estimation approach is predictive deconvolution which is developed on a nonparametric model of the reflected signals. In principle, it is an application of Wiener optimal filtering theory to the deconvolution problem. It is observed in Computer simulation that predicative deconvolution is able to resolve returns closely located in time. Its implementation is carried out by simply solving a set of linear, normal equations, and its operation involves only straightforward linear filtering which demands little computation. However, the performance of predictive deconvolution deteriorates quickly in the presence of even a moderate level of noise in the input signal. This weakness may severely restrict its usefulness in practice.</p> / Doctor of Philosophy (PhD)

<p>Future photonic data communication networks may be based on emerging wavelength multiplexing (WDM) technology. In such networks, stations communicate using some combination of wavelength agile transmission or reception. Unfortunately, the multiple channel environment tends to increase station optical hardware requirements, and complicate the station media access protocols. In this thesis, a number of new network architectures are presented which use either a fully broadcast and select typology, or a partial broadcast and select typology with wavelength routing. The design objective is the creation of multichannel networks and multichannel single hop packet switched media access protocols, which require a minimal amount of station optical hardware, and minimal media access protocol complexity. This is accomplished through the use of channel controllers which assist in the operation of the network. Channel controllers are able to reduce the volume of network state information that a station is required to track and process, assist stations with their media access protocol, and provide synchronization information. The fully broadcast and select single passive star network, and on of the partial broadcast and select networks require a station to have a single fast discrete tunable transmitter and a single fixed or slow tunable receiver, while two other partially broadcast and select networks require a station to be configured with an additional fixed or slow tunable receiver.</p> / Doctor of Philosophy (PhD)

<p>The impact of the array geometry for linear arrays on the estimation of the direction of arrival of incident plane waves is examined in this thesis. The fundamental result is the establishment of the conditions under which array structures, different from uniformly spaced, may provide improved accuracy or reliability in these estimates. We are primarily concerned with the use of high accuracy estimators, attempting to obtain accuracies well within the classical beamwidth of the array.</p> <p>Several different criteria for designing thinned array structures are described, with the principal emphasis on redundancy based designs. For a single plane wave incident on the array, the Weiss-Weinstein bound, which is an estimator independent bound, is applied to a variety of array structures and indicates that the thinned arrays will yield greater accuracy, provided the SNR is sufficiently large. The bound allows us to investigate the effects of a priori information on the estimation performance of the different structures considered.</p> <p>Maximum likelihood estimation is applied to the identical problem and similar tradeoffs are observed. We propose the concept of outlier probability as a measure for discriminating between array structures and provide models for characterization. Certain algorithms in the literature for the design of nonuniform arrays are shown to be poor under these measures.</p> <p>The impact of array structure in a multipath environment consisting a target with a strong specular reflection is illustrated using exact maximum likelihood estimation. It is shown that the most significant gains to be made for thinned arrays occur when the multipath is such that target and image are well within a beamwidth of the array. Under these conditions, it is found that the nonuniform arrays often outperform uniform arrays consisting of many more elements for all values of SNR and phase differences between the two plane waves.</p> <p>An experimental, 32-element array was constructed and brought into the field in order to gather multipath data over water, in a real world environment. For the very closely spaced target and image of this experiment, the nonuniform arrays outperform uniform arrays consisting of even twice as many elements.</p> / Doctor of Philosophy (PhD)

<p>The magnetomyogram (MMG) is a magnetic signal that may be measured external to an active skeletal muscle. It is generated by the transmembrane ion currents associated with the propagation of action potentials along the plasma membranes of the muscle fibres. Compared with other biomagnetic signals, such as the magnetocardiogram and the magnetoencephalogram, the MMG has been studied very little.</p> <p>Presented in this dissertation is the development of a strategy for investigating the potential of the MMG as a research and clinical tool. Investigation of the compound magnetic action flux (MAF), the basic component of the MMG signal, was chosen in favour of direct study of the MMG. The design of a data acquisition system and experimental procedure for investigation of the MAF is also reported. A preliminary set of measurements from an excised frog gastrocnemius under controlled physiological conditions is presented. These measurements were designed to reveal the temporal and spatial characteristics of the MAF and to demonstrate the effect of several physiological variables on these characteristics. Such measurements have never before been reported.</p> <p>The experimental measurements showed that the MAF from a frog gastrocnemius is a biphasic signal with an amplitude of 30 pT and a duration of 4.2 ms. The MAF was found to occur at the same time as the whole muscle action potential and well before the mechanical response of the muscle (generation of twitch force). The MAF was found to have an azimuthal component and a component that looped from end-to-end of the muscle. For submaximal stimulation, the MAF amplitude was found to increase nonlinearly with the twitch force amplitude. For supramaximal stimulation, the MAF amplitude was found to decrease with increasing muscle length. Finally, it was found that the electrical conductivity of the bath medium strongly affected the amplitude of the MAF signal.</p> <p>This research has demonstrated the feasibility of performing MAF measurements in a typically "noisy" laboratory with a commercially available SQUID magnetometer. It has also shown that the MAF signal is reproducible and sensitive to some physiological variables. With these results, a greater knowledge and better understanding of the MAF has developed. Further evaluation of the potential of the MMG as a research and diagnostic tool is justified.</p> / Doctor of Philosophy (PhD)

<p>The material presented in this thesis is a logical extension of and addition to previous work on network sensitivities as applied to power system analysis and planning. The continuing tendency of supplementing the existing extra-high voltage a.c. transmission systems with high-voltage d.c.(HVDC) lines has been taken into consideration, and various relevant component models have been investigated using a new hybrid network formulation based on the methedology developed by Bandler and EI-Kady. The load buses, frequently modelled as PQ-buses at which both the real and reactive injected powers are known, and the generator buses characterized by a constant voltage magnitude and constant real injected power, have been dealt with by exploiting a special complex conjugate notation. In addition, the current, voltage and/or power relationships associated with the transmission network branches have been investigated. A hybrid formulation for generalized power system component models has been developed. This novel formulation not only encompasses the work established on the basis of one-port theory, but it is also capable of manipulating multiport, nonreciprocal, a.c. as well as integrated a.c.d.c. bulk transmission networks. The attractive features of adjoint modeling have been retained, and consequently, exact sensitivity formulas associated with various control variabies have been derived, tabulated and verified. Applications of the novel sensitivity formulation to both, the HVDC link and phase-shifting transformer modeling, are presented. In the first application, a two-port model of an HVDC link connecting two a.c. networks has been used. The terminal relations of the converters have been utilized ingeniously to develop an adjoint converter model. Both firing angle and commutation reactance have been considered as the control variables of interest, and their respective sensitivity formulas have been numerically verified on a test power system. In the second application, a cascaded phase-shifting transformer model comprising an ideal transformer in series with a transformer equivalent impedance has been considered. Exact sensitivity formulas for the control variables representing transformer turns ratio magnitude, phase angle, equivalent resistance and reactance have been derived elegantly and compactly.</p> <p>The theoretical results achieved have been extensively verified using a 2-bus and a 6-bus sample power systems. The functions encountered in steady-state security assessment have been considered, and investigated in applications to two practical IEEE (30-bus and 118-bus) test systems.</p> / Doctor of Philosophy (PhD)

<p>Knowledge of the origin and characteristics of the intestinal pacemaker activity, and the characteristics of intercellular communication throughout the musculature is instrumental for the understanding of the mechanisms through which gastrointestinal (GI) motility is regulated. This thesis makes a significant contribution to provide electrophysiological and morphological evidence supporting the hypothesis that interstitial cells of Cajal (ICCs) are the Gl pacemaker cells. The pacemaker activity of the GI tract triggers the slow-wave.type action potentials (slow waves) which are coherent with the phasic contractions for facilitating peristaltic movement. Origins of the slow waves at different portions of the GJ tract always coincide with the locations of the ICCs. The objectives of this study were to identify the cellular origin of the pacemaker activity and to thoroughly investigate the mechanism of intercellular communication in the canine colon using electrophysiological and microscopic techniques. The cellular origin of the pacemaker activity was further examined by studying simultaneously the ontogenesis of the pacemaker activity and the ICCs in the neonatal mouse small intestine.</p> <p>The cellular origin of the pacemaker activity was studied by employing the photodynamic property of methylene blue (Chapter 3). We previously demonstrated that the submuscular ICCs of the canine colon selectively accumulate methylene blue. In this undertaken study, we further illustrated that incubation with 50 μM methylene blue and subsequent intense illumination resulted in abolition of the pacemaker activity. Following methylene blue incubation, intense illumination first changed the mitochondrial conformation in the ICCs from very condensed to orthodox, and progressively imposed more severe damages, such as swollen and ruptured mitochondria, loss of cytoplasmic contrast and detail, and rupture of the plasma membrane. No damage was seen in smooth muscle cells and nerves. The correlation between selective lesioning of ICCs and loss of the pacemaker activity strongly supports that ICCs play an essential role in the generation of the pacemaker activity.</p> <p>The regulatory mechanism of the pacemaker frequency was investigated with a focus on the effects of cyclopiazonic acid (CPA), a specific inhibitor of the endoplasmic reticulum (ER) Ca²⁺-pump (Chapter 4). CPA dose dependently decreased the pacemaker frequency. Similarly, chelating cytosolic Ca²⁺ with BAPTA also decreased the pacemaker frequency. The pacemaker frequency was also decreased by neomycin (inhibiting inositol 1,4,5-triphosphate (IP₃) synthesis) and caffeine (inhibiting the IP₃-sensitive Ca²⁺ channels in the ER membrane). Electron microscopy showed that the smooth ER forms an extensive network of subsurface cisternae which is closely associated with large areas of the cytoplasmic face of the plasma membrane. These structures were the most extensive in the ICCs, slightly less in branching smooth muscle cells and far less in circular muscle cells. Based on the electrophysiological and morphological observations, we hypothesize that the Ca²⁺ refilling cycle of the IP3-sensitive calcium stores associated with the plasma membrane, determines the frequency of the pacemaker activity generated by the submuscular ICC-smooth-muscle network of the canine colon.</p> <p>The ontogenesis of the pacemaker activity and the ICCs in the small intestine of neonatal mice was studied to further substantiate the pacemaker role of ICCs (Chapter 5). The pacemaker component of the slow waves was fingerprinted by its resistance to L-type Ca²⁺-channel blockers and sensitivity to cyclopiazonic acid, CPA, (Chapter 4). All isolated musculature of the neonatal mouse small intestine (newborn, unfed-7 days old) spontaneously generated action potentials. The presence of the pacemaker component in different age groups was examined by verapamil, a L-type Ca²⁺ channel blocker, and CPA. In conclusion, electrophysiological and morphological evidences were obtained to demonstrate that both the pacemaker activity and the, ICC network were immature at birth but fully developed in 2 days old neonatal mice.</p> <p>Communication between the longitudinal and the circular muscle layers are essential for producing co-ordinated motility in the musculature. Through electrophysiological measurements with microelectrodes, the study of neurobiotin spread using confocal microscopy and the investigation of the cellular structure at the electron-microscopic level, the cellular mechanisms of communication between the two muscle layers was studied. We positively demonstrated the existence of low-resistance pathways. We also provided evidence that the ICCs associated with the myenteric plexus facilitated electrotonic coupling between the two muscle layers across the myenteric plexus of the canine colon (Chapter 7).</p> <p>In the longitudinal muscle layer, no positive evidence for electrical coupling between smooth muscle cells has yet been presented. We thoroughly examined the properties of electrical coupling in the longitudinal muscle layer of the canine colon (Chapter 8). The properties of electrical coupling between longitudinal muscle cells were compared with that between the circular muscle cells. Three electrical coupling parameters were measured: (i) the input resistance, (ii) the space constant (determined by the method developed with a double-electrode technique (Chapter 6)), and (iii) the phase relationship of simultaneously recorded electrical activities. Furthermore, a detailed electron microscopic investigation revealed the absence of gap junctions in the longitudinal muscle layer; whereas, numerous close apposition contacts were observed. These observations put forward the hypothesis that the pathways for electrical coupling between longitudinal muscle cells are consituted by close apposition contacts.</p> <p>Communication between circular muscle (CM) lamellae is necessary to generate propulsive phasic contractions for facilitating peristalsis along the longitudinal axis of the GI tract. The submuscular ICCs are extensively coupled to the underlying branching smooth muscle (bSM) cells forming an ICC-bSM network covering the entire submucosal surface of the canine colon. There is another ICC network located in the myenteric plexus. The roles of the submuscular ICC-bSM network, the myenteric ICC network and the longitudinal muscle layer in mediating communication across the CM lamellae were studied by simultaneous recordings with surface electrodes using different types of muscle strip preparations (Chapter 9). Electrophysiological evidence demonstrated that, within the pure circular musculature, circular muscle cells were electrically coupled along a CM lamella, oriented circumferentially around the lumen, but electrically insulated across CM lamellae. The submuscular ICC-bSM network, but not the longitudinal muscle nor the myenteric plexus, was shown to be essential for mediating communication between CM lamellae such that co-ordinated motility can be exhibited with neig,hbouring CM lamellae through excitation-contraction coupling.</p> <p>In summary, employing a number of electrophysiological and microscopic techniques, this dissertation presents novel evidence (i) to substantiate the pacemaker role of ICCs in the GI tract; (ii) to put forward a hypothesis that the pacemaker-frequency regulatory mechanism is synchronized with the ER Ca²⁺ refilling cycle; and, (iii) on the heterogeneity of mechanisms through which intercellular communication occurs along the radial, circumferential and longitudinal axes of the intestinal musculature.</p> / Doctor of Philosophy (PhD)

<p>A new method for the detection of signals in "noise", which is based on the premise that the "noise" is chaotic with at least one positive Lyapunov exponent, is presented. The method is naturally rooted in nonlinear dynamical systems, and relies on neural networks for its implementation.</p> <p>We first present a theoretical basis for methods of modeling the underlying dynamics of a chaotic system using a time series. The subject matter selected for this part of the thesis is written with emphasis on experimental studies of chaos. Specifically, we discuss the issues involved in the reconstruction of chaotic dynamics, attractor dimensions, and Lyapunov exponents. We describe procedures for the estimation of the correlation dimension and the Lyapunov exponents. The method of false nearest neighbors analysis fur finding the minimum embedding dimension is described. The need for an adequate data length is stressed.</p> <p>In the second part of the thesis we apply the chaos-based method to the radar detection of a small target in sea clutter. To justify the applicability of the new method to this problem, we clearly need to show in a convincing way that sea clutter is indeed the result of a chaotic dynamical system. We do this by presenting the results of a detailed experimental study using surface-truthed real-life data collected by means of an instrument quality radar at different geographic locations. Specifically, we show that (1) sea clutter has a finite correlation dimension, (2) the largest Lyapunov exponent of sea clutter is positive, and (3) sea clutter is locally predictable. Most importantly, we show that both the correlation dimension and the largest Lyapunov exponent are essentially invariant to the choice of radar signal component used to construct the time series, and that the correlation dimension and Lyapunov exponent do not appear to change appreciably with sea state or with geographic locations. These results suggest that there may exist universal chaotic structure responsible for the generation of sea clutter. Perhaps the most dramatic result presented in the thesis is the fact that this prior information (i.e., the knowledge that sea clutter exhibits chaotic behavior) can be exploited to build a chaos-based detettor operating on amplitude information only (as in a noncoherent marine radar), realizing a performance comparable to that of a "conventional" receiver using coherent radar data (i.e., both amplitude and phase). This result points to the potential of trading off sophisticated but inexpensive computer software for expensive microwave hardware. Lastly, we show experimentally that a chaos-based coherent detector can provide a further improvement in radar detection performance.</p> / Doctor of Philosophy (PhD)

<p>Comparison of sequences and trees is an essential problem in computational biology. In this thesis, we investigate some algorithmic problems in sequence and tree comparison. Theoretical issues such as computational complexity, the efficiency of algorithms, and the performance of approximation algorithms are stressed.</p> <p>The most popular approach for comparing a set of sequences is multiple sequence alignment. We show that two popular variants of multiple sequence alignment, multiple alignment with SP-score and tree alignment, are NP-complete. We also design a polynomial time approximation scheme (PTAS) for tree alignment, which is believed to be the first PTAS in computational biology.</p> <p>Tree comparison has applications in the study of RNA secondary structures and evolutionary trees. We propose the notion of alignment of trees as a new method for comparing RNA secondary structure trees and give an efficient algorithm for computing the optimal alignment of such trees. Several other methods for comparing evolutionary trees are also considered.</p> <p>Steiner trees have been studied extensively in the literature, and are closely related to tree alignment. We also design a PTAS for some planar Steiner tree problems.</p> / Doctor of Philosophy (PhD)

<p>This thesis focuses on the joint detection of the model order and the estimation of the parameters of interest, with applications to array signal processing in both the off-line and on-line contexts. In the off-line mode, Markov Chain Monte Carlo methods are applied to obtain a numerical approximation of the joint posterior distribution of the parameters. The on-line approach uses the sequential implementation of Monte Carlo methods applied to probabilistic dynamic systems. Three problems were addressed in the course of this thesis. (1) A method for joint detection of the number of sources and estimation of their respective directions of arrival in coloured noise with unknown arbitrary covariance was developed. (2) The second algorithm represents an extension of the first one with the addition of the joint estimation of the times of arrival of the pulses, in the spirit of channel sounding for characterization of multipath channels. Both methods were successfully applied to real data, acquired on campus with a channel sounder during an extensive measurement campaign. (3) The final part of this thesis focuses on the sequential implementation of the Monte Carlo methods, i.e. particle filters, for probabilistic dynamic systems. The algorithm recursively estimates the posterior distribution of the evolving parameters of interest, allowing for the on-line detection of the number of sources and the estimation of their respective directions of arrival.</p> / Doctor of Philosophy (PhD)

<p>This thesis contributes to computer-aided design and modeling of microwave circuits exploiting space mapping technology. Comprehensive frameworks for enhancing available empirical models or creating new ones are presented. A novel technique for microwave circuit design is also presented. A comprehensive framework to engineering device modeling which we call Generalized Space Mapping (GSM) is introduced. GSM aims at significantly enhancing the accuracy of available empirical models of microwave devices by utilizing a few relevant full-wave EM simulations. Three fundamental illustrations are presented: a basic Space Mapping Super Model (SMSM), Frequency-Space Mapping Super Model (FSMSM) and Multiple Space Mapping (MSM). Two variations of MSM are presented: MSM for Device Responses (MSMDR) and MSM for Frequency Intervals (MSMFI). A novel criterion to discriminate between coarse models of the same device is also presented. A new computer-aided modeling methodology to develop broadband physics-based models for passive components is presented. Full-wave EM simulators, artificial neural networks, multivariable rational functions, dimensional analysis and frequency mapping are coherently integrated to establish broadband models. Frequency mapping is used to develop the frequency-dependent empirical models. Useful properties of the frequency mapping are utilized in the modeling process. Transformations from frequency-dependent models to frequency-independent ones are also considered. The passivity conditions of the frequency-dependent empirical model are also considered. We present a novel design framework for microwave circuits. We expand the original space mapping technique by allowing preassigned parameters (which are not used in optimization) to change in some components of the coarse model. We refer to those components as "relevant" components and we present a method based on sensitivity analysis to identify them. As a result, the coarse model can be calibrated to align with the fine model. Our algorithm establishes a mapping from some of the optimizable parameters to the preassigned parameters of the relevant components. This mapping is updated iteratively until we reach the optimal solution.</p> / Doctor of Philosophy (PhD)