Thesis (Ph. D.)--University of Washington, 2001. / Vita. Includes bibliographical references (p. 120-124).

Thesis (Ph. D.)--Rutgers University, 2009. / "Graduate Program in Mathematical Sciences." Includes bibliographical references (p. 74).

Spectral recomposition splits a seismic spectrum into Ricker components. It provides a tool for imaging and mapping temporal bed thicknesses and geologic discontinuities. I propose an application of separable, nonlinear, least-squares estimation in spectral recomposition. Employing the Gauss-Newton method, this approach estimates fundamental signal parameters such as peak frequencies and amplitudes. I applied spectral recomposition to multicomponent seismic data, which provides new perspectives of seismic attributes and multicomponent data interpretation. Correlating S-wave reflection with P -wave reflection is one of the very first steps in multicomponent data interpretation. In a given stratigraphic interval of a geologic section, registration correlates P and S-wave profiles to determine ts /tp ratios, which are equivalent to Vp /Vs ratios for vertical propagation paths. The registration process is largely driven by the availability of dipole sonic logs. However, dipole sonic logs are not as common as standard sonic logs and tend to be affected by various borehole factors. Therefore, new techniques are needed for accurate P P and P S correlation and registration. Assuming P P and P S reflection events have been correctly positioned laterally in migrated images, and the difference between P P wave image and P S wave image can be explained only by vertical transformation, I adopt a multistep approach to register PP and PS images automatically. Setting PP time as a coordinate system, I was able to squeeze P S traces accordingly while keeping the signal pattern of P S wave data. Local seismic attributes, such as the local similarity, help improve registration accuracy. / text

The post-asymptotic giant branch (post-AGB) phase is a short episode in the life of a star with mass between 0.8 to 8 M⊙. It comes after the AGB phase, and before the planetary nebula phase. A rapid change in many physical properties of a star is suggested to happen in this phase, for example the onset of jets. However, a lot of details are still unknown. In this thesis, three major problems are addressed: insufficient samples of post-AGB stars, identification of post-AGB stars, and the true status of a special class of objects called the “water fountains (WFs)”. WFs are evolved stars associated with high velocity collimated bipolar jets that can be traced by H2O maser emissions. For the first two problems, new searching criteria are introduced with two new maser surveys on oxygen-rich post-AGB stars. It is necessary to collect more samples of post-AGB stars for further studies. Nonetheless, there has been no systematic searching method because most of the post-AGB stars are dim in optical and near-infrared wavelengths, which increases the difficulty in identification. Maser thus becomes a good alternative tool. In the first survey which focused only on H2O masers, over 200 AGB or post-AGB star candidates have been selected and observed. Those candidates were mainly chosen by new colour criteria with the far-infrared AKARI data. In particular, four characteristic maser sources were found, and they are currently suggested as possible very young post-AGB stars. In the second survey, another 100 objects were observed in OH and/or H2O masers. Three possible high velocity objects were discovered, including a new rare member of WFs. The colour criteria are proved to be quite sensitive in distinguishing post-AGB stars from AGB stars or other types of objects, even though there are still some contamination from young stellar objects. A follow-up study shows that the Q-parameters are effective in isolating objects with spherical or aspherical envelopes, which are also useful in finding post-AGB stars. Regarding the third problem, one-dimensional radiative transfer models have been used to study the spectral energy distributions (SEDs) of all known WFs. Owing to the young dynamical ages of the WF jets, it has been suggested that WFs represent the very early post-AGB phase, when the circumstellar envelopes started to depart from spherical symmetry. However, it is shown from the present SED study that WFs in fact could have various form of morphologies, some of them are even spherical which resemble the AGB stars. It implies that WFs do not necessary belong to the very early post-AGB phase, and the present interpretation on the WF status may not be entirely correct. / published_or_final_version / Physics / Doctoral / Doctor of Philosophy

ITC/USA 2013 Conference Proceedings / The Forty-Ninth Annual International Telemetering Conference and Technical Exhibition / October 21-24, 2013 / Bally's Hotel & Convention Center, Las Vegas, NV / The AFSSI-C is a spectral imager that generates spectral classification directly, in fewer measurements than are required by traditional systems that measure the spectral datacube (which is later interpreted to make material classification). By utilizing adaptive features to constantly update conditional probabilities for the different hypotheses, the AFSSI-C avoids the overhead of directly measuring every element in the spectral datacube. The system architecture, feature design methodology, simulation results, and preliminary experimental results are given.

This paper introduces a new 3-D quadratic autonomous system, which can generate two coexisting single-wing chaotic attractors and a pair of diagonal double-wing chaotic attractors. More importantly, the system can generate a fourwing chaotic attractor with very complicated topological structures over a large range of parameters. Some basic dynamical behaviors and the compound structure of the new 3-D system are investigated. Detailed bifurcation analysis illustrates the evolution processes of the system among two coexisting sinks, two coexisting periodic orbits, two coexisting single-wing chaotic attractors, major and minor diagonal double-wing chaotic attractors, and a four-wing chaotic attractor. Poincare´-map analysis shows that the system has extremely rich dynamics. The physical existence of the fourwing chaotic attractor is verified by an electronic circuit. Finally, spectral analysis shows that the system has an extremely broad frequency bandwidth, which is very desirable for engineering applications such as secure communications.

No description available.

This document provides a short history and analysis of Gerard Grisey's Anubis et Nout, written for Harry Sparnaay in 1983 and dedicated to Claude Vivier, a friend and colleague of Grisey's who was murdered that year. Although Anubis et Nout is originally for solo contrabass clarinet, Grisey also created a version in 1990 for Claude Delangle, to be played on either baritone or bass saxophone. Both versions are published by Ricordi.In the first two chapters, I present a brief introduction to the life and music of Gerard Grisey, as well as an introduction to Claude Vivier. In chapter three, I give a chronology of Anubis et Nout; the history presented here is largely informed by interviews conducted by the author with instrumentalists Jean-Noel Crocq, Claude Delangle, Harry Sparnaay, and Taimur Sullivan. The remainder of the document is an original analysis of each movement of Anubis et Nout. The analysis includes discussion and examples of subharmonicity, harmonicity, time concepts presented in Grisey's essay "Tempus ex Machina" , and Grisey's attention to timbre through diacriticals as they pertain to the work.

No description available.

Spreading activation is a common method for searching semantic or neural networks, it iteratively propagates activation for one or more sources through a network { a process that is computationally intensive. Spectral association is a recent technique to approximate spreading activation in one go, and therefore provides very fast computation of activation levels. In this paper we evaluate the characteristics of spectral association as replacement for classic spreading activation in the domain of ontology learning. The evaluation focuses on run-time performance measures of our implementation of both methods for various network sizes. Furthermore, we investigate differences in output, i.e. the resulting ontologies, between spreading activation and spectral association. The experiments confirm an excessive speedup in the computation of activation levels, and also a fast calculation of the spectral association operator if using a variant we called brute force. The paper concludes with pros and cons and usage recommendations for the methods. (authors' abstract)