Refine
Year of publication
Document Type
- Preprint (1037)
- Doctoral Thesis (939)
- Article (609)
- Report (399)
- Master's Thesis (30)
- Conference Proceeding (28)
- Diploma Thesis (24)
- Periodical Part (21)
- Working Paper (15)
- Lecture (11)
Language
- English (3149) (remove)
Keywords
- AG-RESY (47)
- PARO (25)
- Visualisierung (16)
- SKALP (15)
- Wavelet (13)
- finite element method (12)
- Case-Based Reasoning (11)
- Inverses Problem (11)
- Optimization (11)
- RODEO (11)
Faculty / Organisational entity
- Kaiserslautern - Fachbereich Mathematik (1052)
- Kaiserslautern - Fachbereich Informatik (753)
- Kaiserslautern - Fachbereich Maschinenbau und Verfahrenstechnik (302)
- Kaiserslautern - Fachbereich Physik (293)
- Fraunhofer (ITWM) (205)
- Kaiserslautern - Fachbereich Chemie (116)
- Kaiserslautern - Fachbereich Elektrotechnik und Informationstechnik (115)
- Kaiserslautern - Fachbereich Biologie (98)
- Kaiserslautern - Fachbereich Sozialwissenschaften (76)
- Kaiserslautern - Fachbereich Wirtschaftswissenschaften (36)
We consider data generating mechanisms which can be represented as mixtures of finitely many regression or autoregression models. We propose nonparametric estimators for the functions characterizing the various mixture components based on a local quasi maximum likelihood approach and prove their consistency. We present an EM algorithm for calculating the estimates numerically which is mainly based on iteratively applying common local smoothers and discuss its convergence properties.
The goal of a multicriteria program is to explore different possibilities and their respective compromises which adequately represent the nondominated set. An exact description will in most cases fail because the number of efficient solutions is either too large or even infinite. We approximate the nondominated by computing a finite collection of nondominated points. Different ideas have been applied, including nonnegative weighted scalarization, Tchebycheff weighted scalarization, block norms and epsilon-constraints. Block norms are the building blocks for the inner and outer approximation algorithms proposed by Klamroth. We review these algorithms and propose three different variants. However, block norm based algorithms require to solve a sequence of subproblems, the number of subproblems becomes relatively high for six criteria and even intractable for real applications with nine criteria. Thus, we use bilevel linear programming to derive an approximation algorithm. We finally analyze and compare the approximation quality, running time and numerical convergence of the proposed methods.
Interactive visualization of large structured and unstructured data sets is a permanent challenge for scientific visualization. Large data sets are for example created by magnetic resonance imaging (MRI), computed tomography (CT), Computational fluid dynamics (CFD) finite element method (FEM), and computer aided design (CAD). For visualizing those data sets not only accelerated rasterization by means of using specialized hardware i.e. graphics cards is of interest, but also ray casting, as it is perfectly suited for scientific visualization. Ray casting does not only support many rendering modes (e.g., opaque rendering, semi transparent rendering, iso surface rendering, maximum intensity projection, x-ray, absorption emitter model, ...) for which it allows the creation of high quality images, but it also supports many primitives (e.g., not only triangles but also spheres, curved iso surfaces, NURBS, implicit functions, ...). It furthermore scales basically linear to the amount of processor cores used and - this makes it highly interesting for the visualization of large data sets - it scales for static scenes sublinear to data size. Interactive ray casting is currently not widely used within the scientifc visualization community. This is mainly based on historical reasons, as just a few years ago no applicable interactive ray casters for commodity hardware did exist. Interactive scientific visualization has only been possible by using graphics cards or specialized and/or expensive hardware. The goal of this work is to broaden the possibilies for interactive scientific visualization, by showing that interactive CPU based ray casting is today feasible on commodity hardware and that it may efficiently be used together with GPU based rasterization. In this thesis it is first shown that interactive CPU based ray casters may efficiently be integrated into already existing OpenGL frameworks. This is achieved through an OpenGL friendly interface that supports multiple threads and single instruction multiple data (SIMD) operations. For the visualization of rectilinear (and not necessarily cartesian) grids are new implicit kd-trees introduced. They have fast construction times, low memory requirements, and allow ontoday's commodity desktop machines interactive iso surface ray tracing and maximum intensity projection of large scalar fields. A new interactive SIMD ray tracing technique for large tetrahedral meshes is introduced. It is very portable and general and is therefore suited for portation upon different (future) hardware and for usage upon several applications. The thesis ends with a real life commercial application which shows that CPU-based ray casting has already reached the state where it may outperform GPU-based rasterization for scientific visualization.
Today’s high-resolution digital images and videos require large amounts of storage space and transmission bandwidth. To cope with this, compression methods are necessary that reduce the required space while at the same time minimize visual artifacts. We propose a compression method based on a piecewise linear color interpolation induced by a triangulation of the image domain. We present methods to speed up significantly the optimization process for finding the triangulation. Furthermore, we extend the method to digital videos. Laser scanners to capture the surface of three-dimensional objects are widely used in industry nowadays, e.g., for reverse engineering or quality measurement. Hand-held scanning devices have the advantage that the laser device can be moved to any position, permitting a scan of complex objects. But operating a hand-held laser scanner is challenging. The operator has to keep track of the scanned regions in his mind, and has no feedback of the sample density unless he starts the surface reconstruction after finishing the scan. We present a system to support the operator by computing and rendering high-quality surface meshes of the captured data online, i.e., while he is still scanning, and in real time. Furthermore, it color-codes the rendered surface to reflect the surface quality. Thereby, instant feedback is provided, resulting in better scans in less time.
In the context of inverse optimization, inverse versions of maximum flow and minimum cost flow problems have thoroughly been investigated. In these network flow problems there are two important problem parameters: flow capacities of the arcs and costs incurred by sending a unit flow on these arcs. Capacity changes for maximum flow problems and cost changes for minimum cost flow problems have been studied under several distance measures such as rectilinear, Chebyshev, and Hamming distances. This thesis also deals with inverse network flow problems and their counterparts tension problems under the aforementioned distance measures. The major goals are to enrich the inverse optimization theory by introducing new inverse network problems that have not yet been treated in the literature, and to extend the well-known combinatorial results of inverse network flows for more general classes of problems with inherent combinatorial properties such as matroid flows on regular matroids and monotropic programming. To accomplish the first objective, the inverse maximum flow problem under Chebyshev norm is analyzed and the capacity inverse minimum cost flow problem, in which only arc capacities are perturbed, is introduced. In this way, it is attempted to close the gap between the capacity perturbing inverse network problems and the cost perturbing ones. The foremost purpose of studying inverse tension problems on networks is to achieve a well-established generalization of the inverse network problems. Since tensions are duals of network flows, carrying the theoretical results of network flows over to tensions follows quite intuitively. Using this intuitive link between network flows and tensions, a generalization to matroid flows and monotropic programs is built gradually up.
The manuscript divides in 7 chapters. Chapter 2 briefly introduces the reader to the elementary measures of classical continuum mechanics and thus allows to familiarize with the employed notation. Furthermore, deeper insight of the proposed first-order computational homogenization strategy is presented. Based on the need for a discrete representative volume element (rve), Chapter 3 focuses on a proper rve generation algorithm. Therein, the algorithm itself is described in detail. Additionally, we introduce the concept of periodicity. This chapter finalizes by granting multiple representative examples. A potential based soft particle contact method, used for the computations on the microscale level, is defined in Chapter 4. Included are a description of the used discrete element method (dem) as well as the applied macroscopically driven Dirichlet boundary conditions. The chapter closes with the proposition of a proper solution algorithm as well as illustrative representative examples. Homogenization of the discrete microscopic quantities is discussed in Chapter 5. Therein, the focus is on the upscaling of the aggregate energy as well as on the derivation of related macroscopic stress measures. Necessary quantities for coupling between a standard finite element method and the proposed discrete microscale are presented in Chapter 6. Therein, we tend to the derivation of the macroscopic tangent, necessary for the inclusion into the standard finite element programs. Chapter 7 focuses on the incorporation of inter-particle friction. We select to derive a variational based formulation of inter-particle friction forces, founded on a proposed reduced incremental potential. This contribution is closed by providing a discussion as well as an outlook.
Within this thesis we present a novel approach towards the modeling of strong discontinuities in a three dimensional finite element framework for large deformations. This novel finite element framework is modularly constructed containing three essential parts: (i) the bulk problem, ii) the cohesive interface problem and iii) the crack tracking problem. Within this modular design, chapter 2 (Continuous solid mechanics) treats the behavior of the bulk problem (i). It includes the overall description of the continuous kinematics, the required balance equations, the constitutive setting and the finite element formulation with its corresponding discretization and required solution strategy for the emerging highly non-linear equations. Subsequently, we discuss the modeling of strong discontinuities within finite element discretization schemes in chapter 3 (Discontinuous solid mechanics). Starting with an extension of the continuous kinematics to the discontinuous situation, we discuss the phantom-node discretization scheme based on the works of Hansbo & Hansbo. Thereby, in addition to a comparison with the extended finite element method (XFEM), importance is attached to the technical details for the adaptive introduction of the required discontinuous elements: The splitting of finite elements, the numerical integration, the visualization and the formulation of geometrical correct crack tip elements. In chapter 4 (The cohesive crack concept), we consider the treatment of cohesive process zones and the associated treatment of cohesive tractions. By applying this approach we are able to merge all irreversible, crack propagation accompanying, failure mechanisms into an arbitrary traction separation relation. Additionally, this concept ensures bounded crack tip stresses and allows the use of stress-based failure criteria for the determination of crack growth. In summary, the use of the discontinuous elements in conjunction with cohesive traction separation allows the mesh-independent computation of crack propagation along pre-defined crack paths. Therefore, this combination is defined as the interface problem (ii) and represents the next building block in the modular design of this thesis. The description and the computation of the evolving crack surface, based on the actual status of a considered specimen is the key issue of chapter 5 (Crack path tracking strategies). In contrast to the two-dimensional case, where tracking the path in a C0-continuous way is straightforward, three-dimensional crack path tracking requires additional strategies. We discuss the currently available approaches regarding this issue and further compare the approaches by means of usual quality measures. In the modular design of this thesis these algorithms represent the last main part which is classified as the crack tracking problem (iii). Finally chapter 6 (Representative numerical examples) verifies the finite element tool by comparisons of the computational results which experiments and benchmarks of engineering fracture problems in concrete. Afterwards the finite element tool is applied to model folding induced fracture of geological structures.
This thesis is devoted to applying symbolic methods to the problems of decoding linear codes and of algebraic cryptanalysis. The paradigm we employ here is as follows. We reformulate the initial problem in terms of systems of polynomial equations over a finite field. The solution(s) of such systems should yield a way to solve the initial problem. Our main tools for handling polynomials and polynomial systems in such a paradigm is the technique of Gröbner bases and normal form reductions. The first part of the thesis is devoted to formulating and solving specific polynomial systems that reduce the problem of decoding linear codes to the problem of polynomial system solving. We analyze the existing methods (mainly for the cyclic codes) and propose an original method for arbitrary linear codes that in some sense generalizes the Newton identities method widely known for cyclic codes. We investigate the structure of the underlying ideals and show how one can solve the decoding problem - both the so-called bounded decoding and more general nearest codeword decoding - by finding reduced Gröbner bases of these ideals. The main feature of the method is that unlike usual methods based on Gröbner bases for "finite field" situations, we do not add the so-called field equations. This tremendously simplifies the underlying ideals, thus making feasible working with quite large parameters of codes. Further we address complexity issues, by giving some insight to the Macaulay matrix of the underlying systems. By making a series of assumptions we are able to provide an upper bound for the complexity coefficient of our method. We address also finding the minimum distance and the weight distribution. We provide solid experimental material and comparisons with some of the existing methods in this area. In the second part we deal with the algebraic cryptanalysis of block iterative ciphers. Namely, we analyze the small-scale variants of the Advanced Encryption Standard (AES), which is a widely used modern block cipher. Here a cryptanalyst composes the polynomial systems which solutions should yield a secret key used by communicating parties in a symmetric cryptosystem. We analyze the systems formulated by researchers for the algebraic cryptanalysis, and identify the problem that conventional systems have many auxiliary variables that are not actually needed for the key recovery. Moreover, having many such auxiliary variables, specific to a given plaintext/ciphertext pair, complicates the use of several pairs which is common in cryptanalysis. We thus provide a new system where the auxiliary variables are eliminated via normal form reductions. The resulting system in key-variables only is then solved. We present experimental evidence that such an approach is quite good for small scaled ciphers. We investigate further our approach and employ the so-called meet-in-the-middle principle to see how far one can go in analyzing just 2-3 rounds of scaled ciphers. Additional "tuning techniques" are discussed together with experimental material. Overall, we believe that the material of this part of the thesis makes a step further in algebraic cryptanalysis of block ciphers.
In engineering and science, a multitude of problems exhibit an inherently geometric nature. The computational assessment of such problems requires an adequate representation by means of data structures and processing algorithms. One of the most widely adopted and recognized spatial data structures is the Delaunay triangulation which has its canonical dual in the Voronoi diagram. While the Voronoi diagram provides a simple and elegant framework to model spatial proximity, the core of which is the concept of natural neighbors, the Delaunay triangulation provides robust and efficient access to it. This combination explains the immense popularity of Voronoi- and Delaunay-based methods in all areas of science and engineering. This thesis addresses aspects from a variety of applications that share their affinity to the Voronoi diagram and the natural neighbor concept. First, an idea for the generalization of B-spline surfaces to unstructured knot sets over Voronoi diagrams is investigated. Then, a previously proposed method for \(C^2\) smooth natural neighbor interpolation is backed with concrete guidelines for its implementation. Smooth natural neighbor interpolation is also one of many applications requiring derivatives of the input data. The generation of derivative information in scattered data with the help of natural neighbors is described in detail. In a different setting, the computation of a discrete harmonic function in a point cloud is considered, and an observation is presented that relates natural neighbor coordinates to a continuous dependency between discrete harmonic functions and the coordinates of the point cloud. Attention is then turned to integrating the flexibility and meritable properties of natural neighbor interpolation into a framework that allows the algorithmically transparent and smooth extrapolation of any known natural neighbor interpolant. Finally, essential properties are proved for a recently introduced novel finite element tessellation technique in which a Delaunay triangulation is transformed into a unique polygonal tessellation.
This technical report contains the preliminary versions of the regular papers presented at the first workshop on Verification of Adaptive Systems (VerAS) that has been held in Kaiserslautern, Germany, on September 14th, 2007 as part of the 20th International Conference on Theorem Proving in Higher Order Logics. The final versions will be published with Elsevier's Electronic Notes on Theoretical Computer Science (ENTCS). VerAS is the first workshop that aims at considering adaptation as a cross-cutting system aspect that needs to be explicitly addressed in system design and verification. The program committee called for original submissions on formal modeling, specification, verification, and implementation of adaptive systems. There were six submissions from different countries of Europe. Each submission has been reviewed by three programme committee members. Finally, the programme committee decided to accept three of the six submissions. Besides the presentations of the regular papers, the workshop's programme included a tutorial on the `Compositional Verification of Self-Optimizing Mechatronic Systems' held by Holger Giese (University of Paderborn, Germany) as well as three presentations of DASMOD projects on the verification of adaptive systems.