Search results for "matrix decomposition"
showing 10 items of 43 documents
On the implementation of weno schemes for a class of polydisperse sedimentation models
2011
The sedimentation of a polydisperse suspension of small rigid spheres of the same density, but which belong to a finite number of species (size classes), can be described by a spatially one-dimensional system of first-order, nonlinear, strongly coupled conservation laws. The unknowns are the volume fractions (concentrations) of each species as functions of depth and time. Typical solutions, e.g. for batch settling in a column, include discontinuities (kinematic shocks) separating areas of different composition. The accurate numerical approximation of these solutions is a challenge since closed-form eigenvalues and eigenvectors of the flux Jacobian are usually not available, and the characte…
A Flux-Split Algorithm Applied to Relativistic Flows
1998
The equations of RFD can be written as a hyperbolic system of conservation laws by choosing an appropriate vector of unknowns. We give an explicit formulation of the full spectral decomposition of the Jacobian matrices associated with the fluxes in each spatial direction, which is the essential ingredient of the techniques we propose in this paper. These techniques are based on the recently derived flux formula of Marquina, a new way to compute the numerical flux at a cell interface which leads to a conservative, upwind numerical scheme. Using the spectral decompositions in a fundamental way, we construct high order versions of the basic first-order scheme described by R. Donat and A. Marqu…
On the Rigorous Calculation of All Ohmic Losses in Rectangular Waveguide Multi-Port Junctions
2005
In this paper, all ohmic losses effects present in rectangular waveguide multi-port junctions are rigorous and efficiently computed. For this purpose, a new formulation based on the theory of cavities, which provides generalized admittance matrix representations for such junctions, is proposed. To validate this theory, we have successfully compared our results with numerical data of a lossy E-plane T-junction and of a hollow waveguide, as well as with experimental measurements of a real H-plane T-junction.
Sub-wavelength and non-periodic holes array based fully lensless imager
2011
Abstract We present a novel concept for microscopic imaging. The proposed microscope-like device does not include an objective lens neither a condenser. Instead, a metallic plate of sub-wavelength hole-array with a varying pitch is used to illuminate the inspected object that is mounted very close to it. As a result, the transmitted spectrum through each hole differs from the others and therefore, each spot of the detected object is illuminated with a unique spectrum. By measuring a single spectrum that is the sum of all the spectra that are transmitted through the sample and by using spectral decomposition algorithms, the spatial transmission pattern of the object can be extracted.
Computing Strong Shocks in Ultrarelativistic Flows: A Robust Alternative
1999
In recent years, shock capturing methods have started to be used in numerical simulations in Relativistic Fluid Dynamics (RFD). These techniques lead to explicit numerical codes that are able to successfully simulate the extreme conditions of the ultrarelativistic regime. After [2], an explicit, ready-to-use description of the full spectral decomposition of the Jacobian matrices of the RFD system is available, and this allows us to implement Marquina’s scheme [3] in RFD. The scheme is seen to maintain the good behavior shown in [3] with respect to certain numerical pathologies.
Jacobian-free approximate solvers for hyperbolic systems: Application to relativistic magnetohydrodynamics
2017
Abstract We present recent advances in PVM (Polynomial Viscosity Matrix) methods based on internal approximations to the absolute value function, and compare them with Chebyshev-based PVM solvers. These solvers only require a bound on the maximum wave speed, so no spectral decomposition is needed. Another important feature of the proposed methods is that they are suitable to be written in Jacobian-free form, in which only evaluations of the physical flux are used. This is particularly interesting when considering systems for which the Jacobians involve complex expressions, e.g., the relativistic magnetohydrodynamics (RMHD) equations. On the other hand, the proposed Jacobian-free solvers hav…
Jacobian-Free Incomplete Riemann Solvers
2018
The purpose of this work is to present some recent developments about incomplete Riemann solvers for general hyperbolic systems. Polynomial Viscosity Matrix (PVM) methods based on internal approximations to the absolute value function are introduced, and they are compared with Chebyshev-based PVM solvers. These solvers only require a bound on the maximum wave speed, so no spectral decomposition is needed. Moreover, they can be written in Jacobian-free form, in which only evaluations of the physical flux are used. This is particularly interesting when considering systems for which the Jacobians involve complex expressions. Some numerical experiments involving the relativistic magnetohydrodyn…
Scalar glueball spectrum
2006
I discuss scenarios for scalar glueballs using arguments based on sum rules, spectral decomposition, the $\frac{1}{{N}_{c}}$ approximation, the scales of the strong interaction and the topology of the flux tubes. I analyze the phenomenological support of those scenarios and their observational implications. My investigations hint a rich low lying glueball spectrum.
Low-Rank Tucker-2 Model for Multi-Subject fMRI Data Decomposition with Spatial Sparsity Constraint
2022
Tucker decomposition can provide an intuitive summary to understand brain function by decomposing multi-subject fMRI data into a core tensor and multiple factor matrices, and was mostly used to extract functional connectivity patterns across time/subjects using orthogonality constraints. However, these algorithms are unsuitable for extracting common spatial and temporal patterns across subjects due to distinct characteristics such as high-level noise. Motivated by a successful application of Tucker decomposition to image denoising and the intrinsic sparsity of spatial activations in fMRI, we propose a low-rank Tucker-2 model with spatial sparsity constraint to analyze multi-subject fMRI dat…
Convolutional Matrix Factorization for Recommendation Explanation
2018
In this paper, we introduce a novel recommendation model, which harnesses a convolutional neural network to mine meaningful information from customer reviews, and integrates it with matrix factorization algorithm seamlessly. It is a valid method to improve the transparency of CF algorithms.