2017

group

Authors' Class Standing

Austin Ogle, Junior Jacki Qi, Graduated Senior

Austin Ogle

Sirani M. Perera

Abstract

Modeling phenomenon of the interpolation problems can be seen in propagation of waves, weather conditions, real-time traffic patterns, signal processing, etc. There are different interpolation methods like polynomial interpolation, spline interpolation, rational interpolation, exponential interpolation, trigonometric interpolation, etc. In this situation, a fast Bj rck-Pereyra-type algorithm can be derived to solve this problem.

In this poster we will present the most general trigonometric interpolation problems to solve complex-Vandermonde system. We present a fast algorithm for solving a system where the coefficient-matrix is a complex Vandermonde matrix. This method is much more favorable than the Gaussian elimination, which ignores the structure of the Vandermonde Matrix.

The new algorithm applies to a fairly general new class to solve trigonometric interpolation problems. We present numerical experiments while elaborating better forward error bound than the Gaussian elimination. Moreover, we analyze, compare, and contrast the connection of trigonometric interpolation problems to phase polynomials interpolation problems through complexity, forward error, and stability of algorithms.

No

Share

COinS

A Fast Bjo ̈rck-Pereyra-type Algorithm for Solving Complex-Vandermonde Systems

Modeling phenomenon of the interpolation problems can be seen in propagation of waves, weather conditions, real-time traffic patterns, signal processing, etc. There are different interpolation methods like polynomial interpolation, spline interpolation, rational interpolation, exponential interpolation, trigonometric interpolation, etc. In this situation, a fast Bj rck-Pereyra-type algorithm can be derived to solve this problem.

In this poster we will present the most general trigonometric interpolation problems to solve complex-Vandermonde system. We present a fast algorithm for solving a system where the coefficient-matrix is a complex Vandermonde matrix. This method is much more favorable than the Gaussian elimination, which ignores the structure of the Vandermonde Matrix.

The new algorithm applies to a fairly general new class to solve trigonometric interpolation problems. We present numerical experiments while elaborating better forward error bound than the Gaussian elimination. Moreover, we analyze, compare, and contrast the connection of trigonometric interpolation problems to phase polynomials interpolation problems through complexity, forward error, and stability of algorithms.