Generalized prolate spheroidal functions: algorithms and analysis

Philip Greengard
2024-05-07
Abstract:Generalized prolate spheroidal functions (GPSFs) arise naturally in the study of bandlimited functions as the eigenfunctions of a certain truncated Fourier transform. In one dimension, the theory of GPSFs (typically referred to as prolate spheroidal wave functions) has a long history and is fairly complete. Furthermore, more recent work has led to the development of numerical algorithms for their computation and use in applications. In this paper we consider the more general problem, extending the one dimensional analysis and algorithms to the case of arbitrary dimension. Specifically, we introduce algorithms for efficient evaluation of GPSFs and their corresponding eigenvalues, quadrature rules for bandlimited functions, formulae for interpolation via GPSF expansion, and various analytical properties of GPSFs. We illustrate the numerical and analytical results with several numerical examples.
Numerical Analysis
What problem does this paper attempt to address?
The paper attempts to address the problem of extending the theoretical analysis and numerical algorithms of Generalized Prolate Spheroidal Functions (GPSFs) in arbitrary dimensions. Specifically, the paper focuses on the following aspects: 1. **Definition and Properties of Generalized Prolate Spheroidal Functions (GPSFs)**: Defining GPSFs in arbitrary dimensions and studying their fundamental properties, including their role as eigenfunctions of truncated Fourier transforms. 2. **Numerical Algorithms**: Proposing efficient algorithms to evaluate GPSFs and their corresponding eigenvalues. These algorithms include: - Numerical methods for eigenvalue computation. - Quadrature rules for the integration of band-limited functions. - Numerical schemes for representing band-limited functions using GPSF expansions. 3. **Numerical Experiments**: Validating the effectiveness and accuracy of the proposed algorithms through numerical experiments and demonstrating the properties of GPSFs and their eigenvalues. In summary, the main objective of this paper is to extend the theory and numerical methods of one-dimensional Prolate Spheroidal Wave Functions (PSWFs) to multi-dimensional spaces and to propose new algorithms for efficiently computing and applying GPSFs.