Generalized hypergroups and orthogonal polynomials

Rupert Lasser, Josef Obermaier, Holger RauhutJournal of the Australian Mathematical Society 2007 · DOI

Abstract

The concept of semi-bounded generalized hypergroups (SBG hypergroups) is developed. These hypergroups are more special than generalized hypergroups introduced by Obata and Wildberger and more general than discrete hypergroups or even discrete signed hypergroups. The convolution of measures and functions is studied. In the case of commutativity, we define the dual objects and prove some basic theorems of Fourier analysis. Furthermore, we investigate the relationship between orthogonal polynomials and generalized hypergroups, using Jacobi polynomials as a primary example.

Introduction

In the field of abstract harmonic analysis, hypergroups serve as a vital generalization of groups. While groups provide a rigid structure for multiplication, hypergroups allow for a probabilistic or "fuzzy" convolution. This flexibility is essential for performing Fourier analysis on spaces that do not possess a natural group structure. However, a persistent gap has existed between standard hypergroups—which are often too restrictive for many systems of orthogonal polynomials—and the generalized hypergroups introduced by Obata and Wildberger, which lacked the constraints necessary for a robust Fourier theory.

This paper introduces semi-bounded generalized hypergroups (SBG hypergroups) to bridge this gap, providing a framework that is flexible enough to encompass broad classes of polynomials while remaining mathematically tractable for harmonic analysis.

The Problem: Balancing Flexibility and Structure

Traditional hypergroup theory requires strict boundedness, which many important mathematical systems fail to satisfy. Conversely, previous attempts at generalized hypergroups were so broad that they lacked the necessary control over "structure constants"—the values that describe how basis elements interact. Without this control, defining a well-behaved convolution of measures and functions becomes nearly impossible.

SBG hypergroups resolve this by being "semi-bounded." This means they provide just enough constraint on their internal structure to allow for meaningful functional analysis without requiring the rigid stability of traditional signed hypergroups.

Methods and Axiomatic Framework

The authors construct the theory of SBG hypergroups through several rigorous steps:

1. Axiom A4 and Semi-Boundedness

The defining feature of this work is the introduction of Axiom A4. This axiom requires that the structure constants of the hypergroup grow at a controlled, semi-bounded rate. This ensures that the convolution operator remains well-behaved when applied to various function spaces.

2. Functional Analysis in Banach Spaces

To prove the stability of their convolution, the authors utilize the Riesz-Thorin interpolation theorem. This allows them to demonstrate that the convolution of functions extends naturally across different $l^p$ spaces, establishing a foundation for more complex operations.

3. Dual Spaces and Fourier Theory

For commutative SBG hypergroups, the authors construct "dual objects." These are sets of characters—functions that preserve the underlying hypergroup structure. By defining these duals, the authors can establish a formal Fourier transform, moving analysis from the discrete spatial domain to a frequency-like domain.

4. Recurrence Relations

The paper explicitly links algebraic hypergroup structures to the three-term recurrence relations that define orthogonal polynomials: $$P_1 P_n = a_n P_{n+1} + b_n P_n + c_n P_{n-1}$$ This mapping allows the properties of the polynomials to be studied through the lens of hypergroup theory.

Key Insights and Results

The "Goldilocks" Constraint

The core theoretical contribution is the realization that Fourier analysis does not require absolute boundedness. By allowing structure constants to grow at a controlled rate, the authors encompass systems where "multiplication" evolves as the index increases, provided that growth is not exponential.

Plancherel’s Theorem and Uniqueness

The authors successfully prove a version of Plancherel’s theorem for SBG hypergroups. This theorem shows that the Fourier transform acts as an isometric isomorphism between specific function spaces, meaning that the "energy" or information of a function is preserved during the transformation. Additionally, they prove Fourier Uniqueness (Theorem 5.5): if the Fourier transform of a function is zero, the function itself must be zero.

Application to Jacobi Polynomials

The theory is validated using Jacobi polynomials. While these polynomials often fail the criteria for standard hypergroups, the authors demonstrate that they almost always satisfy the requirements for SBG hypergroups. In the "ultraspherical" case, they derive explicit formulas showing that the structure constants grow polynomially, fitting perfectly within the semi-bounded framework.

Conclusion and Future Work

By formalizing SBG hypergroups, this paper provides a new set of tools for researchers in approximation theory and harmonic analysis. However, several questions remain open:

  • Compactness: It is not yet proven if the dual space of an SBG polynomial hypergroup is always compact in the general case.
  • Dimensionality: The current focus is on discrete structures (sequences); extending this to locally compact continuous spaces is a logical next step.
  • Non-Commutativity: Most results here rely on commutativity. Exploring non-commutative SBG hypergroups could open new avenues in non-abelian harmonic analysis.

Summary updated

Abstract

Abstract The concept of semi-bounded generalized hypergroups (SBG hypergroups) is developed. These hypergroups are more special than generalized hypergroups introduced by Obata and Wildberger and more general than discrete hypergroups or even discrete signed hypergroups. The convolution of measures and functions is studied. In the case of commutativity we define the dual objects and prove some basic theorems of Fourier analysis. Furthermore, we investigate the relationship between orthogonal polynomials and generalized hypergroups. We discuss the Jacobi polynomials as an example.
BibTeX
@article{lasser2007generalized,
  author = {Rupert Lasser and Josef Obermaier and Holger Rauhut},
  title = {{Generalized hypergroups and orthogonal polynomials}},
  journal = {Journal of the Australian Mathematical Society},
  year = {2007},
  doi = {10.1017/s144678870003617x},
}
Figure