A new and efficient method for the computation of Legendre coefficients

Enrico De Micheli,Giovanni Alberto Viano

DOI: https://doi.org/10.48550/arXiv.1106.0463

2011-06-03

Abstract:An efficient procedure for the computation of the coefficients of Legendre expansions is here presented. We prove that the Legendre coefficients associated with a function f(x) can be represented as the Fourier coefficients of an Abel-type transform of f(x). The computation of N Legendre coefficients can then be performed in O(N log N) operations with a single Fast Fourier Transform of the Abel-type transform of f(x).

Numerical Analysis,Mathematical Physics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the efficient calculation of Legendre coefficients. Specifically, the paper proposes a new method to calculate the Legendre expansion coefficients of the function \( f(x) \) in order to overcome the inefficiency of traditional methods in computational efficiency.

### Problem Background

The Legendre polynomial expansion is of great significance in numerical analysis and applied mathematics, and is widely used in approximation theory, partial differential equation solving, numerical integration and other fields. However, the traditional calculation methods have the following problems:

1. **High computational complexity**: Using the standard quadrature formula to calculate \( N \) Legendre coefficients requires \( O(N^2) \) operations.
2. **Integral oscillation**: When the order of the polynomial is high, the integrand of the integral will oscillate rapidly, resulting in computational difficulties.

### Solution Proposed in the Paper

The paper proposes a method based on the Abel - type transform to relate the Legendre coefficients to the Fourier coefficients. The specific steps are as follows:

1. **Abel transform**: Define an Abel - type transform \( \hat{f}(y) \) such that the Legendre coefficients can be represented by the Fourier coefficients of this transform.

\[
\hat{f}(y)=\frac{1}{2\pi i}\varepsilon(y)e^{iy / 2}\int_{\cos y}^{1}\frac{f(x)}{\sqrt{2(x-\cos y)}}dx
\]

2. **Fourier coefficient representation**: Prove that the Legendre coefficient \( c_{n} \) can be represented by the Fourier coefficient \( a_{n} \) of the Abel transform as:

\[
a_{n}=\frac{c_{n}}{2n + 1}=\int_{-\pi}^{\pi}\hat{f}(y)e^{iny}dy\quad(n\geq0)
\]

3. **Fast Fourier Transform (FFT)**: By using the Fast Fourier Transform (FFT), \( N \) Legendre coefficients can be efficiently calculated within \( O(N\log N) \) operations.

### Numerical Implementation

The numerical implementation of this algorithm includes the following steps:

1. **Calculate the Abel transform**: First, calculate the Abel transform \( \hat{f}(y) \) or the equivalent function \( \varphi(y) \).
2. **Apply FFT**: Apply FFT to the result of the Abel transform to obtain the required Fourier coefficients, and then obtain the Legendre coefficients.

Through this method, not only the calculation speed is improved, but also a high precision is ensured. The experimental results show that for a variety of test functions, this algorithm can significantly improve the calculation speed with almost the same precision, especially in multivariate Legendre transformation and Gegenbauer polynomial expansion.

### Summary

The main contribution of the paper is to provide a new and efficient method for calculating Legendre coefficients. By relating the Legendre coefficients to the Fourier coefficients and using the Fast Fourier Transform, the computational complexity is significantly reduced from \( O(N^{2}) \) to \( O(N\log N) \). This provides a more efficient solution for many problems in numerical analysis and applied mathematics.