Chebyshev admissible meshes and Lebesgue constants of complex polynomial projections

Leokadia Bialas-Ciez,Dimitri Jordan Kenne,Alvise Sommariva,Marco Vianello
2023-11-11
Abstract:We construct admissible polynomial meshes on piecewise polynomial or trigonometric curves of the complex plane, by mapping univariate Chebyshev points. Such meshes can be used for polynomial least-squares, for the extraction of Fekete-like and Leja-like interpolation sets, and also for the evaluation of their Lebesgue constants.
Numerical Analysis
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to construct admissible polynomial meshes of piecewise polynomial or trigonometric curves on the complex plane by mapping one - dimensional Chebyshev points. These meshes can be used in polynomial least - squares methods, for extracting interpolation sets similar to Fekete and Leja types, and for evaluating their Lebesgue constants. Specifically: 1. **Construction of Admissible Polynomial Meshes**: The paper mainly focuses on constructing optimal admissible polynomial meshes on the complex plane, and these meshes can be generated by mapping one - dimensional Chebyshev points. 2. **Application Areas**: These meshes can be applied to polynomial least - squares methods, for extracting interpolation sets similar to Fekete and Leja types, and for evaluating the Lebesgue constants of these interpolation sets. 3. **Theoretical Basis**: Based on the work of Calvi and Levenberg, admissible polynomial meshes have become important tools in polynomial approximation. The paper extends this concept, especially in applications on the complex plane. ### Formula Summary - **Definition of Admissible Polynomial Meshes**: \[ \|p\|_K \leq c \|p\|_{Z_n}, \quad \forall p \in P_n(\mathbb{C}) \] where $\| \cdot \|_Y$ represents the supremum norm on the continuous or discrete compact set $Y$, $\text{card}(Z_n) = O(n^\alpha)$, $\alpha \geq 1$, and $c$ is a constant independent of $n$. - **Inequality in Lemma 1**: \[ \| \varphi \|_{[a,b]} \leq c_m \| \varphi \|_{X_m^\nu}, \quad c_m := \frac{1}{\cos\left(\frac{\pi}{2m}\right)} \] - **Inequality in Proposition 1**: \[ \| p \|_\Gamma \leq c_m \| p \|_{Z_m^n(k)} \] - **Estimation of Lebesgue Constants**: \[ \|\lambda_n\|_{Z_m^n} \leq \|L_n\| \leq c_m \|\lambda_n\|_{Z_m^n} \] \[ 0 \leq \|L_n\| - \|\lambda_n\|_{Z_m^n} \leq (c_m - 1) \|\lambda_n\|_{Z_m^n} \] ### Conclusion By constructing these admissible polynomial meshes, the paper provides an effective method for dealing with polynomial approximation problems on the complex plane, especially for the applications of interpolation and least - squares methods. At the same time, the paper also provides strict error bounds for evaluating Lebesgue constants, which are very important for understanding the quality of interpolation and approximation.