c1c2c6ad50eeac7b.tex
1: \begin{abstract}
2: In this paper, the  numerical
3: differentiation by integration method based on Jacobi polynomials originally introduced by Mboup, Fliess and Join \cite{num,num0} is revisited in the central case where the used integration window is centered.  Such a method based on Jacobi polynomials was introduced through an algebraic approach \cite{num,num0} and extends the numerical
4: differentiation by integration method introduced by  Lanczos (1956) \cite{C. Lanczos}. The method  proposed here, rooted in \cite{num,num0}, is used to estimate the $n${th} ($n \in \mathbb{N}$) order derivative from
5: noisy data of a smooth function belonging to at least $C^{n+1+q}$
6: $(q \in \mathbb{N})$. In \cite{num,num0}, where the causal and anti-causal cases were investigated, the mismodelling due to the truncation of the Taylor expansion was investigated and improved allowing a small time-delay in the derivative estimation. Here, for the central case, we show that the bias error is $O(h^{q+2})$ where $h$ is the integration window length for $f\in C^{n+q+2}$ in the noise free case and the corresponding convergence
7: rate is $O(\delta^{\frac{q+1}{n+1+q}})$ where $\delta$ is the noise
8: level for a well-chosen integration window length.
9: Numerical examples show that this proposed method is stable and effective.
10: 
11: \end{abstract}