Skip to Main content Skip to Navigation
Theses

Débruitage et interpolation par analyse de la régularité Hölderienne. Application à la modélisation du frottement pneumatique-chaussée.

Abstract : Wavelet and fractal analysis are both adapted to the study of irregular non-stationnary signals. This thesis deals with the analysis of signals for which the local regularity bears important information. In that view, several tools are developped which center on the notion of Hölder exponent. First, some techniques allowing to estimate this exponent are compared. Then, an interpolation method based on preserving the local regularity is proposed. In the third chapter, new denoising methods are developped. They allow a control of the Hölder exponent of the denoised signal. In addition, their convergence rate is, in many cases, rate-minimax. The second part of the thesis deals with the application of Hölder regularity to the processing of road profiles, in a view to modeling road-tyre friction. Work performed at LCPC has proved that the microtexture of raod profiles plays an important role in the friction process. We first show that road profiles do exhibit a fractal behaviour over a large range of scales. Then we explain how the interpolation and denoising methods developped in the first part allow to improve the results of the Stefani model developped at LCPC. Finally, a multi-scale version of this model is developped.
Complete list of metadatas

Cited literature [75 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00643450
Contributor : Pierrick Legrand <>
Submitted on : Tuesday, November 22, 2011 - 3:04:59 PM
Last modification on : Thursday, May 2, 2019 - 2:10:03 PM
Long-term archiving on: : Thursday, February 23, 2012 - 2:20:41 AM

Identifiers

  • HAL Id : tel-00643450, version 1

Citation

Pierrick Legrand. Débruitage et interpolation par analyse de la régularité Hölderienne. Application à la modélisation du frottement pneumatique-chaussée.. Traitement du signal et de l'image [eess.SP]. Ecole Centrale de Nantes (ECN); Université de Nantes, 2004. Français. ⟨tel-00643450⟩

Share

Metrics

Record views

641

Files downloads

1021