Wavelet based estimators for self-similar /spl alpha/-stable processes
- Resource Type
- Conference
- Authors
- Abry, P.; Taqqu, M.S.; Pesquet-Popescu, B.
- Source
- WCC 2000 - ICSP 2000. 2000 5th International Conference on Signal Processing Proceedings. 16th World Computer Congress 2000 Signal processing - Beijing Signal Processing Proceedings, 2000. WCCC-ICSP 2000. 5th International Conference on. 1:369-375 vol.1 2000
- Subject
- Signal Processing and Analysis
Computing and Processing
Communication, Networking and Broadcast Technologies
Wavelet coefficients
Continuous wavelet transforms
Discrete wavelet transforms
Kernel
Virtual colonoscopy
Wavelet transforms
Random variables
Statistics
Frequency
Virtual manufacturing
- Language
The aim of this work is to propose wavelet based estimators for the self-similarity and stability parameters of stable self-similar processes. These estimators are simple both conceptually and practically, consisting of linear fits performed on the wavelet coefficients of stable self-similar processes. Their statistical performance are studied both theoretically, from the analysis of the statistical properties of the wavelet coefficients, and using numerical simulations.