Title :
Quality of synthesis and analysis methods for fractional Brownian motion
Author :
Jennane, Rachid ; Harba, Rachid ; Jacquet, Gérard
Author_Institution :
Orleans Univ., France
Abstract :
The article compares the quality of synthesis and analysis methods for fractional Brownian motion (fBm). Three conditions are necessary and sufficient for a process to be fBm. This allows one to check the correspondence between a given synthesis method and the fBm model. Two synthesis methods among 6 tested give good results. Reference signals obtained by these 2 methods are used to evaluate 8 analysis methods. Three estimators have a low bias and a variance close to the Cramer-Rao lower bound
Keywords :
Brownian motion; parameter estimation; signal processing; signal synthesis; Cramer-Rao lower bound; analysis methods; fractional Brownian motion; linear regresion methods; low bias; maximum likelihood estimator; necessary conditions; quality; reference signals; signal analysis; signal estimators; signal processing; signal synthesis; sufficient conditions; synthesis methods; variance; Automatic testing; Brownian motion; Discrete wavelet transforms; Fourier transforms; Parameter estimation; Random processes; Signal analysis; Signal synthesis; Statistical analysis; Sufficient conditions;
Conference_Titel :
Digital Signal Processing Workshop Proceedings, 1996., IEEE
Conference_Location :
Loen
Print_ISBN :
0-7803-3629-1
DOI :
10.1109/DSPWS.1996.555522
