A Novel Parameter Estimation Method of Alpha-stable Distribution Based on Extreme Value

Author

Li, Li ; Yu, Li ; Zhu, Guangxi ; Pei, Xuebing

Author_Institution

Huazhong Univ. of Sci. & Technol., Wuhan

fYear

2007

fDate

22-24 Aug. 2007

Firstpage

621

Lastpage

625

Abstract

Since Alpha-stable distribution was introduced as a generalization of Gaussian distribution, it has been applied to model and to analyze signals with high variability. After analyzing characteristic of PDF of Alpha-stable distribution, we propose a novel parameter estimation method of Alpha-stable distribution, based on asymptotic Pareto characteristic and asymmetry of its tails. Simulation and analysis results prove that this method could achieve much more accuracy and stability compared with other estimation methods.

Keywords

Gaussian distribution; Pareto distribution; parameter estimation; signal processing; Gaussian distribution; alpha-stable distribution; asymptotic Pareto characteristic; parameter estimation method; signal analysis; Analytical models; Gaussian distribution; Information analysis; Maximum likelihood estimation; Parameter estimation; Pareto analysis; Probability distribution; Signal analysis; Signal processing; Stability analysis; Alpha-stable distribution; asymmetric tails; asymptotic Pareto characteristic; parameter estimation;

fLanguage

English

Publisher

ieee

Conference_Titel

Communications and Networking in China, 2007. CHINACOM '07. Second International Conference on

Conference_Location

Shanghai

Print_ISBN

978-1-4244-1009-5

Electronic_ISBN

978-1-4244-1009-5

Type

conf

DOI

10.1109/CHINACOM.2007.4469467

Filename

4469467