
	Information Technology Journal
	Year: 2012 \| Volume: 11 \| Issue: 8 \| Page No.: 1121-1126 DOI: 10.3923/itj.2012.1121.1126

A Robust and Simple Piecewise Approximation to SαS Distribution with Bi-Region Model

Kang Wang, Zhi-Jiang Xu, Yuan Wu and Jing-Yu Hua

Abstract: As a non-Gaussian model, the alpha stable distribution has gained much attention because of its generality to model the heavy-tail and impulsive noise which is widely observed in many communication channels. Unfortunately, there exists no analytic expression for the Probability Density Function (PDF) of Symmetric Alpha Stable (SαS) distribution. In order to approximate the PDF of SαS, we propose a bi-region curve approximation algorithm with the bi-region separated by the triple divergence. Specially, within the triple divergence, we propose a penalty function with two special parameters and adopt a very simple and effective exponential function for approximation. Different from the existing algorithms using the series expansion, our model avoids the problem of selecting the number of the series items and the risk of series expansion divergence. Compared with the conventional Cauchy-Gaussian mixture approximation, our derivation emploits the simple bi-region approximation and yields a very simple and closed-form expression. Numerical results verify that our approximation is very close to the actual PDF of SαS.

PDF Fulltext XML References Citation Report Citation

RELATED ARTICLES:

•	A Novel Structure for Covert Communication Based on Alpha Stable Distribution

•	A Novel Adaptive Regularized Possibilistic Linear Models Based Median Filter ARBMF for Image Noise Suppression

•	Quality Assessment of Pedochemical Data Using Extreme Value Methodology

•	An Effective Method for Exact Reliability Analysis

•	Probabilistic Analysis of Eigenvalue of Stochastic Systems

How to cite this article:

Kang Wang, Zhi-Jiang Xu, Yuan Wu and Jing-Yu Hua, 2012. A Robust and Simple Piecewise Approximation to SαS Distribution with Bi-Region Model. Information Technology Journal, 11: 1121-1126.

DOI: 10.3923/itj.2012.1121.1126

URL: https://scialert.net/abstract/?doi=itj.2012.1121.1126

COMMENT ON THIS PAPER

Navigation


		Online First
		Current Issue
		Previous Issues
		Editorial Board
		Submit a Manuscript
		Guide to Authors
		Subscribe to E-alerts

Indexed In


		ASCI-Database
		Asian Digital Library
		Google Scholar

Science Alert