Title :
On the relationship between 1/f and alpha-stable processes
Author :
Yin, Jijun ; Petropulu, A.P.
Author_Institution :
Drexel Univ., Philadelphia, PA, USA
Abstract :
1/fβ-type spectral behavior has received considerable attention in the literature because it arises from a wide range of natural phenomena. In this paper we consider a filtered Poisson process model, and show that this model gives rise to a power-law α-stable process if β<1, or to a process with power-law α-stable increments if 1<β<3. Based on the relationship between α and β we propose a new method for the estimation of the spectral exponent of the sample spectrum, which is particularly robust in the case 0<β<1
Keywords :
Poisson distribution; digital filters; estimation theory; shot noise; spectral analysis; statistical analysis; stochastic processes; 1/f; alpha-stable processes; filtered Poisson process model; power-law α-stable increments; power-law α-stable process; sample spectrum; spectral behavior; spectral exponent; 1f noise; Economic forecasting; Fuel economy; Gaussian distribution; Maximum likelihood estimation; Power generation economics; Robustness; Stochastic processes; Telecommunication traffic; Wavelet analysis;
Conference_Titel :
Information Theory, 1998. Proceedings. 1998 IEEE International Symposium on
Conference_Location :
Cambridge, MA
Print_ISBN :
0-7803-5000-6
DOI :
10.1109/ISIT.1998.708907
