Title :
Performing dimensionality reduction of nonstationary stochastic processes using perturbation analysis
Author :
Vigneron, V. ; Maaref, H.
Author_Institution :
LSC, Evry, France
Abstract :
In this paper, we propose strategies based on sensor selection. We have shown that sensor selection is a viable approach in the absence of reliable and detailed prior information by simply using the autocorrelation of the signals and its derivative.
Keywords :
adaptive filters; digital filters; eigenvalues and eigenfunctions; perturbation theory; stochastic processes; SVD; adaptive filter; digital filter; dimensionality reduction; eigenvalues; nonstationary stochastic processes; perturbation analysis; singular value decomposition; stochastic approximation; Fetus; IP networks;
Conference_Titel :
Signal Processing Conference, 2005 13th European
Conference_Location :
Antalya
Print_ISBN :
978-160-4238-21-1
