The USC Andrew and Erna Viterbi School of Engineering USC Signal and Image Processing Institute USC Ming Hsieh Department of Electrical and Computer Engineering University of Southern California

Technical Report USC-SIPI-230

“Infinite and Finite Observation Interval Whitening Filters for Certain Fractional Stable Noises”

by George A. Tsihrintzis and Chrysostomos L. Nikias

January 1993

We extend the Mandelbrot-Van Ness models of fractional Gaussian noise to define linear and log fractional symmetric _-stable (S_S) noises. We present both "engineering" definitions, based on certain fractional integrals of white S_S noise, as well as rigorous mathematical definitions via appropriate linear operators acting on subsets of the Banach space L_, 1< _ < 2. For these noises we find a canonical representation and proceed to derive whitening filters for linear and log fractional S_S noises for both infinite and finite observation intervals. The overall style of the report is tutorial complemented with a large number of references which review the relevant literature.

To download the report in PDF format click here: USC-SIPI-230.pdf (1.0Mb)