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

Technical Report USC-SIPI-253

“Maximum A Posteriori Filter Estimation for Cauchy Data Using Lower-Order Statistics”

by Russell H. Lambert and Chrysostomos L. Nikias

March 1994

A Maximum A Posteriori (MAP) estimation approach is used to determine the multipath channel parameters of a system driven with Cauchy data. This is a blind deconvolution problem with the input driving sequence having a Cauchy probability density function. The fact that the Cauchy distribution is not Bussgang, and has infinite variance, requires the development of a new blind deconvolution formulation which estimates the forward filter, instead of the inverse filter as in traditional deconvolution schemes.

To download the report in PDF format click here: USC-SIPI-253.pdf (0.5Mb)