Accession Number : ADA140906

Title :   Statistical Techniques for Signal Processing.

Descriptive Note : Interim scientific rept. 1 Nov 82-31 Oct 83,

Corporate Author : MOORE SCHOOL OF ELECTRICAL ENGINEERING PHILADELPHIA PA DEPT OF ELECTRICAL ENGINEERING AND SCIENCE

Personal Author(s) : Kassam,S A

PDF Url : ADA140906

Report Date : 15 Dec 1983

Pagination or Media Count : 6

Abstract : This report describes the results obtained during the grant year 1 November 1982 to 31 October 1983. References are made to twelve publications produced or in preparation. The research supported was in the area of robust signal processing, particularly nonlinear filtering and smoothing. Results were obtained on quantization of filters and on multiple-input robust matched filters. Considerable progress was made in the area of nonlinear filters and smoothers, where it appears that generalizations of median filters may be of considerable practical significance in applications such as speech and image processing. In addition, some further analytical results have been obtained on the statistical properties of median filter outputs for Markov models on signal-plus-noise. Results on output power spectral densities for white inputs have also been obtained for M-filters based on robust M-estimation. (Author)

Descriptors :   *Statistical processes, *Signal processing, Mathematical filters, Matched filters, Power spectra, Markov processes

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE