Accession Number : ADA186725

Title :   Outlier Resistant Predictive Source Encoding for a Gaussian Stationary Nominal Source.

Descriptive Note : Technical rept. for period ending 1 Jul 87,

Corporate Author : VIRGINIA UNIV CHARLOTTESVILLE DEPT OF ELECTRICAL ENGINEERING

Personal Author(s) : Kazakos, P

PDF Url : ADA186725

Report Date : 18 Sep 1987

Pagination or Media Count : 43

Abstract : A sequence of qualitatively robust predictive source encoders, for a Gaussian stationary source with outlier contaminated observation data, is proposed and analyzed. Performance measures include mean difference-sequence distortion and output entropy at the nominal Gaussian source, as well as breakdown point and influence function. The proposed sequence of predictive encoders attains strictly positive breakdown point and uniformly bounded influence function, at the expense of increased mean difference-squared distortion and differential entropy, at the Gaussian nominal source.

Descriptors :   *ENTROPY, *CODING, *STATISTICAL ANALYSIS, CONTAMINATION, OBSERVATION, COSTS, OUTPUT, PROBABILITY DENSITY FUNCTIONS

Subject Categories : Cybernetics

Distribution Statement : APPROVED FOR PUBLIC RELEASE