Files in this item



application/pdf8803020.pdf (5MB)Restricted to U of Illinois
(no description provided)PDF


Title:A Linear, Time-Varying System Framework for Bandlimited Signal Extrapolation
Author(s):Diethorn, Eric John
Department / Program:Electrical Engineering
Discipline:Electrical Engineering
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Engineering, Electronics and Electrical
Abstract:A Fourier theory for modeling, analyzing, and designing linear, time-varying systems is used to study several aspects of bandlimited signal extrapolation, the problem of reconstructing the whole of a bandlimited signal from a known piece. The extrapolation of both continuous-time and discrete-time signals is considered.
The principal component of the approach is that of modeling the extrapolation process as a linear, time-varying operation on the known data, where the operator is defined by a linear, time-varying system--the extrapolation system. A frequency-domain representation facilitates the characterization of this system. In particular, this representation provides clear, visual evidence of the extreme noise-sensitivity associated with extrapolation--one of the most pressing issues of study today. A least-squares approximation procedure for designing extrapolation systems is also developed. This design methodology is shown to unify a large class of well-documented algorithms for extrapolating both continuous-time and discrete-time data. New techniques for extrapolating data in the presence of noise are also developed.
Issue Date:1987
Description:146 p.
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1987.
Other Identifier(s):(UMI)AAI8803020
Date Available in IDEALS:2014-12-15
Date Deposited:1987

This item appears in the following Collection(s)

Item Statistics