Title :
The discrete Gerchberg algorithm
Author_Institution :
University of Surrey, Guildford Surrey, U.K.
fDate :
6/1/1986 12:00:00 AM
Abstract :
The discrete version of the Gerchberg algorithm for iterative restoration of a time-constrained function from only partial knowledge of its spectrum (or vice versa) is analyzed. Although convergence is guaranteed, eigenvalues close to unity inhibit iteration to the limit. Identification of these large eigenvalues, allowing extrapolation to the limit, is described.
Keywords :
Algorithm design and analysis; Convergence; Distributed computing; Eigenvalues and eigenfunctions; Extrapolation; Fourier transforms; Frequency; Iterative algorithms; Physics; Signal processing algorithms;
Journal_Title :
Acoustics, Speech and Signal Processing, IEEE Transactions on
DOI :
10.1109/TASSP.1986.1164861
