IDENTIFICATION OF LINEAR SYSTEMS USING FAST FOURIER TRANSFORM TECHNIQUES.
Abstract
An algorithm is proposed for the identification of the impulse response function of a linear system from input-output records by means of Fast Fourier Transform (FFT) techniques. The errors resulting from the need to truncate input-output data are reduced for a large class of data sets by the introduction of Contraction Mapping iterations--the Contracting Operator FFT Identification (COFFTI) method A new input-output relation is derived based upon truncated functions in the time domain which has the form required by the hypotheses of the Contraction Mapping Theorem, x = G(X). Computer studies are presented for single-input, single-output systems using a variety of known impulse response functions. The COFFTI algorithm is extended to multivariable systems. A comparison is made between the computational complexities of the COFFTI algorithm and the solution obtained by direct matrix inversion. This comparison illustrates the computational advantages of the algorithm.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1970
- Accession Number
- AD0711104
Entities
People
- Harvey Fox Silverman
Organizations
- Brown University