CANONICAL REPRESENTATION OF A CLASS OF LINEAR DISCRETE-TIME TIME-VARIANT SYSTEMS
dc.contributor.advisor | Wacker, A. G. | |
dc.creator | Lohar, Gautam | |
dc.date.accessioned | 2024-07-12T16:42:51Z | |
dc.date.available | 2024-07-12T16:42:51Z | |
dc.date.issued | 1985-08 | |
dc.date.submitted | August 1985 | |
dc.description.abstract | A particular class of linear discrete-time systems can be characterized by a (square) matrix of system coefficients [H]. The system is time-invariant if [H ] is circulant and time-variant if [In is non-circulant. It is shown that a linear discrete-time time-variant system can be decomposed into a combination of subsystems where each subsystem consists of a linear discrete-time time-invariant system in cascade with a modulator (canonical representation). The canonical representation results from a linear system interpretation of the following matrix decomposition : A PXP matrix [II] (complex in general) can be decomposed as P-1 [H] = [AT(i)] [H,(1)] i=0 where [H,(1)] is a circulant matrix and the diagonal matrix[AT(1)] diag {T }, with T(i) denoting the complex conjugate of the i-th column of a PXP Discrete Fourier Transform matrix. The above matrix decomposition can also be interpreted as a (two dimensional) symmetrical component decomposition of [H ]. Since a sampled and quantized image can be represented by a matrix, an application involving image representation in terms of its symmetrical components is presented. | |
dc.identifier.uri | https://hdl.handle.net/10388/15795 | |
dc.title | CANONICAL REPRESENTATION OF A CLASS OF LINEAR DISCRETE-TIME TIME-VARIANT SYSTEMS | |
dc.type.genre | Thesis | |
thesis.degree.department | Electrical Engineering | |
thesis.degree.grantor | University of Saskatchewan | en_US |
thesis.degree.level | Masters | |
thesis.degree.name | Master of Science (M.Sc.) |