Design of a Multiple-Valued Systolic System for the Computation of the Chrestenson Spectrum
IEEE Transactions on Computers
Digital filter design
An Architecture for a Video Rate Two-Dimensional Fast Fourier Transform Processor
IEEE Transactions on Computers
Spectral Techniques and Fault Detection
Spectral Techniques and Fault Detection
IEEE Transactions on Computers
Generalized transforms for DSP and generalized spectral analysis
Systems Analysis Modelling Simulation - Special issue: Digital signal processing and control
| Hi-index | 14.98 |
A novel generalized spectral analysis approach which applies a weighting to z-transform spectra evaluated on contours in the z-transform plane is proposed. A parallel algorithm and 1D and 2D parallel processor architectures for the estimation of the pole-zero mathematical model of a system from a truncated version of its impulse response by succesive parallel evaluations of the proposed weighted z-transform spectra are subsequently presented. The algorithm is applicable to system identification and digital filter synthesis. The proposed weighted z-transform spectra and associated energy spectra make possible the evaluation of the poles and zeros of an infinite impulse response system of which the order is unknown from a truncated version of its impulse response with reasonable accuracy. A parallel dynamic weighting of z-transform spectra that is a function of |z| is shown to overcome the effect of exponential divergence of the z-transform of finite duration sequences as the z-plane center is approached. The resulting "bi-dimensional spectral decomposition," in terms of both hyperbolic and circular function content, is in contrast with that of Fourier and z-transforms which for finite duration sequences are shown to effect a decomposition in terms of only circular function content. Using parallel fast transform operations the proposed algorithm unmasks pole-zero peaks and through interpolation estimates the associated residues. The corresponding time-sequence component is deleted, the process repeated and the zeros determined using the evaluated residues. Optimal parallel and pipelined 1D- and 2D-type processor architectures are proposed, leading to the possibility of on line adaptive modeling of fast varying systems.