Title :
Optimal design of two-dimensioal FIR-filters
Author_Institution :
University of Boon, Boon, FRG
Abstract :
The design problem of two-dimensional FIR-filters is solved by quadratic programming. It is shown, that the add of inequalities to the L2-norm method (least-squares) tends to the L∞-norm method (Chebyshev or minimax) depending On the frequency bands where inequalities have been formulated. An Example of optimal and suboptimal design will illustrate the method proposed in this paper.
Keywords :
Frequency domain analysis; Frequency response; Lagrangian functions; Matrices; Nonlinear filters; Sampling methods; Vectors;
Conference_Titel :
Acoustics, Speech, and Signal Processing, IEEE International Conference on ICASSP '83.
DOI :
10.1109/ICASSP.1983.1172152
