Title :
Design of Linear-Phase FIR Multiple-Notch Filters via an Iterative Reweighted OMP Scheme
Author :
Wei Xu ; Jiaxiang Zhao ; Chao Gu
Author_Institution :
Sch. of Electron. & Inf. Eng., Tianjin Polytech. Univ., Tianjin, China
Abstract :
This brief considers the design of sparse linear-phase finite-impulse-response multiple-notch filters whose notch frequencies are arbitrarily spaced across the spectrum [0, π]. The proposed scheme, i.e., iteratively reweighted orthogonal matching pursuit (IROMP), is based on the orthogonal matching pursuit performed under the weighted l2-norm whose weights are iteratively computed through the hybrid l1/l2-norm minimization. At each iteration, the IROMP is employed to yield the filter whose response best approximates the desired filter. The computed filter is then modified through solving a linear programming problem. The iteration stops when the modified filter meets the design specifications. The simulation demonstrates the potentials of the IROMP scheme.
Keywords :
FIR filters; iterative methods; linear phase filters; linear programming; notch filters; hybrid l1/l2-norm minimization; iterative reweighted orthogonal matching pursuit; linear phase FIR multiple notch filters; linear programming problem; Attenuation; Bandwidth; Finite impulse response filters; Matching pursuit algorithms; Passband; Vectors; Hybrid $l_{1}/l_{2}$-norm minimization; iteratively reweighted orthogonal matching pursuit (IROMP); linear phase; multiple-notch filter; sparse;
Journal_Title :
Circuits and Systems II: Express Briefs, IEEE Transactions on
DOI :
10.1109/TCSII.2014.2345299