A fast design algorithm for elliptic-error and phase-error constrained LS 2-D FIR filters

Abstract

Two-dimensional (2-D) nonlinear-phase finite impulse response (FIR) filters have found many applications in signal processing and communication systems. This paper considers the elliptic-error and phase-error constrained least-squares design of 2-D nonlinear-phase FIR filters, and develops a matrix-based algorithm to solve the design problem directly for the filter's coefficient matrix rather than vectorizing it first as in the conventional methods. The matrix-based algorithm makes the design to consume much less design time than existing algorithms. Design examples and comparisons with existing methods demonstrate the effectiveness and high efficiency of the proposed design method.

A matrix-based IRLS algorithm for the least $${l}_{p}$$ l p -norm design of 2-D FIR filters

“…On the other hand, several matrix-based design algorithms (Hsieh et al (1997); Aravena and Gu (1996); Zhu et al (1999); Lai (2011, 2013); Hong et al (2013Hong et al ( , 2016; Zhao et al (2016)) have been proposed for designing 2-D filters. They formulate the design problem and solve for the coefficients of 2-D filters in their natural matrix forms.…”