Design of Quadrature Mirror Filter bank using constrained optimisation

Abstract

In this paper, a constrained optimisation technique is used for designing a two-channel Quadrature Mirror Filter Bank (QMFR). A new method is developed based on ideal characteristics of a prototype filter and overall response of the filter bank in transition band. The design problem is formulated as nonlinear constrained optimisation of an objective function, which is weighted sum of square of error in passband, stop band and overall filter bank response at frequency (?= 0.5?). For solving optimisation problem, Sequential Quadratic Programming (SQP) optimisation technique is used. The simulation results included illustrate the superiority of the proposed method over the other techniques terms of reconstruction error, mean square error in passband, stopband and transition band. Copyright © 2009 Inderscience Enterprises Ltd.