An Iterakive Method for the Design of FIR Part i a1 Filter Banks
Weigang Lu, A. Antoniou · 1997
An iterative algorithm for the de- sign of FIR partial filter banks (PFBs) is de- scribed. The design piroblem is formulated as a 4th-order nonlinear optimization problem in which the objective function is a weighted sum of a reconstruction-error term and an aliasing-error term. The optimization :problem is solved by iter- atively minimizing a paiir of quadratic functions. Explicit formulas for evalluating the minimums of these quadratic functions are derived, which lead to an efficient and fast algorithm. Two design ex- amples are presented to illustrate the proposed algorithm. The PFBs obtained are then used to compress bathymetric signals and electrocardio- grams.