Design of 2-D FIR digital filters with specified magnitude and group delay responses by GA approach

An effective genetic algorithm (GA) approach is proposed for designing two-dimensional (2-D) real finite-duration impulse response (FIR) digital filters with complex-valued frequency responses. The method is extended from the one-dimensional (1-D) filter designs. By minimizing a quadratic measure of the error in the frequency band, the real-valued chromosomes of a population are evolved to get the filter coefficients. It is used to design two major classes of 2-D FIR digital filters including multi-band filters and all-pass phase equalizers. Two examples are presented to demonstrate the efficiency and effectiveness of this proposed approach.