Design of Iir Filters Having Specified Magnitude and Group Delay Characteristics Using Evolutionary Immune Algorithm

This paper presents a new method based on antigen-antibody reaction of immune system in biology, has been adopted for the design of IIR digital filter using an evolutionary Tagushi Immune Algorithm (TIA). The design process is reduced to a constrained optimization problem the solution of which is achieved by the convergence of the TIA which implements a learning technique inspired by human immune system. The main feature of the immune algorithm is its ability to find global optimal solution in a nonlinear search space with stability guaranteed. In the design process, all the filter coefficients are collected as antibodies, and a population of these antibodies is evolved by using immune algorithm operations of reproduction, and clonal proliferation within hyper-mutation. Taguashi quality control technique has been adopted for a better selection of antibodies to generate more excellent offsprings. It is applied to the design of IIR digital filters exhibiting either specified time-domain or specified magnitude and group delay characteristics. Illustrative examples are presented to verify the effectiveness of the proposed method compared to the existing hierarchical genetic [22]-[23], and immune based algorithm [24].