Optimal design of fractional delay FIR filter using cuckoo search algorithm

Abstract

The conventional gradient‐based optimization methods are not sufficient to optimize nonlinear, multimodal, and nonuniform objective functions of fractional delay FIR (FD‐FIR) filters, and the objective function cannot converge to the global minimum solution. So a population‐based meta‐heuristic optimization algorithm called as cuckoo search algorithm (CSA) has been implemented in the design of optimal FD‐FIR filter. Cuckoo search algorithm is based on lifestyle and unique parasitic behavior in egg laying and breeding of some cuckoo species along with Lévy flight behavior of some birds and fruit flies. To attain a balance between exploration and exploitation in the search space, different set of control parameters is tested by simulation. Extensive simulations were performed to ensure how CSA exploits in the design of optimal FD‐FIR filter. A quantitative assessment of the proposed CSA‐based method is accomplished using several performance metric such as magnitude error, convergence rate, and optimal solution. The simulation results reveal the advantages of the proposed FD filter using CSA compared with the FD filter designed using evolutionary algorithm like genetic algorithm and conventional FD filter design methods such as Lagrange, discrete Hartley transform, discrete Fourier transform, discrete cosine transform, and radial basis function methods