A Hybrid Differential Evolution Method for the Design of IIR Digital Filter

This paper establishes methodology for the robust and stable design of infinite impulse response (IIR) digital filters using hybrid differential evolution method. Differential Evolution (DE) is undertaken as a global search technique and exploratory search is exploited as a local search technique. DE is a population based stochastic real parameter optimization technique relating to evolutionary computation, whose simple yet powerful and straight forward features make it very attractive for numerical optimization. Exploratory search aims to fine tune the solution locally in promising search area. This proposed DE method augments the capability to explore and exploit the search space locally as well globally to achieve the optimal filter design parameters by applying the opposition learning strategy and random migration. A multivariable optimization is employed as the design criterion to obtain the optimal stable IIR filter that minimizes the magnitude approximation error and ripple magnitude. DE method is implemented to design low-pass, high-pass, bandpass, and band-stop digital IIR filters. The achieved design of IIR digital filters by applying DE method authenticates that its results are comparable to other algorithms and can be effectively applied for higher filter design.