Design of FIR filters with arbitrary amplitude and phase specifications using genetic algorithm

In this paper we present efficient design algorithm for FIR filters with arbitrary amplitude and phase specifications using genetic algorithm. This algorithm provides flexibility in control the genetic algorithms parameters such as population size, number of generations, crossover probability and so on. Filters with binary, integer or real coefficients can be easily handled by the proposed method. This includes the multiplierless coefficients case. It has been tested for the design of filters with different amplitude and phase specifications. The fitness function was defined as the simple reciprocal of sum squared of the error.