The optimal design method of FIR filter using the improved genetic algorithm

The design target of Finite Impulse Response (FIR) filter is to approximate the ideal filters on the request of a given designing filter specifications. Genetic algorithm is one of global optimal search algorithm of mimic biological evolution, but it has the precocious and slow convergence problems if it is directly applicated. Based on these problems, the adaptive selection algorithm was used to improve the choice of the crossover operator and its mutation operator for genetic algorithm. And then, the improved genetic algorithm was used to design and optimize the Finite Impulse Response (FIR) filter. Based on design features of FIR filter frequency sampling method, we used the improved genetic algorithm, Look-up table and references methods for compare their results. The results of the proposed method by MATLAB simulation showd that the proposed method can obtain better results in the same design specifications.