The design of FIR filter based on APA Genetic Algorithms

In a given performance requirements, Frequency sampling FIR filter design's goal is as far as possible approximate the ideal filter to reduce the approximation error. Genetic algorithms is a mimic biological evolution global optimization search algorithm, which has strong global search ability and has been successfully applied to a variety of optimization problems, in this paper the frequency sampling FIR filter design method is also an optimization problem in essence. In this paper, the GA's parameters was improved based on the procedure of evolutionary so that to prevent premature and improve the speed of convergence, the improved genetic algorithm(APA Genetic Algorithms) was applied in the frequency sampling FIR filter design. Simulation results show that, in the same design requirements, the AGAPA method of design the low- and high-pass FIR is much more advantages than the traditional method and the reference look-up table.