Digital Filter Design Based On Intelligent Optimization Algorithm

The optimal design of the digital filter is a multi-parameter optimization problem, now a lot of existing algorithm cannot very good out the best solution. Improved in this paper, a novel have rapid global search algorithm of swarm intelligence - Cultural Colony (Cultural Bee Colony, CBC) algorithm and its application in the digital filter.This paper FIR and IIR digital filter was deduced in detail the specific optimization design model and the objective function. Then the Particle Swarm (Particle Swarm Optimization, PSO) Algorithm, the Quantum Particle Swarm (Quantum - behaved Particle Swarm Optimization, QPSO) Algorithm, Artificial Colony Algorithm (Artificial Bee Colony,ABC), cultural Algorithm (Culture Algorithm, CA)) to represent sets forth the principle of classic Swarm intelligence Algorithm, and applied to digital filter design, simulation comparison of the classic filter design and performance of the Algorithm.For optimizing the design of digital filter based on ABC algorithm shortcoming, double evolution both CA algorithm was proposed and the advantages of the ABC algorithm three complementary advantages search mechanism, with the help of the CBC algorithm of reverse learning mechanism. After establishing CBC algorithm framework, and gives out the CBC algorithm was applied to digital filter optimization design of the specific process and the simulation comparison with classical algorithm performance difference.The simulation experiment results show that the CBC algorithm in the design of FIR digital filter, its performance is better than PSO algorithm and QPSO algorithm, the CA algorithm and the design of the ABC algorithm; For optimal design of IIR digital filter, the performance of the CBC algorithm is better than PSO algorithm and QPSO algorithm, the CA algorithm.At the same time, on the basis of the FPGA implementation of the project, design the hardware digital FIR filter in a distributed algorithm, using the filter coefficient of CBC optimization algorithm, with good QPSO filter coefficient filtering effect of contrast, show that the improved algorithm's CBC superior effect.