基于差分文化算法的FIR数字滤波器设计

有限脉冲响应（FIR）数字滤波器的设计实质可看作是多参数优化问题。为高效实现FIR数字滤波器,将滤波器的设计转化为滤波器参数优化问题,然后提出差分文化粒子群（DC）算法在参数空间进行并行搜索以获得滤波器设计的最优参数值。提出的差分文化算法结合文化原理差分演进原理,是一种可用于实数优化的多维搜索算法。计算机仿真实验表明在设计FIR数字滤波器设计时,差分文化算法的收敛速度和性能都优于粒子群,量子粒子群以及自适应量子粒子群优化等算法,证明了该方法的有效性和优越性。

FIR Digital Filters Design Based on Differential Cultural Algorithm

The digital filter design is a multi-parameter optimization problem.This paper presents a joint objective function to design finite impulse response（FIR） digital filters,and a differential cultural algorithm is proposed to implement filter designs.The differential cultural algorithm is used to search optimal value of filter design parameters in the parameter space with parallel search.The differential cultural algorithm is a multi-dimensional search algorithm for optimization of real number,mechanisms of the quantum and cultural evolution are used to update the quantum population space.Computer simulations show that FIR digital filters based on DC is superior to filters based on particle swarm optimization（PSO）,quantum particle swarm optimization（QPSO） and adaptive quantum particle swarm optimization（AQPSO） in convergent speed and the optimization results.The effectiveness and superiority of DC are also demonstrated by simulations.