基于粒子群算法的数字滤波器优化与仿真

Digital Filter Optimization Based on Momentum Crossover Particle Swarm Optimization Algorithm

研究数字滤波器优化问题,针对传统算法在数字滤波器优化过程中易出现"早熟"和后期收敛速度慢等等问题,提出了一种动量交叉粒子群算法的数字滤波器优化方法。首先把求解数字滤波器参数的问题数学化为性能指标优化模型,然后采用动量交叉粒子群算法找到符合特征要求的数字滤波器参数值,并通过仿真对性能进行测试。仿真结果表明,动量交叉粒子群算法较好地解决了传统算法的易出现"早熟"和后期收敛速度慢等等难题,设计数字滤波器的频域响应十分逼近理想频域响应,提高数字滤波器的设计效率。

This paper proposed a digital filter method based on momentum crossover particle swarm optimization algorithm.Firstly,the mathematical problem of digital filter was taken as optimization objection,and then the momentum crossover particle swarm algorithm was used to find parameters which met with the features of digital filter,the simulation experiments were carried out to test the performance.The simulation results show that the proposed algorithm can effectively solve the defects of the traditional algorithms,and the frequency response of the designed digital filter is very ideal and improves the design efficiency of digital filter.