基于智能优化算法的FIR数字滤波器设计  

Design of FIR Digital Filter Based on Intelligent Optimization Algorithm

作　　者：王伟 冉昌艳 李维文[3] 祝执 WANG Wei;RAN Chang-yan;LI Wei-wen;ZHU Zhi(Hubei Key Laboratory of Intelligent Vision Monitoring for Hydropower Engineering,China Three Gorges University,Yichang Hubei 43002,China;Yichang Key Laboratory of Intelligent Vision Monitoring for Hydropower Engineering,China Three Gorges University,Yichang Hubei 43002,China;College of Computer and Information Technology,China Three Gorges University,Yichang Hubei 43002,China;Electronic and Communication Institute,China Three Gorges University,Yichang Hubei 443002,China)

机构地区：[1]三峡大学湖北省水电工程智能视觉监测重点实验室,湖北宜昌443002 [2]三峡大学水电工程智能视觉监测宜昌市重点实验室,湖北宜昌443002 [3]三峡大学计算机与信息学院,湖北宜昌443002 [4]三峡大学电子与通信研究所,湖北宜昌443002

出　　处：《计算机仿真》2023年第12期416-421,528,共7页Computer Simulation

基　　金：湖北省水电工程智能视觉检测重点实验室开放基金(2020SDSJ07);水电工程智能视觉监测湖北省重点实验室建设(2019ZYYD007);宜昌市自然科学研究项目(A20-3-004);“一流课程建设”背景下数字信号处理实验课程教学改革研究(J2021022)。

摘　　要：使用频率采样法设计FIR数字滤波器时,在通带与阻带之间设置适当的非零过渡点可极大改善滤波器的幅度特性,但过渡点最佳采样值难以确定,可以使用智能优化算法解决上述问题。在设置一个,两个和三个过渡点时,使用遗传算法、粒子群算法和模拟退火算法,在最大误差最小,阻带最大误差最小,最小二乘以及阻带最小二乘准则下,分别求解过渡点最佳采样值,并使用五种评价指标对比不同算法在不同滤波器优化准则下的优化效果。实验结果表明,在设置不同数量过渡点的情况下,使用智能优化算法均可得到阻带幅度特性优于查找表数据的过渡点最佳采样值,三种算法中模拟退火算法优化效果最佳,四种滤波器优化准则中阻带最大误差最小准则下的优化效果最佳。When designing FIR digital filters using the frequency sampling method,the amplitude characteristics of the filters can be greatly improved by appropriately setting non-zero transition samples between the passband and the stopband.However,it is difficult to determine the optimal transition coefficients,and the problem can be solved using intelligent optimization algorithms.The Genetic Algorithm(GA),Particle Swarm Optimization(PSO)and Simu⁃lated Annealing(SA)algorithms were respectively used for one,two and three transition samples.The Minimax(MM),Stop Band Minimax(SBMM),Least Squares(LS),and Stop Band Least Squares(SBLS)optimization criteria were applied to determine the optimal transition coefficients,and five evaluation indicators were also used to compare the optimization effects of different algorithms under different criteria.The experimental results show that the optimal transition coefficients which have better stop band amplitude characteristics than the lookup table data can be obtained using the intelligent optimization algorithms under the condition of different number of transition samples,and the best optimization effect was achieved by the SA algorithm among the three algorithms,and the best optimiza⁃tion effect was achieved by the SBMM criterion among the four filter optimization criteria.

关 键 词：有限冲激响应滤波器 频率采样法 遗传算法 粒子群优化 模拟退火算法 

分 类 号：TN713[电子电信—电路与系统]