IIR数字滤波器设计的搜寻者优化算法

为进一步提高无限冲击响应(IIR)数字滤波器的性能,提出了一种基于搜寻者优化算法(SOA)的IIR数字滤波器设计方法.SOA基于模拟人的随机搜索行为,由利用位置变化评价得到的经验梯度确定搜索方向,由采用简单模糊规则的不确定性推理确定搜索步长,通过搜寻者在搜索空间的位置更新,实现对优化问题的求解.2个典型设计实例的仿真结果表明,与差分进化算法(DE)和3种改进的粒子群算法(PSO)相比,SOA具有较好的全局寻优能力和较快的收敛速度,能有效地应用于IIR数字滤波器的没计.

IIR Digital Filter Design via Seeker Optimization Algorithm

To further improve the performances of infinite impulse response (IIR) digital filters, a new approach based on the seeker optimization algorithm (SOA) was proposed for IIR digital filter design. The SOA is aimed to simulate the random action in human searching behaviors for solving an optimization problem through the update of seekers' positions. In the algorithm, the search direction is determined by the empirical gradients based on evaluating the responses to the changes of the seekers'positions, and the step length is decided by uncertainty reasoning based on a simple fuzzy rule. The performance of the SOA was investigated by two typical cases of IIR digital filter design. The simulation results show that compared with the differential evolution (DE) and three modified particle swarm optimization (PSO) algorithms, the proposed approach has both a good global search ability and a fast convergence speed, as a result, the SOA can be efficiently used for the design of IIR digital filters.