疯狂蝙蝠算法的低通FIR滤波器设计

针对处理滤波器设计问题，蝙蝠算法（BA）很容易陷入早熟收敛，提出改进的疯狂蝙蝠算法（CBA）旨在设计低通有限脉冲响应（FIR）滤波器。首先，在频率更新中引入两个随机数来控制全局与局部的平衡；其次，在标准蝙蝠算法中引入疯狂因子，使得蝙蝠算法以一定疯狂概率维持其多样性；最后，在局部搜索时，对算法的响度和脉冲发射率进行动态修正，便于精细搜索。将CBA与其他算法进行比较，仿真结果表明，疯狂蝙蝠算法在应用实例中优于其他算法，且能够有效、稳定用于低通FIR滤波器设计。

Design of low-pass FIR filter for craziness bat algorithm

In order to deal with the problem that the filter design and bat algorithm(BA) is prone to premature convergence, this paper proposed an improved craziness bat algorithm(CBA) to design a low-pass finite impulse response(FIR) filter. Firstly, it introduced two random numbers to control the global and local equilibrium in the frequency update. Secondly, the bat algorithm could maintain its diversity with a certain crazy probability by introducing the crazy factor in the standard bat algorithm. Finally, the loudness and pulse emissivity of the algorithm were modified dynamically in the local search, which was convenient for fine searching. Compared with other algorithm, simulation experiments show that the craziness bat algorithm is superior to other algorithms in the application and can be effectively and stably used for low-pass FIR filter design.