进化规划用于自适应IIR滤波器的优化设计

Evolutionary Programming for Optimizing Parametersof Adaptive IIR Filtering

针对自适应ＩＲ滤波器（ＡＩＲＦ）潜在的不稳定性和性能指标函数容易陷入局部极小点而导致性能下降等问题，本文将进化规划用于直接、并联、级联和格型结构的ＡＩＩＲＦ的优化设计。基于进化规划的自适应滤波算法不依赖于梯度信息，能够有效地实现ＡＩＩＲＦ参数的全局寻优。大量的仿真实验结果表明不同结构的ＡＩＲＦ用进化规划进行参数寻优，不仅解决了自适应滤波器性能指标函数容易陷入局部极小点的问题，亦解决了ＡＩＲＦ的稳定性问题。

Adaptive IIR filtering(AIIRF) suffers from potential instability and converges to a local minimum of the objective function. This paper proposes a evolutionary programming method for the optimizing the design of three AIIRF structures:cascade, parallel and lattice. A number of experimental results show that AIIRF based on evolutionary programming can reach the global optimization of objective function and quickly converges. Evolutionary programming for AIIRF problems provide guaranteed filter stabilty.