ⅡR DF设计的高阶关联神经网络优化方法

Designing ⅡR Digital Filter by Using a High-order Neural Network-based Optimization Technique

  • 摘要: 基于全反馈高阶关联神经网络优化理论,提出了一种ⅡR数字滤波器神经网络优化设计方法(ⅡR-NNO)的理论框架。表征设计质量的加权均方误差当作神经网络能量函数,导出了ⅡR-NNO的Lyapunov方法。简要说明了该方法计算机实现的基本原则,通过设计实例及与其他方法的比较,证明了它的有效性。

     

    Abstract: Recently,neural network optimization (NNO) technique has been proved to be an efficient approach for designing FIR filters.This paper makes an attemp to generalize the NNO technique to design ⅡR digital filters.For this aim a high order neural network (HONN) is used to form the Lyapunov function which corresponds to the MSE in designing the ⅡR.The convergency of the algorithm is addressed and an LP ⅡR simulation example is given to verify the effectiveness of the NNO design technique.

     

/

返回文章
返回