论文部分内容阅读
研究了FIR线性相位滤波器的幅频特性与余弦基函数神经网络算法之间的关系,给出了FIR高阶多带阻数字滤波器的优化设计实例,提出并证明了神经网络算法的收敛性定理.仿真结果表明,算法不仅是有效的,而且在FIR高阶数字滤波器优化设计领域具有很大的优越性.且算法不涉及逆矩阵的计算,有效解决了国内外其它优化设计方法无法设计高阶FIR线性相位滤波器的问题.
The relationship between the amplitude-frequency characteristic of the FIR linear phase filter and the algorithm of cosine basis function neural network is studied. The optimal design example of the FIR high-order multi-band-stop digital filter is given. The convergence of the neural network algorithm is proposed and proved Theorem.The simulation results show that the algorithm is not only effective but also has great advantages in the field of FIR high-order digital filter optimization design.The algorithm does not involve the calculation of inverse matrix and effectively solves the problem that other optimization methods at home and abroad can not be designed High order FIR linear phase filter problem.