Optimal Design of FIR Notch Filters Using Neural-Network Algorithm

摘要

A novel optimal design approach for linear phase finite impulse response (FIR) notch filters is proposed based on a new parallel neural networks algorithm (FNNA), the main idea is to minimize the squared-error function in the frequency-domain. By using the PNNA, the coefficients of the designed notch filter can be obtained directly from the specified magnitude response. The convergence theorem is presented and proved to illustrate the stability of the PNNA. Finally, some examples are given to illustrate the effectiveness of the proposed linear phase FIR notch filters design method.