【摘要】 详细讨论了神经网络在FIR线性相位带阻滤波器中的应用。用神经网络方法设计的线性相位带阻滤波器 ,其幅频响应在通带与阻带范围均无过冲与波动现象 ,衰减特性好 ,在 10 0dB以上。给出了FIR线性相位带阻滤波器的设计实例 ,计算结果表明 ,该算法在FIR线性相位滤波器设计中非常有效 ,具有优异性能。更多还原

【Abstract】 We discusses the application of the neural networks in the field of FIR(finite impulse response) band-block filters with linear phase. The linear-phase band-block filters are designed by the neural networks, which characteristic of the amplitute-frequency is no excess-impulsion and ripple. Its attenuation performance is very good and over 100dB in the block-band. The example of the FIR band-block filter with a lineary-phase is presented. The result shows that the algorithm is efficient and excellent in the design of the FIR filters.更多还原