论文部分内容阅读
ＦＩＲ数字滤波器（ＦＩＲＤＦ）软件设计方法一般采用Ｆｏｕｒｉｅｒ级数法、频率抽样法和切比雪夫最佳一致逼近法．由于常规ＦＩＲＤＦ设计方法的自身特点，要取得通带、阻带和过渡带性能都最优的滤波器很困难，并且难以实现灵活可调的多通带（阻带）线性相位滤波器．文中分析了ＦＩＲＤＦ的特点之后，根据小波理论中连续小波变换的性质和函数逼近理论，提出了ＦＩＲＤＦ设计的新方法——小波逼近方法．利用多个小波函数来逼近理想滤波器的频率特性．结果表明，ＦＩＲＤＦ设计的小波方法比切比雪夫逼近法和离散小波变换共轭镜象滤波器方法，其计算过程更灵活，带宽选择更方便，计算速度更快．根据不同的滤波要求，很容易设计多通带的带通滤波器和多通带移相器．
FIR digital filter (FIRDF) software design methods commonly used Fourier series method, frequency sampling method and Chebyshev best consistent approach. Due to the inherent characteristics of the conventional FIRDF design method, it is difficult to obtain the optimal filter in the passband, stopband and transition band, and it is difficult to realize a flexible and adjustable multipass (stopband) linear phase filter. After analyzing the characteristics of FIRDF, a new method of FIRDF - wavelet approximation is proposed based on the properties of continuous wavelet transform in wavelet theory and the function approximation theory. Using multiple wavelet functions to approximate the ideal filter ’s frequency characteristics. The results show that the FIRDF wavelet method is more flexible than the Chebyshev approximation method and the discrete wavelet transform conjugate image filter method, the bandwidth selection is more convenient and the calculation speed is faster. According to different filtering requirements, it is easy to design a multi-pass bandpass filter and multi-passband phase shifter.