一种新的中频采样滤波器的设计与实现

许多信号处理算法是针对复信号的,因此,从实带通信号中产生复信号是信号处理中的一个重要问题.中频采样技术是对中频信号直接采集产生正交复基带信号的技术,一般是采用2b或4b采样频率(b为信号带宽),而且滤波器的系数是以表格形式给出的,因此使用者不能根据实际情况对滤波器加以修正.本文提出了一种新的滤波器设计方法,它没有限定采样频率和信号带宽之间的关系,换句话说,信号的归一化带宽是可变的.此外,新滤波器的系数是以解析形式表示的,可以方便地根据实际信号的带宽计算出最佳滤波器的系数.当信号带宽较窄时,新滤波器的性能明显优于现有滤波器.文章的最后介绍了一种基于SHARC的中频采样的实现方案.该方案已经用于实际系统,取得了很好效果.

Sinusoid Frequency Estimation in Colored Noise Based on Nonlinear Prefiltering in Frequency Domain

Based on the observation that sinusoids show themselves as outliers in frequency domain,this paper proposes a new approach for the superresolution frequency estimation of sinusoids in an unknown colored noise environment by utilizing the capability of running median filters to remove outliers from data.The outstanding advantages of the method are its strong adaptability to various noise distributions and the simplicity of its implementing algorithm.Simulations are provided to show its performances.