论文部分内容阅读
一.前言 多年来,估算平稳随机过程的功率谱常用的方法是FTA(自相关函数的福氏变换)法和FFT(快速福氏变换)法.FTA法,是通过随机过程样本序列的自相关函数乘以某些窗函数,然后经过福氏变换及某种平滑而得到功率谱估计值.FFT法,是通过样本序列乘以某种数据窗,经过快速福氏变换和频率平滑而得到功率谱估计值.这两种谱分析法实际上对样本序列作了如下不切实际的假定:(1)过程是以样本长度为其周期之一的周
I. INTRODUCTION Over the years, commonly used methods for estimating the power spectrum of a stationary stochastic process have been the FTA (Freunds Transformation) method and the FFT (Fast Fourier Transform) method. The FTA method is based on the autocorrelation of random process sample sequences The function is multiplied by some window functions, and then Fourier transform and some smoothing to get the power spectrum estimation.FFT method, through the sample sequence multiplied by a data window, after fast Fourier transform and frequency smoothing to get the power spectrum The two spectral analyzes actually make the following unrealistic assumptions about the sample sequence: (1) The process is based on a sample week whose length is one of its periods