平稳时间序列的Hartley变换谱估计及其在遥测系统中的应用

本文给出遥测信号数据相关函数和功率谱的计算方法——快速Hartley变换(FHT),用之取代快速Fourier(FFT)算法。首先研讨了FHT法,继而讨论相关和功率谱估计、自/互功率谱估计并对两种算法的计算量作了对比。结果表明,随机数据处理软件系统对一系列伪噪声功率谱的测试结果在统计意义下与理论值一致。从而得出结论:用相关函数法对实型数据进行功率谱估计,采用FHT比采用FFT更有效。

Stationary Series Spectral Estimation with FHT and it's Telemetry Application

A Telemetry signal processing system is realized with FORTRAN in AST-286. The functions imclude linear trand removal, stationary test and caculation of correlation, spectrum, transfer function as well as coherence. The sampling data of telemetry sinai is analysed as a stationary times series, if it fits sampling therom and passes stationary test. The algorithm of correlation and spectrum caculating with Fast Hartley Transform(FHT) replacing Fast Fourier Transform (FFT) is deduced in the paper. The testing results of random data spectrum is statistically equal to the theoreieal values. The conclusion is that the caculating of corrolation and spectrum using FHT is more efficient than that using FFT.