Analyzing large quantities of physical data during transient or other changing state conditions requires special treatment if results are to be useful. In this paper, results are derived for the accuracy to be associated with average response computations on a wide class of nonstationary data as a function of underlying signal and noise variations and the available sample size. Input and output signal-to-noise ratios are defined in terms of these quantities. Confidence bands are determined for both arbitrary probability distributions and for Gaussian probability distributions. A special result shows the increase in sample sizes required to analyze unknown distributions as opposed to Gaussian distributions in order to achieve a desired confidence band.
Methods for Analyzing Nonstationary Data
Sae Technical Papers
Pre-1964 SAE Technical Papers ; 1906
1962-01-01
Conference paper
English
Analysis of Nonstationary Vibration Data
NTIS | 1987
|The analysis of nonstationary vibration data
NTRS | 1987
|Spectral analysis of nonstationary spacecraft vibration data
TIBKAT | 1965
|Missile structural loads by nonstationary statistical methods
Engineering Index Backfile | 1961
|