ESTIMATING NOISE CHARACTERISTICS FROM FLIGHT TEST DATA USING OPTIMAL FOURIER SMOOTHING

A technique based on Fourier series analysis was developed to separate signal from noise for flight test data. This was done with an optimal filter designed in the frequency domain. The method is general, and separates signal and noise based on the spectral content of the measurement time history. Smoothed time histories with no time lag were computed, and noise characteristics were accurately estimated. The technique can be used independently of other procedures, and does not require assumptions about the independence of the noise processes or the frequency content of the measurements. Simulated data was used to demonstrate the technique and to evaluate the accuracy of estimated noise characteristics. For 20 simulation cases, noise standard errors were estimated within 5% of the true values. Flight test data from a lateral maneuver of the F-18 High Alpha Research Vehicle was then analyzed. The theoretical analysis was shown to be sound for both simulated data and flight test data.