Detection of nonstationary processes
Harold A. Titus, D. Kirk, J. Arfman · 1971
A Tsunami detector is developed for a pressure signal which is measured digitally in a deep ocean capsule. Spurious data is rejected via first differencing and interpolation. Kalman filter theory is applied to the problem of the rejection of the colored noise of the tides. The detection thresholds are determined from a stochastic study of the background or sea noise from previous experiments.