ΜΚ13 | Identification and Estimation of Stochastic Systems

Postgraduate Study Program | Design and Production

Theοry and practical application of the modeling and analysis of stochastic signals and systems. The need for information extraction from and mathematical modeling of measured signals. Overview of the basic concepts of stochastic signals and systems in the time and frequency (Fourier) domains: moments, stationarity, ergodicity, spectral representations. Transfer function models, including ARX, ARMAX, and Box-Jenkins models. State space models. Vector models, uniqueness, identifiability, canonical and pseudo-canonical forms. Prediction theory and the Kalman Filter. Estimators and their properties. Least Squares, Prediction Error and Maximum Likelihood estimation methods. Experiment design. Time-recursive and adaptive estimation methods. Model structure identification. Model validation. Practical mechanical and aeronautical engineering examples using MATLAB.