2.687  Time Series Analysis and System Identification

Matched filtering, power spectral (PSD) estimation and adaptive signal processing/ system identification algorithms are covered.  Algorithm development is framed as an optimization problem and optimal and approximate solutions are described.  Time-varying systems, first and second moment representations of stochastic processes, and state-space models are reviewed.  Algorithm derivation, performance analysis and robustness to modeling errors are covered.  Algorithms for PSD estimation, the LMS and RLS algorithms, and the Kalman Filter are treated in detail.