Torsten Söderström Livres

Focusing on errors-in-variables (EIV) methods, this book delves into their role in system identification aimed at uncovering physical laws rather than predicting future behavior. It highlights the challenges of parameter identifiability in EIV problems and provides sufficient conditions for achieving it. The author discusses various modeling aspects, including noise characterization and extensions to multivariable and continuous-time systems. Unique solutions are presented that effectively handle noisy data, contrasting with traditional methods like total least squares.

Discrete-time Stochastic Systems offers a thorough introduction to the estimation and control of dynamic stochastic systems, covering Wiener filtering, state-space methods, and polynomial approaches. It includes algorithms for analysis, spectral factorization, and complex-valued models for signal processing, making it valuable for M.Sc. students and self-study.