This is a complete, coherent description of the theory, methodology and practice of System Identification. The completely revised Second Edition introduces subspace methods, methods that utilize frequency domain data, and these key non-linear black box methods: neural networks, wavelet transforms, neuro-fuzzy modeling and hinging hyperplanes.KEY TOPICS:Leader in the field Lennart Ljung introduces systems and models, time-invariant linear systems, time-varying and nonlinear systems. He presents several approaches to system identification, including nonparametric time- and frequency-domain methods; parameter estimation; convergence and consistency; asymptotic distribution of parameter estimates; linear regressions, iterative search and recursive estimation. He also presents detailed coverage of key issues that can make or break system identification projects: defining objectives, designing experiments, selecting criteria, and controlling the bias distribution of transfer-function estimates.MARKET:For all engineering and control systems professionals, faculty and students.