Classical System Identification

Abstract System identification as a field has been around since the 1950s with roots from statistical theory. A substantial body of concepts, theory, algorithms and experience developed then. Indeed, there is very extensive literature on subject, many text books, like [5, 8, 12]. Some main points this “classical” are summarized in chapter, just pointing to basic structure problem area. The centres four pillars: (1) observed data system, (2) parametrized set candidate models, “the Model structure”, (3) an estimation method that fits model parameters (4) validation process helps taking decisions about choice structure. crucial archetypical for linear models ARX model, difference equation between system’s input output signals. This universal approximator systems—for sufficiently high orders equations, arbitrarily good descriptions system obtained. For “good” proper choices structural parameters, orders, required. An essential part classical theory deals asymptotic quality measures, bias variance, aim at giving best mean square error true system. reviewed chapter methods maximum likelihood character.