This paper provides a comprehensive survey of system identification, a field that has developed rapidly in the past decade. The authors discuss the state-of-the-art in system identification, emphasizing general properties, classification of identification problems, and model structures. They highlight the importance of choosing the right model structure based on the purpose of identification and available prior knowledge. The paper covers various methods for identifying linear systems, including least squares, maximum likelihood, and instrumental variable methods. It also addresses non-linear systems, on-line and real-time identification, and the challenges associated with these areas. The authors emphasize the importance of accuracy in identification, noting that the accuracy of an identification should be judged based on the performance of the control system designed from the results. The paper also discusses the relationship between identification and control, noting that the separation hypothesis is often used in control system design. The authors conclude that identification is a crucial area in control engineering, with many applications in various fields such as biology, economy, and medicine. The paper is supported by 230 references, and includes appendices that summarize parameter estimation principles and provide an example of least squares estimation. The paper is structured into several sections, including an introduction, general properties of identification problems, classification of identification methods, choice of model structure, identification of linear systems, identification of non-linear systems, on-line and real-time identification, and conclusions. The authors also discuss the importance of computational aspects, the choice of input signals, and the criterion for identification. The paper concludes that system identification is a complex and important area of research with many applications in various fields.This paper provides a comprehensive survey of system identification, a field that has developed rapidly in the past decade. The authors discuss the state-of-the-art in system identification, emphasizing general properties, classification of identification problems, and model structures. They highlight the importance of choosing the right model structure based on the purpose of identification and available prior knowledge. The paper covers various methods for identifying linear systems, including least squares, maximum likelihood, and instrumental variable methods. It also addresses non-linear systems, on-line and real-time identification, and the challenges associated with these areas. The authors emphasize the importance of accuracy in identification, noting that the accuracy of an identification should be judged based on the performance of the control system designed from the results. The paper also discusses the relationship between identification and control, noting that the separation hypothesis is often used in control system design. The authors conclude that identification is a crucial area in control engineering, with many applications in various fields such as biology, economy, and medicine. The paper is supported by 230 references, and includes appendices that summarize parameter estimation principles and provide an example of least squares estimation. The paper is structured into several sections, including an introduction, general properties of identification problems, classification of identification methods, choice of model structure, identification of linear systems, identification of non-linear systems, on-line and real-time identification, and conclusions. The authors also discuss the importance of computational aspects, the choice of input signals, and the criterion for identification. The paper concludes that system identification is a complex and important area of research with many applications in various fields.