The paper "System Identification: A Survey" by Åström and Eykhoff provides a comprehensive overview of the field of system identification and process parameter estimation, which has seen rapid development over the past decade. The authors present the state-of-the-art in a systematic manner, focusing on general properties and the classification of identification problems. They discuss various model structures, emphasizing that the choice of model depends on the purpose of identification and available prior knowledge. For linear parameter models, the paper covers the most commonly used least squares methods and their variants, such as repeated and generalized least squares, maximum likelihood, instrumental variable, and tally principles. It also addresses the challenges of nonlinear identification, online and real-time identification, and recent developments in these areas. The authors highlight the importance of planning experiments and selecting appropriate input signals to achieve realistic models. They discuss the limitations of identification methods and the need for a unified framework to treat identification problems. The paper includes a detailed exposition of least squares estimation and an example to illustrate the concepts. The paper is structured into several sections, including an introduction, a discussion of general properties of identification problems, a classification of identification methods, and a section on the choice of model structure. It concludes with a summary of the key principles and a detailed example of least squares estimation. The authors emphasize the importance of identifying systems for control engineering applications and the need to balance the accuracy of identification with the practical constraints of experimental design.The paper "System Identification: A Survey" by Åström and Eykhoff provides a comprehensive overview of the field of system identification and process parameter estimation, which has seen rapid development over the past decade. The authors present the state-of-the-art in a systematic manner, focusing on general properties and the classification of identification problems. They discuss various model structures, emphasizing that the choice of model depends on the purpose of identification and available prior knowledge. For linear parameter models, the paper covers the most commonly used least squares methods and their variants, such as repeated and generalized least squares, maximum likelihood, instrumental variable, and tally principles. It also addresses the challenges of nonlinear identification, online and real-time identification, and recent developments in these areas. The authors highlight the importance of planning experiments and selecting appropriate input signals to achieve realistic models. They discuss the limitations of identification methods and the need for a unified framework to treat identification problems. The paper includes a detailed exposition of least squares estimation and an example to illustrate the concepts. The paper is structured into several sections, including an introduction, a discussion of general properties of identification problems, a classification of identification methods, and a section on the choice of model structure. It concludes with a summary of the key principles and a detailed example of least squares estimation. The authors emphasize the importance of identifying systems for control engineering applications and the need to balance the accuracy of identification with the practical constraints of experimental design.