Information processing is an important factor in various fields such as navigation, industry, agriculture, transport, communications, trade etc. The concept of an information processor includes measuring and acquisition system, data and signal processor, and measurement and converter systems for the information transmission in explicit form in the real world. Functional design of signal processors, as a part of an information device, is based on estimation and classification theory. The main difference between these two areas is the information type obtained as a result of processing. The classification output is discrete, i.e. a class, feature or category. In estimation problems that is real scalar or vector variable. Since such problems occur in both static and dynamic environment, estimation problem is used for dynamic events, which may be continuous or discrete in time. The similarity between these two areas allows unique methodology based on Bayes decision theory to be used. The course covers the mathematical foundations of this theory, and special emphasis will be placed to the practical aspects of the theoretical results. In the first part of the course the classification and estimation theory in the case of static and dynamic models which exactly and adequately describe the physical process will be considered. In the second part course will consider more realistic situations in which the model of the process is not fully understood and there is certain uncertainty or unmodeled dynamics. These models are obtained from experimental data or experimental data are directly used for classification and estimation algorithm training. Areas of application for this kind of methodology are different and include mechanical engineering, electrical engineering, construction management, engineering processes, environmental engineering, etc.