Abstract:

At present, the demand for solving analysis, synthesis, modeling and processing multidimensional signals tasks is high. It is connected with developments in areas such as medical equipment, telecommunication, automotive industry, finance, which require real-time processing and compression of changing multidimensional signals.
Simulation modeling is an important tool for research on objects and processes. The problem of testing technical systems that process multidimensional signals has become more relevant with increasing quantity and difficulty of such kind of systems in our world recently. This problem is more fully solved with the use of simulation modeling.
Traditional teaching methodologies for the "Simulation modeling of digital signals in real-time" course usually include theoretical methods using "dry" mathematical apparatus. As a result, the perception and learning of the course are difficult. Usually, the practical use of simulation signals modeling hasn't been told. Additionally, less attention has been paid to enhancing practical skills in simulation signals modeling.
A detailed analysis of the discipline structure was done from the students' perception viewpoint. This viewpoint has formed the basis of teaching simulation modeling of signals to master's students. The simulation modeling signals methodology consists of unique methods developed by the authors of this article. It utilizes neural network architecture and clustering methods.
The introduced methodology has been applied to the BMSTU magistracy in the following departments: Computer systems and networks, biomedical systems, and electric vehicles for the last 3 years. Semester tests and feedback surveys have shown the following results: the use of this methodology has improved the perceived quality of studying materials by an average of 20%, and the level of interest in this course has increased by 35% compared to the average values of the previous years.

Keywords:

Real-time, simulation modelling, multidimensional signals, neural network, clustering methods.