Method of automated identification of group air trajectories on the basis of system analysis and information processing in military automation complexes

The paper presents different methods for solving the prob-lem of target identification in radar information processing systems. Bayesian and non-Bayesian approaches are ana-lyzed: the first ones use a priori information about the probabilities of identification hypotheses, taking into ac-count both current data and previous knowledge; the sec-ond ones are based solely on current observations, which improves performance at the expense of less information. The JV method and its extended versions, such as JVC, are newer trajectory processing approaches that take into account both static and dynamic aspects of the identifica-tion problem. These methods allow for a more flexible ap-proach to problem solving, especially in situations where real-time data is available and decisions need to be made quickly. The choice of a method for solving the identification problem depends on specific conditions, such as the amount of resources, the need to take into account a priori information and the requirements for the speed of data processing. It is important to take into account that each method has its own advantages and disadvantages, and their proper application can significantly improve the effi-ciency of solving problems in the field of radar. Special attention is paid to modern technologies of radar infor-mation processing, including methods of machine learning and artificial intelligence. These methods can significantly improve the quality of data processing due to adaptability and learnability on new data. However, their application is limited due to high computational complexity and the need for large volumes of relevant data for training, which is difficult in real combat conditions.