Open AccessPattern Radar Images Formation’s Like a Stochastic Differential Equations for Recognition of Space Objects
Author(s) -
А П Кадочников,
А. А. Казанцев,
O.A. Mishukov,
С А Шигорев
Publication year - 2019
Publication title -
trudy učebnyh zavedenij svâzi
Language(s) - English
Resource type - Journals
eISSN - 2712-8830
pISSN - 1813-324X
DOI - 10.31854/1813-324x-2019-5-4-106-113
Subject(s) - radar , artificial intelligence , space (punctuation) , computer vision , computer science , object (grammar) , radar imaging , pattern recognition (psychology) , mathematics , telecommunications , operating system
The resource problems of the traditional use of detailed radar images for reliable recognition of space objects are shown. The urgent task of forming a new type of model of radar images of space objects to determine the signs of their recognition is posed. Corresponding mathematical models of such images based on stochastic differential equations of elliptic type are presented. The adequacy of the developed models to the real radar images of a space object was assessed. It is established that for the description of radar images of space objects the most suitable is a modified model in the form of a mixed derivative of an elliptical model. To test the hypothesis about the possibility of using the radar image model when constructing descriptive recognition signs, an experiment was conducted to distinguish four different types of space objects. The experimental results showed the possibility of using a mixed derivative of the elliptical model to determine signs of recognition of space objects.