The beginning of the mass use of UAVs for various purposes gave rise to the problem of their safe and controlled movement in space, especially in dense urban areas. Examples of recent events covered, including in the media (airports, private property, secret objects, etc.), showed the relevance of solving this problem, which is related to the need to create an appropriate multi-position detection and tracking system to avoid air incidents.
     It is supposed to use in the system not only radar, but also other sources of primary detection of air objects: optical, acoustic, thermal imaging, providing information about the sector where the UAV is located. Since the correct functioning of each of the listed sensors is limited by building parameters and weather conditions, a distributed system based on data integration and centralized control is able to function correctly.
    An intelligent control system for distributed radar (and other) means should manage in real time the modes of operation of each data source for scanning an area of airspace, obtaining information about moving objects, calculating the coordinates and direction of movement (components of velocity vectors) from the processed data, as well as forecasting the position of the air object to make a decision on the issuance of information to the associated system.
      A scientific novelty is the ability of such a control system to intelligently reconfigure depending on the building profile, the number of information sources available via wireless communication lines, weather conditions, as well as the ability to make real-time decisions about the mode of operation of each source of information and the parameters of the system as a whole.
     The expected results and significance lie in the fact that the projected intelligent control system for distributed radar facilities can be used as an air traffic control system in places with an increased probability of the appearance of air objects, as a security system for detecting unauthorized crossing of territory borders, and also as a system of external trajectory measurements during testing of air objects (including small ones). As radar means, it is planned to use cascaded antenna sections, as signal coding - new code structures based on nesting and asymmetric representation, which allows you to create mobile radar stations with customizable characteristics depending on the required detection range and the required sector of view of space, interference environment.

​

Research results for 1 year

​

      During the reporting period, studies were conducted aimed at developing general theoretical and practical aspects for the implementation of distributed radar control systems that detect UAVs in dense urban areas, including using artificial intelligence methods.
      An analysis of UAV variants capable of carrying a payload at a given speed and the required flight time has been carried out. The corresponding databases have been compiled. An aerial target simulator based on a hexacopter UAV (DJI S900) with a mobile corner reflector was designed and implemented to achieve the calculated reference characteristics.
  A comparison of multi-position (spatially distributed) and single-position radar stations is given. The expediency of application and features of options for building multi-position systems and processing information in them, in relation to the detection of UAVs in dense urban areas, are shown.
      Interfacing with portable computers of experimental samples of small-sized radar stations "Mars-1M", "Mars-B", "Ares", implemented using the technology of cascaded active phased waveguide-slot antenna arrays, has been implemented. Field experiments were carried out to detect and track an aerial target simulator of a two-position radar system operating in an autonomous mode. UAV coordinates are measured by combining information from two positions of small-sized radars placed in parallel or in series. Theoretical and practical issues of designing radar stations for sector coverage with the use of electronic scanning and the rangefinder-goniometer method of measurement are considered.
     The applicability of algorithms for classifying and recognizing objects on radar portraits formed using distributed mobile radars is analyzed. A method for recognizing small-sized objects based on the use of neural network algorithms has been developed and tested, and appropriate field tests have been carried out, including in conditions of dense urban development.  
     The program "Radar environment simulator of an intelligent control system of a distributed radar system" has been developed, in particular, the following modules:
-- formation of relative position and determination of trajectory coordinates;
-- formation of flight trajectories of observed objects.
    The process of modeling a distributed system consisting of two mobile radar stations is described, which is applicable for testing methods for detecting and estimating the coordinates of air objects. To implement the simulation model, analytical relationships were obtained to calculate the coordinates of the observed physical objects, both from range-finding information and from goniometric information. A block diagram of the stages of modeling for determining the trajectory coordinates of an object is proposed. The expediency of using a distributed system in order to improve the accuracy of measuring trajectory coordinates is shown.
    The analysis of methods for determining the location of objects in the local navigation field, formed by two radar stations, is carried out, the structure of the local navigation field, consisting of three navigation posts, is determined.
     An analysis of the dependence of the specific attenuation of electromagnetic waves by forest vegetation due to absorption and scattering on the frequency of propagation has been carried out. The processing of experimental data of radar measurements on the route passing through the forest within the city, confirming the calculated results.
    The analysis of the effect of precipitation on the propagation of radio waves in natural conditions is carried out, and a model for the specific attenuation of radio waves in rain is proposed to take into account their attenuation in a multi-position radar system. An algorithm for estimating the long-term statistics of attenuation of electromagnetic radio waves in rain is proposed.
     To ensure noise-immune communication between distributed mobile radars and a processing and control center, new code structures are proposed that are built on the basis of rows of cyclic persymmetric quasi-orthogonal matrices and provide better selection of the frame preamble in the data stream. The issues of searching and researching promising error-correcting codes with better autocorrelation and periodic correlation functions based on special quasi-orthogonal matrices of circulant and two circulant symmetric structures are considered.
 

​